Price Elasticity – Sniffie

How to forecast discount campaigns with forecast module

Tomi Grönfors — Thu, 01 Jun 2023 10:29:46 +0000

Discount campaigns can be a powerful tool in a marketeer’s arsenal, driving customer engagement and sales volumes while moving inventory. However, executing successful discount campaigns requires careful planning and precise forecasting to ensure the right balance between promotional offers, profitability, and supply chain effectiveness. In this post, we’re going to examine how you can use Sniffie’s forecasting module to plan effective discount campaigns.

What is Sniffie’s Forecasting Module?

Sniffie’s forecasting module uses advanced enforced learning AI to predict future sales by understanding the interaction between product demand, price, and product costs. It incorporates your sales history, seasonal patterns, and uses regression analysis to create a robust forecast, allowing you to compare different scenarios against the status quo.

How to Use Sniffie’s Forecasting Module

1. Select Your Product

Begin by choosing the product or products you wish to simulate and forecast. Use any filters to narrow down your selection. You can combine both static product filters and AI insights like Dynamic ABC Analysis.

2. Open the price simulator

After you have selected the products the blue top bar appears. Click on the forecasting icon on the top bar to open up the price simulator.

3. Adjust pricing, cost and stock

Adjust the price, the costs and modify the stock to see how changes impact your forecast. You have three distinct views to understand your forecast accuracy: Insights, Future Forecast, and Current Situation AI Insights.

Insights: This tab shows how price changes will affect your profitability and sales volumes.
Future Forecast: The future forecast gives a timeline of your profitability and sales volumes for a selected period, showing how they develop. The green line is the new scenario and blue line is the current scenario. The confidence interval is shown as the base colour around the line.
Current Situation AI Insights: This section demonstrates how price elasticity can affect your sales volumes, revenue, and profit. It fits all the product elasticities into one and shows you how much a percentual change up or down will impact volume, profit and revenue.

Planning a Discount Campaign Using Sniffie’s Price Simulator

Once you’ve selected your products, it’s time to start planning your discount campaign.

1. Establish Preliminary Discount

Begin by setting a preliminary discount rate. Note how this affects your profitability. You can adjust the discount % up or down and see how your profitability and volume changes.

2. Adjust Discount Levels

Play around with different discount levels to observe how changes impact profitability and sales volumes. Especially important is to monitor how the difference between current stock and infinite stock develops.

3. Assess Stock Levels

As mentioned above, if you notice a significant difference between the infinite and current stock levels, adjust first the timeframe to see how the sales volumes change. If the sales volume is projected to dip close to zero, it means your current stock might be insufficient for the campaign.

4. Modify Stock Levels

Use the stock modifier to increase your stock in response to the projected demand. Experiment with this feature to strike a balance between increased sales, inventory, and profitability.

5. Adjust Costs

Finally, consider additional costs associated with your campaign. You can add a percentage increase to account for extra costs or elements like kickback discounts. This will finalize your scenario planning and you are ready in minutes, instead of hours with Excel sheets.

Sniffie’s forecasting module offers an intuitive way to plan your discount campaigns, using AI to predict future demand and simulate different scenarios. By adjusting parameters like price, stock levels, and costs, you can find the perfect balance that optimizes both sales volumes and profitability. By integrating Sniffie’s forecasting module into your campaign planning, you can ensure you’re making data-driven decisions that will help your business thrive.

Conclusion

In conclusion, Sniffie’s forecasting module provides marketers with a powerful AI-driven tool to plan and execute effective discount campaigns. By leveraging advanced enforced learning AI and analyzing factors such as demand, price, and costs, this module offers precise predictions and scenario simulations. With the ability to adjust parameters like pricing, stock levels, and costs, businesses can optimize sales volumes and profitability. By incorporating Sniffie’s forecasting module into campaign planning, marketers can make data-driven decisions that drive success and help their business thrive.

The post How to forecast discount campaigns with forecast module appeared first on Sniffie.

Dynamic ABC Analysis: Powering Your Product Portfolio Management

Tomi Grönfors — Thu, 01 Jun 2023 06:26:04 +0000

In the constantly evolving world of e-commerce, effectively managing your product portfolio can be a challenge. Let alone to understand what products bring in the value and how that changes over time. ABC analysis is a handy tool, yet it often takes a lot of resources or manual work. We at Sniffie are on a path to make pricing simple for e-commerce, and to help you with that we released a new feature: Dynamic ABC analysis.

Dynamic ABC analysis

At its core, the Dynamic ABC Analysis feature is a pioneering AI algorithm that recalculates your product portfolio into ABC groups based on three key parameters: revenue, profit, and volume. Designed to streamline filtering, pricing analytics and dashboards, and dynamic pricing workflows, this feature is a game-changer for businesses looking to boost their pricing efficiency and profitability.

A Deeper Dive into Dynamic ABC Analysis

Dynamic ABC Analysis takes the traditional ABC inventory categorization and adds a new layer of intelligence to it. Earlier on your ABC classifications were updated maybe quarterly and were mainly done manually in Excel. With the help of AI, we generated a Dynamic ABC Analysis that is reliable, quick and dynamic. For a quick refresher, ABC classification is a method of sorting inventory into three categories:

Class A represents about 20% of the products, which are the highest profit contributors.
Class B signifies the next 30% of products with moderate profitability.
Class C encompasses the remaining 50%, usually the least profitable products.

However, Dynamic ABC Analysis goes one step further by dividing each category into three subcategories. For instance, class A products are further divided into subcategories

A1 (the best 0%-6.7%),
A2 (the following 6.7%-13.4%), and
A3 (the rest of A1 representing 13.4%-20%),

providing a more granular view of your product portfolio’s profitability.

Why machine learning and AI are the future of product pricing?

AI is a buzzword like no other. We sat down, wrote all we know about it and let you be the judge. Here is a seriously indepth 50-page insights on how to use artificial intelligence in pricing. Hope you find it useful!

How to use Dynamic ABC Analysis in different pricing workflows

Let’s explore how this innovative new feature can supercharge your operations across three key areas:

1. Filtering

Using Sniffie’s global filters, you can streamline your product catalog by selecting specific ABC categories. This means you can focus solely on for example ‘A1’ products, if desired, allowing you to zero in on the products that matter most. Further refine your filtering by incorporating other parameters such as revenue, category, or AI insights, creating a tailored pricing or analytics workflow.

2. Analytics and dashboard

Sniffie’s pricing dashboards incorporate Dynamic ABC Analysis, enabling you to create widgets and compare crucial aspects like revenue, profit, or volume development across different ABC categories. This provides rich, comparative insights to guide your strategic decision-making.

3. Automate your pricing strategies

Harness Dynamic ABC Analysis to shape your pricing strategies. By creating filters that utilize this feature, you can trigger automatic pricing changes based on a product’s category. For instance, if an ‘A1’ product’s volume is lower than expected, workflows could automatically adjust the price to stimulate sales.

A Tool for Every Team

The versatility of Dynamic ABC Analysis extends its utility across various teams within an organization. Whether you’re part of the pricing, category management, or supply chain team, this feature can provide essential insights into how product performance and pricing influence your daily operations.

In conclusion, Sniffie’s Dynamic ABC Analysis feature offers a powerful, sophisticated way of optimizing your product portfolio management. By providing real-time, AI-driven classification based on revenue, profit, and volume, this tool not only simplifies operations but also empowers businesses to make data-driven decisions that drive profitability. Welcome to the future of e-commerce – it’s as dynamic as your business needs to be.

The post Dynamic ABC Analysis: Powering Your Product Portfolio Management appeared first on Sniffie.

Pricing Automation for a Consumer Electronics Retailer

Tomi Grönfors — Thu, 25 May 2023 10:01:48 +0000

Streamlined pricing processes with pricing automation

How did Lumise, a prominent Consumer Electronics Retailer, achieve significant cost-savings and received streamlined pricing processes with our pricing automation solution

Consumer Electronics
Omnichannel
Europe

5/5

A must have tool for modern Ecommerce companies.

Lumise

Lumise is a prominent Ecommerce retailer that specializes in high-quality lighting products for various purposes. Their extensive range of products cater to different segments of the market, including professional, commercial, and consumer lighting products. With over a decade of experience in the industry, Lumise has established itself as a trusted name in the European lighting market and beyond, with an annual turnover of close to 8 million euros. Their selection of products includes state-of-the-art lighting solutions that are cost-effective, making them a preferred choice for customers. Their customer service team is dedicated to providing exceptional support and guidance throughout the purchasing process, ensuring that customers receive the best possible experience.

Need a Pricing Tool for Consumer Electronics

Take the first step and see how we might be able to help you with a pricing automation solution. We offer a 14-day free Shopify trial!

No credit card needed – no strings attached.

Case background

Lumise, a European consumer electronics Ecommerce store, has been on the hunt for a pricing automation solution to boost their pricing profitability. But not just any solution would do. They were looking for an AI-driven tool that could effortlessly manage all their pricing workflows. With a large catalogue of products and a complicated master data setup, Lumise needed a solution that could handle the complexities of their business. They were looking for something that would help them stay ahead of the curve and keep their pricing strategies innovative and profitable.

The challenge

One of the biggest challenges faced by Lumise was managing the pricing of their vast product portfolio. Pricing was a manual process that took a lot of time and was error prone. Naturally, with a large product portfolio on offer, it can be difficult to determine the right price for each one and make sure the overall profitability is monitored as well.

This is where Sniffie comes in, offering Lumise a pricing automation solution that takes care of all their pricing needs. By using Sniffie, Lumise is now able to run profitable sale campaigns, forecast price changes, and position their pricing in a way that maximizes their profits.In addition to these benefits, Sniffie also provides Lumise with detailed insights into their pricing products and pricing strategy. With access to real-time data on consumer demand and market trends, Lumise can now make informed decisions about their pricing strategy that will help them stay ahead of the competition. Furthermore, Sniffie’s pricing automation features have helped Lumise streamline their operations, freeing up valuable time and resources that can be better spent on other aspects of their business. With Sniffie, E-commerce team can focus on delivering high-quality products and services to their customers, secure in the knowledge that their pricing is being taken care of by a trusted partner.

Sniffie has been a super valuable investment that really has increased our profitability and reaction time. A must have tool for modern Ecommerce companies.

Kari Kauhanen

CEO, Lumise

Try out our Pricing Automation Solution!

Take a free 14-day trial for your Shopify store!

The post Pricing Automation for a Consumer Electronics Retailer appeared first on Sniffie.

Why are consumer price index searches soaring in Google?

Tomi Grönfors — Thu, 12 Jan 2023 13:15:20 +0000

Now that Consumer Price Index (CPI) came out in USA we found out how it impacts the Google searches for Consumer Price Index. Yes, the searches are soaring. Why is that and how does CPI figures help companies to understand the economy and do pricing? So we will dig into the basics here.

What is a Consumer Price Index?

The Consumer Price Index (CPI) measures the average change over time in the prices consumers pay for a market basket of consumer goods and services. It is used to measure inflation and cost of living. So it tells a lot about where the purchasing power is heading. That said, that is why companies should monitor it. The Bureau of Labor Statistics (BLS) calculates the CPI for the United States.

How does it help companies to understand the environment they are in?

A company can use the Consumer Price Index (CPI) to forecast future demand by analyzing the relationship between inflation changes and consumer spending. Especially in their own store. Understanding quickly how much increasing inflation affect them, is crucial to make quick changes in their assortments and pricing. Here are a few ways a company can use the CPI to forecast future demand:

1. Identifying trends:

By analyzing historical data on the CPI and their own sales, a company can identify trends in how inflation and prices affect consumer spending patterns in their business. This helps companies predict how future inflation changes might impact their product demand and how to set prices in the future.

2. Analyzing customer behavior:

By understanding how their different customer groups are affected by inflation and price changes, companies can make more accurate predictions about how demand for their products will change. For example, if the company’s target market comprises primarily low-income consumers, they will be more sensitive to price increases.

3. Using inflation expectations:

A company can also take into account inflation expectations. If the inflation rate is expected to be high in the future, consumers may spend more now before prices rise.

4. Combining with other factors:

A company can also combine the CPI data with other factors that can impact the demand for its products, such as the state of the economy, consumer sentiment, and their marketing efforts. In many cases having high stock that can be offered affordably with relevant campaigning, a company can grow its customer base, whereas its competitors can’t.

How can you use Consumer Price Index in pricing?

In practicality, you can use the Consumer Price Index (CPI) in following ways:

Understand how your Cost of Goods Sold might change

You should use the CPI to track changes in the cost of your raw materials, purchase prices, labor, and other inputs, and adjust your pricing accordingly to maintain a desired outcome. It would help if you planned price changes so that your overall profit will also cover your operating costs. Although you would obtain a high gross margin it might mean you are selling less and then getting lower overall profitability than you earlier did.

Use inflation figures periodically

By monitoring the CPI periodically, you can anticipate changes in inflation and adjust your pricing in good time to maintain the value of your products. As above, measure closely how much you can raise prices to keep the demand focused on your products.

Understand how your competitors are affected by inflation

You can also use the CPI to estimate how changes in inflation and prices will affect the prices of similar products offered by your competitors, and adjust its prices accordingly to remain competitive. This is especially hard in retail where price changes might not be that often compared to Ecommerce, where prices change much more often. In cases where competitors use dynamic pricing software to change their prices it will be hard to predict what part of the price change is due to inflation.

Make clear estimation of overall consumer demand and how your customers react

One way you can also use the CPI to estimate how changes in inflation affect consumer spending patterns. Are there new products that you could offer? Is there products you should get rid of? Can you offer your most wanted items on more competitive pricing to maintain or increase demand for your products. NOTE: You should research and understand the prices of similar products that your competitors offer to remain competitive and maintain your market share.

Understand your brand positioning and brand value

Your pricing strategy should align with your overall brand positioning and target market. If you are the desired brand, most likely, you won’t be affected as much as the less desired ones. For example, if you are a luxury brand, you will more room to wiggle than companies targeting budget-conscious consumers.

How to modify your distribution and promotion strategy in an inflation

Lastly, you can also plan how your distribution and promotion works. Are you still able to keep up with your current distribution model or should you look for more cost-effective ways to do the distribution? Another factor that comes into play is also the role of promotions. When consumers’ purchase power decreases, they look for more offers. For many skillful Retail and Ecommerce companies, this time will open a niche to carve more market share with clever promotions to obtain new markets or consumer groups.

Understandably, companies look for latest details on consumer price index. Still, as with everything, it’s crucial to note that the CPI is just one of the many factors that can and should impact your pricing strategy. It should be considered against other data and use also broader market analysis. Hopefully you found this article useful. G

FAQ

Most frequent questions and answers

What is the Consumer Price Index (CPI)?

The Consumer Price Index (CPI) is a measure of the average change over time in the prices paid by consumers for a market basket of consumer goods and services. It is used to measure inflation and cost of living.

How is the CPI calculated?

The Bureau of Labor Statistics (BLS) calculates the CPI for the United States. The BLS collects prices for a representative basket of goods and services, and then compares the cost of the basket in the current period to the cost of the same basket in a base period. The percentage increase or decrease in the cost of the basket is the CPI.

How is the CPI used?

The CPI is used to measure inflation and cost of living, and is also used as a deflator in the calculation of real gross domestic product (GDP). Additionally, it’s also used by companies to understand the economic environment they operate in, and adjust their pricing, wages, and other business decisions accordingly.

Can the CPI be used to forecast future demand?

The CPI can be used to forecast future demand by analyzing the relationship between changes in inflation and changes in consumer spending, but it should be considered in conjunction with other data and market analysis, as well as other factors such as distribution costs, promotion costs, and competition.

Can the CPI be used to compare prices across countries?

The CPI is calculated differently in different countries, and may not be directly comparable. However, the Organization for Economic Cooperation and Development (OECD) and the International Monetary Fund (IMF) both publish purchasing power parities (PPP) that are used to compare prices across countries.

The post Why are consumer price index searches soaring in Google? appeared first on Sniffie.

How to calculate a perfect sale price after the Omnibus Directive came into force?

Tomi Grönfors — Mon, 19 Sep 2022 09:10:56 +0000

How do you calculate a perfect sale price for your next campaign? Running a sale like Black Friday to boost your product sales is often tempting. Earlier on, previous campaign results and estimated outcomes were enough to make a solid plan for the next sale. You probably used the same approach, most likely, every time. However, as the EU’s new Omnibus Directive is in force, the legal requirements mean you must change how you prepare for sale campaigns. You must also change the way you plan your sale price. We decided to study the difference between two approaches on pricing to create a stellar sale campaign. We used the demand models used by our AI to enable as accurate a forecast model as possible.

The background for the sale price simulation

As said above, we base our simulation on actual demand. As the product, we are selling a chair. A light wooden chair called creatively Light Wooden Chair.

The original price we brought to market was 500€, and now the price is 450€. We use that as a given premise in both cases. Firstly, we are 31 days from the 30-day discount campaign. That is why, we must decide whether to raise the price higher to get a better comparison price (Omnibus Directive regulation) or use a lower price to start the sale.

The simulation considers in both scenarios Omnibus Directive and will comply with that legislation. We will go to that at the end of the blog post. But now, we must decide on the market we are in and select the price sensitivity of our customers.

Show your Omnibus price automatically

We keep an eye on the lowest prices for 30 days and dynamically show them on your website. Show off your discounted pricing with confidence!

Determining the customer profile first

As pricing highly depends on your customers’ price sensitivity, we will consider that as the starting point. In case multiple retailers are selling the same products around the internet, it is more likely that your customers will look for the lowest option.

To make this simulation real, we used aggregated data from Sniffie AI to create five different customer profiles (Sniffie demand function). The price sensitivity takes into account the market competition as well. The more competed the market is, the higher sensitivity you should choose. The more options there are on the market, the more possibilities consumers can choose from. The customer profiles based on price sensitivity use price elasticity: (see also: Price elasticity basics: Understand the Strategic Pricing Gateway)

1. Extremely price sensitive customers

For these customers, the price is of utmost importance. They know the market and price and compare every purchase from multiple retailers. They love a good deal. So for them, your promise of a low price makes no difference unless it’s true. You can see the demand and profit curves below every profile against the cost of our product.

Picture 1. Profit and demand curve for Extremely price sensitive customers

2. Very price sensitive customers

For this segment, price plays an important part. Because they know the market price, a more negligible price difference will make them buy elsewhere.

Picture 2. Profit and demand curve for Very price sensitive customers. Note how the demand curve starts to smooth out in the end.

3. Price sensitive customers

For these customers, price is essential but not the only thing they consider in their purchase decision. They might appreciate fast delivery, good reviews, branding, earlier purchases etc.

Picture 3. Profit and demand curve for price sensitive customers. The demand and profit curves start to be fairly smooth.

4. Price affects very little to these customers

These customers do not care about the price. They love shopping for products on impulse or from the same retailer. They also care more about the saved time on fast shopping experience and delivery than the actual sale price.

Picture 4. For these customers, for whom price affects very little, the price range is already really high.

5. Price has no effect to these customers

For these customers, it is irrelevant what they pay for your products. So naturally, for this simulation, we will consider the above examples. An example of this type of product is a Ferrari. On the contrary, people buying a Ferrari might love the fact that it costs more. Then they know fewer people are driving those.

Picture 5. Totally inelastic profit and demand curves

Use Our Free Price Elasticity Calculator

Use this simple calculator to calculate your product’s price elasticity and understand how price changes affect your product’s demand.

Specifying the sale price before the sale

As Omnibus Directive requires us to show the 30-day lowest price before the campaign, the comparison price and discount percentage will play a part. According to studies, the minimum discount should be at least 5%. The minimum does not sound like a discount but still is the lowest amount, attracting customers’ attention. Therefore, it would be best if you chose the discount’s upper limit by evaluating our campaign’s cost structure and aim. In this example, our costs are only 18% of our price (450€), which gives us great flexibility when designing the on sale price.

Let’s assume we aim to create as much profit as possible and clear stock to make room for newer models. For this example, we evaluate that we want our minimum margin to be around 160€ price point, and to clear stock, we want the minimum sale percentage to be at least 30%. The 30% we have learned by analyzing previous sales. These premises paint our range of possible sale percentages to 30% – 64%.

Note that the discount will cut our margin more than it cuts the price the customer has to pay. So in this example, giving a 50% discount from 450€ for the customer cuts our margin by roughly 60%.

In these scenarios, we also assume that we are not the only ones selling light wooden chairs. According to studies, this makes our customers more price sensitive. Based on studies, we can also assume that since we are an eCommerce company, the customers respond better to “30% cheaper than market” signs than to “30% discount from our previous price”. This encourages us to focus our marketing more on the discount from the market price than the discount percentage from our original price.

The question is: What is the best price for the 31 days before the campaign? How will that affect the overall profitability of our sale campaign?

We will lose sales in the previous 30 days if we have a very price-sensitive market. So to cover that, we need to understand the impact of the price change before the sale campaign.

Selecting the sale campaign discount %

A successful sales campaign depends highly on how good deals you can provide—knowing your purchase price, demand from earlier campaigns and your marketing budget, the questions are:

What will be the best sale discount % in your upcoming campaign?

What is the best option for the sale to be as profitable for us as possible?

Finally, before you make the decision you could use two thumb rules. First, the percentage should be more than 30% (see above the explanation) and the end price inline with other sale prices on the market.

To see the difference, we will compare the two scenarios and think about how realistic those are for us.

Returning to normal pricing after sale campaign

We are naturally returning our pricing to normal, which also needs some consideration. For example, what should be the regular price of our product after the sale? We again examine the price impact on the overall profitability 30 days after the sale. Many eCommerce companies use dynamic pricing, so this issue might be something they consider more on the gateway they give the pricing engine to set the prices.

Running the sale price simulation

Eventually, we analyze the results between the scenarios with detailed analyses and see how revenue and profit develop at each stage. To make this visually clear we highlighted the three 30 days periods with different colors. The areas are:

Lighter pink

30-day period before the campaign.

Green

30-day campaign period.

Purple

30-day period after campaign.

So let’s start.

Ensuring Price Display Compliance

Make sure to adhere to the Omnibus Directive when reducing the price of your products.

Determine the market for Light Wooden designer chair

As we are selling a Light Wooden designer chair, the market we are in is price sensitive (see above for the reasoning). Our customers value the price, but it’s not the only criteria they consider. They appreciate fast delivery, which we are known for. So the price sensitivity is set to the second highest level. While using our campaign pricing simulator, you can always change the demand for both scenarios.

Setting the 30 day price before the sale

As the price is now 450€, we need to determine the strategy for both scenarios. To see how we perform, we create two highly different techniques. Scenario 1. is built on the expectation that a very high sale discount % will help us create a good campaign and get new customers for our customer pool. So for scenario 1. we raise the price to 500€ (a 50€ increase).

For scenario 2. we remain at the 450€ level. The ration behind this approach is that we have higher volumes before the campaign. That way we make a more constant flow of customers to our shop.

Table 1. Price set before campaign	Scenario 1.	Scenario 2.
Price set Before Campaign	500,00 €	450,00 €

Finalising the sale price and sale discount %

Now that we have raised scenario 1. price to 500€, we even have the legal right to compare the price to the original launch level ie. 500€. This will give us a higher levy for setting the sale % in the campaign.

Our purchase price is 80€ for both scenarios, so the discount % can be fairly high in both cases. For scenario 1. we set the discount to 60%. That is a very high discount. For scenario 2 we set the sale discount to 45%. It is higher than 30% and in the expected line with the market pricing. See the sale campaign details from the below table.

Table 2. Sale Campaign details	Scenario 1.	Scenario 2.
Campaign Discount %	60 %	45 %
Sale price	180 €	247,50 €
Comparison prices	450 €	500 €
Lowest 30-day price	450 €	500 €
Savings %	-60 %	-51 %

Returning to normal pricing after campaign

As we now have good sales volumes and profitability of 450€ level, we decided to try out how our simulation works if in scenario 1. we raise the price to 450€ and in scenario 2. to 500€. Below you can find the returned prices.

Table 3. Sale price comparison	Scenario 1.	Scenario 2.
Sale price	450 €	500,00 €

The outcome of the sale scenarios

With these settings, scenario 2. will win with an overall 6% profitability ratio, although scenario 1. outperforms scenario 2. with a whopping 34% profit difference before the campaign. The campaign performance of campaign 2 is 64% higher than scenario 1. when we compare the profits. On revenue, the difference is 34%, which is also high. Lets dive deeper in the analytics.

Table 4. Detailed analyses	Campaign1	Campaign2	Difference	Difference %
Overall Revenue	4 378 524 €	4 586 132 €	207 608 €	5 %
Revenue Before Campaign	1 189 080 €	764 636 €	-424 444 €	-36 %
Profit Before Campaign	977 685 €	641 589 €	– 336 095 €	-34 %
Campaign Revenue	540 000 €	732 972 €	192 972 €	36 %
Campaign Profit	299 999 €	492 981 €	192 981 €	64 %
Overall Profit	3 482 779 €	3 687 350 €	204 571 €	6 %

Picture 6. Comparison of pricing in scenario 1 and 2.

Picture 7. Sales volumes in different scenarios

Picture 8. Revenue comparison in different scenarios

Picture 9. Profit scenarios in different stages

Use our sale price planning tool

Are you interested to know what is the best on sale price for your next discount campaign? Is raising the price before the campaign wise? To know it, try out our latest tool Discount Campaign Pricing Simulator and play with two scenarios and real-life demand.

Visual difference on the product pages

There will be a psychological difference on the product pages. For example, if you are selling private labels, you might focus more on the discount percentage as the market price. Here are the two examples of our campaign:

Scenario 1. product page with the discount

Scenario 2. product page with the discount

Analyzing the results to understand why the difference is so high

The simple answer for the difference is that we were too eager to give a discount in scenario 1., although our customers are price sensitive. A 60% discount is enormous, even too high. But even if we started from 500€, 60% eats most of our profits during the campaign. Anything above 50%, is most likely too much for your profits.

For example, setting the deal in scenario 1. to 50% would have narrowed the gap to 2% on overall profitability. Moreover, putting the final price to 480€ for scenario 1. would mean that scenario 1. would win with an overall 2% and 70 139€ difference.

This example shows well that, when planning a campaign, it is too luring to higher prices before the sale and to set high discounts, even if you have price-sensitive customers.

Naturally, every case is different, and only you can know the actual impact on your customers. That said, using clear and comprehensive scenario calculations for your campaign will help you create profitable deals in your next discount campaign.

Notice that even you are human. Thinking that a high discount percentage will make the best results might be psychologically biased. Therefore we suggest you use this type of exercise to help your decision-making. Enlighten your team about the outcomes, even on a example product like this.

We uploaded the Excel tool we used to our Resource Library to help you make decisions. You can also download our price elasticity tool to know what kind of customers you are serving. Here is a list how you do the same scenario by yourself.

Download the tools
Calculate your customers price elasticity
Use that as a basis in the scenarios
Play the scenario in our Excel tool

Hopefully it will help you to make more informed decisions.

Use our FREE tools from Resource Library

Are you interested to know what is the best price for your next discount campaign? Is raising the price before the campaign wise? To know it, try out our latest tool Discount Campaign Pricing Simulator and play with two scenarios and real-life demand.

The post How to calculate a perfect sale price after the Omnibus Directive came into force? appeared first on Sniffie.

Should you optimize individual SKUs or bigger optimizable product groups?

Jonas Timonen — Mon, 17 May 2021 07:00:00 +0000

What is an optimizable unit?

When optimizing your product prices with an Price Optimization like Sniffie, the most typical scenario is to optimize individual SKUs. However, sometimes profit optimization becomes more effective when you group SKUs into bigger optimizable product groups. Let’s explore the different options.

Optimizable product or optimizable product group refers to a single SKU or a group of SKUs whose price is optimized with an AI-based pricing tool.

To start with, it is not mandatory to group products into an optimizable product group. If the sales volumes of an individual SKU are high enough, the SKU by itself can (and typically should) be an optimizable product.

The minimum amount of sales volume for an individual SKU to be considered an optimizable product is 0.5 sold units in a week.

Another criteria for any SKU to be considered an optimizable product is that you have to be able to change the price point of the SKU in question.

Are you in need of profit optimization?

Maximizing profitability in eCommerce is vital. We have a ready made module for optimizing profitability in our pricing tool.

When does it makes sense to group multiple products as optimizable product groups?

If the products you are selling are very similar to each other, grouping them together as bigger optimizable product groups can be a good idea instead of having them compete with each other in price. This becomes especially important, when you are optimizing profits based on demand.

Let’s say you’re selling t-shirts in five sizes and in five different colors for the same price. It is a good idea to group these five together so that all their prices change when a new price suggestion is made by our tool.

In some industries, grouping products with the same use case and similar price can be a good idea. Think about a hardware store in which they sell many different models of simple screws that are in a similar price range. No matter the model, all the screws could be grouped together to neutralize the risk for cannibalization.

A good rule of thumb is that if you have many similar products within the same price range in your product catalogue, such as t-shirts, grouping them together is often a good solution.

We at Sniffie are more than happy to help you with the process of deciding which products could be grouped together and which products should be optimized singularly.

Benefits of grouping products together into bigger optimizable product groups

Grouping products into bigger optimizable product groups has two main benefits:

First, grouping can minimize cannibalization when optimizing products that are at risk of cannibalizing each other.

Secondly, grouping allows you to price your products based on price elasticity in a situation where individual SKUs are not providing enough sales for them to be optimized efficiently.

Read more about our AI-based Price Optimization tool for ecommerce and retail businesses.

Interested in profit optimization?

Maximize the profitability of your products by setting prices that your customers are happy to pay for. See how we can help.

The post Should you optimize individual SKUs or bigger optimizable product groups? appeared first on Sniffie.

5 things you should know about price elasticity in e-commerce

Jonas Timonen — Thu, 22 Apr 2021 16:21:56 +0000

A quick primer: What is price elasticity in e-commerce and in general?

Let’s start with a quick summary of the basics.

In short, price elasticity tells you how much a price increase or decrease will impact the demand for a product. For a more detailed explanation, we would recommend reading this article about the definition and causes behind price elasticity.

The majority of all products sold in the world are elastic – some more than others – which means that their demand will go up when prices go down or vice versa. Some products, on the other hand, are relatively inelastic – meaning that their demand won’t be affected much by changes in price (but this is not very common in e-commerce).

Let’s first see a few examples of both categories to get a better grasp on the difference.

Products with extremely elastic prices are typically luxury or nice-to-have goods where the consumer has a lot of room for choices and where there are a lot of alternatives at the time of purchase, such as:

Consumer electronics. Most consumer electronics have multiple options at varying price points. If you are selling wireless earbuds and you increase your prices by 20%, you’ll probably see a significant drop in both demand and revenue.

Clothing. If you think about generic white t-shirts, sweaters, or jeans, consumers have a massive number of substitute options at hand, which means that in many cases, the demand of your white t-shirt is largely affected by price. Conversely, people will typically pay “extra” for luxury goods with added perceived value.

Products with extremely inelastic prices are typically necessities with no substitutes available, monopolized products, or products with extremely tough competition (let’s get back to competition in a minute) or poor availability, such as:

Insulin. If a person needs a life-saving drug like insulin, they’ll need to buy it, no matter the price. Changes to price will have a small effect on the demand (until it’s not affordable anymore).

Toothpaste. Most people agree you need toothpaste to keep your teeth clean. If the price fluctuated a little, most consumers would still be likely to purchase it because of its usefulness.

If you want to understand better what actually causes price elasticity in e-commerce or inelasticity, we recommend reading our complete guide to price elasticity here.

Benefits of pricing based on price elasticity – and the risks of not understanding it

All this being said, it’s obvious that elastic and inelastic products should be treated differently when it comes to pricing. This is also why understanding the concept of price elasticity will help you make better and more profitable pricing decisions — regardless of if your products are elastic or inelastic.

Some of the benefits of pricing according to a product’s price elasticity include:

Understand your buyers’ behavior better to inform your pricing strategy: Understanding how consumers will respond to price changes can not only help in mitigating the risks involved but also help in reducing the uncertainty in making pricing decisions.

More accurate sales forecasting: By knowing the optimal price of a product, you can forecast your sales and even set up prices for the future.

Increased profit margins: If you’re selling inelastic products, you might have more flexibility when it comes to increasing the prices of your products because people won’t be turned off by the increased price. If you’re selling elastic products, you can increase demand by giving out discounts and see an increase in revenue.

In turn, failing to understand how price elasticity affects demand can do a lot of harm to your profitability. A typical scenario is that the price elasticity of a product is not taken into account in pricing decisions and products are being sold at a too low price, which in turn results in low profit margins.

Calculate your products' price elasticity with a free calculator

5 common price elasticity myths debunked

Myth #1: Taking price elasticity into account always results in lower prices

Probably the most common misunderstanding about price elasticity in e-commerce and in general is that pricing according to your products’ price elasticity always results in lower prices.

Good news: you don’t have to be the cheapest on the market to benefit from pricing based on price elasticity.

In our experience, the optimal price point (where profit margins are at the highest) is actually often a lot higher than the starting price — which means many companies are essentially leaving money on the table.

Myth #2: The more you decrease prices, the more demand you will have

Decreasing a product’s price often results in increased demand, but not always. There are a few (surprisingly common) scenarios where this logic fails miserably:

Sometimes if you decrease the price of a product too much, people can get suspicious and may start questioning the quality of the product, and therefore the demand can drop (e.g. if a $1.000 iPhone is sold at $500).

You can run out of stock faster than expected, in which case the increased demand won’t help.

You’ve already reached the whole market (or a big enough proportion of it), which means lowering prices won’t help you to acquire new customers — because there are not potential customers left in the market.

In some cases, no matter how much you lower your prices, your competitors will match your price, while the size of the market stays the same — and all of you will be left with lower profit margins and nothing more.

Myth #3: Price elasticity is something intangible and non-scientific that cannot be verified

Wrong! Price elasticity is not just a gut feeling that you could maybe charge more; it can indeed be calculated and verified in a very precise way which helps you as a merchant to find the optimal price point for each product.

In mathematical terms, price elasticity can be calculated in a couple of different ways (disclaimer: which formula you should use depends on the situation, and the example presented here is just for the sake of illustration).

One way to do this is to divide the percentage change in quantity demanded by the percentage change in price.

Let’s say you’re selling shirts online for $80 each and currently, you sell 500 shirts a day. You decide to increase the selling price by 12.5% which adds up to $90 for a shirt. Because of the price increase, your sales drop from 500 shirts sold every day to 400 shirts sold — a 20% decrease.

Now, if we use the formula above, we get the following result:

20% / 12.5% = 1.6.

Price elasticity of demand is then 1.6. So, what does this value tell us?

Here’s what different price elasticity values signify:

Value is 0: Perfectly inelastic product → Price changes have no effect on demand
Value is between 0 and 1: Relatively inelastic product → Big price changes have a small effect on demand
Values is exactly 1: Unit elastic product → Increase in price and decrease in demand are equally matched
Value is greater than 1: Price elastic product → Demand affected significantly with a change in price

The value of 1.6 tells us that this particular product’s price is elastic.

If you’re interested in learning more about your products’ price elasticity, check out our handy price elasticity calculator.

Myth #4: “We have too much competition to benefit from price elasticity”

Newsflash: chances are you actually don’t have as much competition as you think – and even if you did, it still probably makes sense to make small, controlled price adjustments to understand the price elasticity and find the optimal price for the most competitive products, too.

So, what does it mean to be in a ‘competitive’ market, really?

Scenario 1: When most e-commerce businesses say that their market is competitive, they simply mean that there are other stores selling more or less similar products and that customers have the opportunity to choose from multiple options. In this case, you still have some room for price optimization, even if the changes are modest (but remember: the bigger the sales volumes, the bigger the compound impacts in profits).

Scenario 2: When we say that a market is competitive, we mean that you’re selling the latest iPhone and you’re fighting for profit margins with everyone else who’s selling exactly the same product for exactly the same price. In this case, you simply can’t change the price without seeing changes in demand right away. If you underprice your iPhones by too much (e.g. sell for 50% lower than everyone else in the market), you’ll see a huge increase in demand but lose in margins. If you overprice compared to everyone else, people will react by going for your competitors.

The second scenario is what we mean by an ‘extremely competitive’ situation, and to be brutally honest – most companies don’t fall into that category, which means they typically have a lot of room for price optimization.

And even if you sold iPhones or other extremely competitive products, making small price changes (think 1% change rather than 10% change) will actually help you find the optimal price point for those, too. Additionally, an iPhone seller is most likely also selling some other products (like phone accessories) with more wiggle room — and therefore has some untapped opportunity in other categories.

Myth #5: It’s not worth the time to optimize all prices based on price elasticity

Having hundreds or thousands of products in your catalog might seem like a lot to manage and reprice.

And we agree: in many cases, it usually doesn’t make any sense to reprice tons of products and analyze data manually.

In fact, it’s typically so time-consuming to make any sense of that much data that it’s practically impossible.

Enter an AI-based pricing tool that can help you automate your price optimization and price changes fully and help you find the optimal price point and maximize profit margins for as many products as you need (we’re talking about hundreds of thousands and up) without any manual work whatsoever

Learn the benefits of pricing with price elasticity

Conclusion: Pricing based on price elasticity is the best choice for (almost) all e-commerce businesses

Hopefully, by now you have a good understanding of why it’s important to understand the price elasticity (or in-elasticity) of all your products, and how this should inform your pricing strategy.

To sum it up:

Price elasticity refers to the changes in demand when the price changes. This is also the reason why you don’t need any competitor data to benefit from dynamic price optimization that builds on the price elasticity (or inelasticity) of your products.

The best way to price your products according to price elasticity is a reinforcement learning based pricing tool that helps you automatically find the optimal price point for each product.

Psst! If you’re still not sure if dynamic pricing is the right choice for your business, make sure to check out some of the most common use cases of dynamic price optimization in this article.

Learn more about e-commerce pricing

All things e-commerce pricing & price optimization right in your inbox. No spam.

The post 5 things you should know about price elasticity in e-commerce appeared first on Sniffie.

Price elasticity basics: Understand the Strategic Pricing Gateway

Jonas Timonen — Tue, 20 Apr 2021 12:20:37 +0000

What is the Strategic Pricing Gateway?

Strategic Pricing Gateway (or the gateway) can be used to understand how your volumes and prices are connected to the market you’re in.

The Pricing Gateway offers an excellent and simple way to understand what your optimum price point for profitability is and how your demand curve is changing in relation to the market price.

The gateway gives you simply minimum and maximum boundaries and your pricing should be done within those boundaries. If you go under or over the boundaries you are going to be losing money. Stay on the middle and you´re making the best profit. This concept is especially helpful for price optimization and keeping your pricing competitive.

“Think about it: If the competitor is pricing his products blindly, should you follow that model? Probably not.”

With Pricing Gateway you make sure that you have the right momentum on the market. This is not a gimmick we invented, it´s a simple tool based on price elasticity and market elasticity. When you envision your pricing as a gateway, you can easily think of it as an airplane flying in a secured tunnel. Go too high or low and your run into trouble Stay in-between the boundaries and you´re good to go.

How does the strategic pricing gateway work?

As explained above the Strategic Pricing Gateway needs three pieces of information:

Competitor pricing
Your sales volumes and pricing
Your costs base divided for the product

It goes without saying that the most successful eCommerce companies will take all three into consideration when planning their pricing strategies. The powers of Pricing Gateway is simple. You get an understanding of where your optimum price point lies and what are the minimum and maximum margins you can have. Not knowing these means that you are either leaving margins or volumes for your competition.

To start you will need to monitor market pricing of your competitors. This lets you understand what the real market prices are. The other thing you need is your pricing and sales volumes for each product used in this concept.

Remember that you will reach the best outcome when you do price testing to understand the price elasticity (how volumes change when price changes) of your products. Lastly you will need the cost base divided into a single product. This can be done in several ways, but in the simplest, you can take the same percentage of costs as the product is responsible for sales.

“Your cost structure determines you strategic pricing gateway.”

Strategic pricing gateway calculated

First you need to understand how your volumes will change if you change your pricing. This concept is called price elasticity. (Feel free to download our tool to calculate your price elasticity). With price elasticity you can estimate your demand curve (see picture 1.). Demand curve will tell you that how much your price increase or decrease will affect on your volumes.

Now that you have your demand curve drawn you can calculate your profit curve. This can be done by calculating a curve where you compare your gross profit against your price. Finally, you compare that to your demand curve (see picture 2.) to understand where your limits are. First your cost base will set the hard boundaries to your pricing. Then you must determine the minimum acceptable margin that you cannot cut. Now you have minimum profit determined. When you compare the minimum profit to your demand curve you will notice that if you raise the price high enough the volumes will drop to the level where your gross profit drops to the same level as in the case of minimum margin. This point is called maximum margin. After that point increase of prices will drop volumes so much that you will do a loss with every single product sold.

In between minimum and maximum margin boundaries is a point where the number of the products sold multiplied with the gross margin is the highest possible. This point is called optimum price point. Keeping prices always on this point will yield the best possible profit.

The market will determine your strategic pricing gateway

Strategic Pricing Gateway is not a permanent position on the market. It is a dynamic (that’s why we have developed automatic tools for that) gateway which is highly dependent on your demand curve position. Let’s say that your market has four competitors and each of them sell the same product with prices ranging from 90€ to 100€. In that case your demand curve will be highly dependent on that market pricing. But if someone changes their pricing permanently to 80€, it means that your demand curve will spread out further and eat volumes from competitors staying at 100€. eCommerce companies who utilize pricing gateway approach will have a huge advantage over those who follow competitors blindly. They understand how customers react on pricing and how much this will have impact on volumes. This enables faster growth and better profitability.

Understanding price elasticity

The strategic pricing gateway is a concept we developed to help our customers. It is no new invention, we just made it an approach to visualize the unseen force on the market: consumer demand. To utilize this tool you must understand how your products differ in terms of price elasticity. Let’s take another example and compare two different product types: inelastic and elastic. Inelastic is a product type where relative change in volume is smaller than the relative change in price. A good example of such a product would be a critical medicine. No matter if it costs 100€, 1000€ or 10000€ the demand is still the same (see picture 3.).

But if we compare that to a perfectly elastic product such as Euro coin (see picture 4.) the difference is clear. If I were to sell you a 1€ coin for 0,99€ you would buy all I have. Lowering the price even further would turn the now elastic curve to even inelastic as people would buy all your euros enabling them to get wealthier. Between 0,95€ and 1€ most of us would suffer from financing problems and thus still keep the curve elastic. Although the example is simple, it does paint you the point. The price of a product has a major impact on how many pieces people will buy in a perfect market. Naturally if you are selling potatoes in a country where everyone is allergic to potatoes, it will not give you the same curve as in a perfect market. When calculating price elasticity you should consider following aspects:

1. Product type, category and characteristics

2. Market size

3. Competition

4. Consumer characteristics

5. Availability

Why should you care about the strategic pricing gateway?

The theory behind the strategic pricing gateway sounds complex, but with simple tools you can easily understand where your gateway lies. That enables your product and pricing teams a safe, dynamic and easy gateway in which to do pricing. Because all pricing actions done in your company have an overall effect on your company’s profitability and sales of other products, it is vital to understand what is the optimum price and which are the gateway boundaries to keep your product pricing profitable.

The web has become a market place, where Amazon is in the next booth and each product is turned into a commodity that can be easily compared. There are five forces that form a tailwind for the growth of the importance of price.

1. Growth of mobile shopping

2. Faster logistics

3. Market place price competition

4. Retailer loyalty decreasing

5. Ad block growth & privacy

Where to start

So now we know what the theory is. So what. You should get it into practice and fast. So let’s recap and give you a few tips on how to start. You need the following data:

1. Price monitoring and understanding of the market

2. Volume data about your products

3. Cost structure per product

4. Simple tools

You can easily collect the market data with price monitoring tools like Sniffie. The volume data you should easily get from your ERP or even Google Analytics. Make sure you collect those prices daily against the price those were sold with. Only then you can calculate how a price change affected on volumes. Your cost structure you can determine as you best see fit, but be realistic how much one product should yield for your overall profitability. Finally, use simple tools like our Gateway Excel and understand where your Strategic Pricing Gateway lies.

Checklist for setting up a Strategic Pricing Gateway

1. Pricing gateway models

Pricing gateway models should be built on the understanding on what is your overall marketing strategy. If you want to be a higher end value added pricing retailer, you naturally will not fight with volumes. Then you want to optimize profit. On the other hand if your marketing strategy is to be a cost leader in the industry your strategic pricing gateway will be optimized for volume.

2. Calculations

Calculations are good but testing is better. Understanding the real elasticity also means taking into consideration imperfect market information such as: marketing, imperfect information, bad competitor SEO and other relevant factors. So make sure that you have done your homework around the market competition before doing the calculations.

3. Monitoring

In today’s eCommerce it is unrealistic that you would survive with some kind of monitoring. Your customer can at any time google your products and find market pricing for those. If you are selling 100.000 items, for you googling those from your mobile is unreal. Using tools like Sniffie will help you to get an automated and structured view of the market.

4. Time to time rechecking

Understanding that pricing gateways move and change is vital. Recheck your volumes and if any change is seen take into consideration if you should calculate again where your strategic gateway lies.

Wrap-up

Strategic Pricing Gateway will help you to optimize your pricing. Without doing that you will always leave money on the table in terms of lost volumes or margin. Make sure you get more power to your pricing by harnessing the simple concept of pricing gateway.

Fill out the form to download the tool

The post Price elasticity basics: Understand the Strategic Pricing Gateway appeared first on Sniffie.

Using machine learning to calculate price-elasticity

Jonas Timonen — Mon, 19 Apr 2021 11:13:30 +0000

Machine learning in price and profit optimization

Machine learning refers to the study of computer algorithms that improve automatically through experience. It is seen as a branch of artificial intelligence and it focuses on automating analytical model building. It is based on the idea that systems are able to learn from data, identify patterns and make decisions with minimal human interaction. As our profit optimization module is based on machine learning and artificial intelligence, we took the time to write down a comprehensive article on this topic.

Why is Machine learning so useful in pricing

Even though humans are capable of incredible feats, we have our limitations. We invented computers, developed them and will continue to develop them, at least to the point of singularity, at the moment there are already some things computers can do better than humans.

Computers are able to process a vast amount of data that would take humans a really long time and could even be impossible for humans to process manually. Where humans might be able to provide models on a category or SKU level for key products by processing them in Excel, a computer can do the same for all products, regardless of the amount. On a similar note, computers don’t have the processing limitations that leave only key products with optimized prices since computers can process entire inventories and reprice all SKUs globally.

One thing that surfaced as a fear when thinking about whether computers would take over human’s jobs was the fact that computers don’t really need to rest. They can work 24/7/365 whereas most humans work 9 to 5 and only 5 days a week, from Monday to Friday.

With the COVID-19 pandemic we have seen that changes in the market can be almost instantaneous. The demand for a specific product can explode in a matter of hours and the people benefiting from these changes are the ones who are awake and working when the changes occur. A computer is able to process the information in almost real time, making it possible to react to changes on the market rapidly. A human would need more time and effort to do this and sometimes still be too slow to take advantage of the situation.

As mentioned above, one of the key functions of machine learning algorithms is to learn patterns from data. In this situation, machines often spot patterns that humans miss or simply wouldn’t take into account. By using machine learning there is no need to program logic for the patterns and the constant integration makes detection of new patterns or changes in demand extremely fast.

If you look at the challenges of pricing and the benefits of machine learning it should come as no surprise that the reason we talk so much about it, is because by using machine learning assisted pricing YOU will become a better pricing professional. It is not about letting a computer take your job, it’s about using computers to your advantage by letting a computer do the work that computers are good at. Becoming a good pricing professional is largely based on having a good understanding of the market and insights into what is changing.

As we saw earlier, computers are good at processing large amounts of data and that is
exactly what a pricing professional needs when making pricing decisions. You need real time data about the market and a quick way to spot trends and changes that you should react to. Very few people enjoy making uninformed decisions and when it comes to pricing, there are many factors to take into consideration.

Here are some of the frequent questions that machine learning can help you answer:

• How were the sales impacted by an increase or decrease in price?
• How does the price increase/decrease in a product like lemonades affect related products like
juices (i.e. possible market cannibalization)?
• When a sale occurs, what is the likelihood of a customer also buying something else in the following days, weeks or months?
• Are promotional campaigns effective in activating customers?

By using computerized methods for tracking and modelling trends you can be sure that your decisions are backed by data, not by hunches and blind luck. Predictive models allow pricing professionals to test different scenarios and determine the best price for each product or service. You are able to evaluate the impact of sales promotions, estimate the right price if you need to take into account factors like best before date.

Predictive models can also be used for predicting price and demand for products that have never been sold before e.g. introduction to market situations. The more data you have and the more you are able to understand the consumers reactions to changes, the better you will be as a pricing professional.

When it comes to why you would use machine learning instead of traditional rule based learning it all boils down to the amount of possibilities that machine learning offers.

You are able to develop much more complex strategies than with rule based pricing strategies. As long as you define simple goals and limits your machine learning model will be able to adapt easily and live within the parameters you have set. Machine learning in pricing also makes adapting to crisis situations like the COVID-19 pandemic faster, increasing your business’s survival rate. The models can be adjusted to consider the near past more important than the older past and thus focusing more on recent times when making predictions for the future.

This blogpost is based on our e-book "AI in pricing"

Neural networks

Artificial neural networks (ANNs) are a set of algorithms that learn to perform tasks without being coded with task-specific rules.

A biological brain like the human brain is made up of neural networks. Neural networks consist of neurons; a single unit that is able to perform some very basic functions or actions. Computers are able to “simulate” functionality of biological neural networks with artificial neural networks that are in principle quite similar to the biological neural networks in our brains.

The neural networks used in artificial intelligence are inspired by biological neural networks. These artificial neural networks only tend to be simpler than their biological counterparts. As mentioned earlier, neurons are able to perform simple tasks when they are alone, but when they are combined with other neurons, they form a network and the tasks the network is able to perform suddenly becomes increasingly complex leading to a highly advanced and large amount of processing capacity. In a neural network, both in the biological and in the artificial, the task of the neuron is to receive and transmit signals and that is what they do, enabling a vast network to emerge.

“In neural networks, each neuron processes data independently from each other, making them able to process more info.”

In artificial intelligence, deep learning is a concept where several layers of simple processing units (neurons) are connected into a network where the input is passed through each of them one after another. When low-level data is processed within these layers, neural networks will adapt and learn and so called “intelligence” will emerge without e.g. a human programming explicit instructions.

One of the advantages of neural networks compared to traditional programming is that instead of traditionally programming in explicit instructions for a program, where a central processing unit (CPU) executes and processes these one after another focusing on one thing at a time, neural networks are able to process lots of information simultaneously. In neural networks, each neuron processes data independently from each other, making them able to process more info . You can sort of think of this as the difference between parallel and series circuits in electrical devices. One (series) runs after another whereas the other (parallel) runs simultaneously.

Like discussed above, all neural networks have a layer-like structure. This means that there is an input layer that gets the initial data and an output layer that produces the output of the whole network. In between these layers there are usually one or more hidden layers that take the output of one layer as its input and output that to the following layer as an input. To recap, all neurons in a layer get their input from a previous layer and output that to the neurons in the following layer.

Each neuron in the neural network has a set of adaptive parameters called weights. These weights are similar to linear and logistic regression and act as multipliers on the inputs to get a linear combination. In this situation using the word “weight” is really quite descriptive as it gives you an idea of what the function does: it places weights on different inputs and thus steers the process in a certain direction.

There are lots of different sizes of artificial neural networks from small to large to very large. Depending on the application, the largest ones can contain hundreds, thousands, millions or even billions of weights.

Let’s take a quick practical example and return to our hypothetical coffee shop presented in our Market cannibalization blog post. If the amount of cupcakes sold depends on two factors; the amount of frosting on the cupcake and price of the cake, you would need to study how the amount of frosting affects the purchase decision and the same goes for the price and after that you would be able to put a weight on both factors. It might be that the amount of frosting modifies the purchase decision threshold with double the intensity of modifying the price, but this might all change when we reach a certain price point. This is just a rough example but illustrates quite well how complex the data the neural networks need to be able to process as the inputs and outputs change when travelling in the layers.

After taking the weight into account and computing to a linear combination the neuron performs an activation function, which defines the final output of a neural network layer.

There are a few different types of activation functions, such as the following:

Identity function

The neuron does nothing to the value of the linear combination and passes it to the next layer, basically being exactly the same as a linear function.

Binary step function

Output 1 if above some value or threshold, else output 0.

Sigmoidal function

A “soft” version of a step function.

Rectified Linear Unit

A.k.a (ReLU) it is a special case of a ramp function (a “truncated” version of a linear function): 0 for values under a certain threshold and linear function for values above the threshold.

The final output of the neuron, i.e. linear combination + activation function, is used in getting a prediction or a decision. In a neural network, learning occurs when weights are adjusted to make the network produce correct outputs, quite similarly to classical linear regression.

The top layers in an artificial network need to be taught to the task using e.g backpropagation or the neural weights. Backpropagation is a technique, where in neural network fitting the gradient of a loss-function is computed using the chain rule (a mathematical formula to calculate the derivative of a composite function) with respect to the weights of the network for a single input-output example, one layer at a time.

The computation in back propagation is done backwards (like its name suggests) from the last layer in order to avoid redundant calculations in the chain rule. This makes back propagation more effective than naive direct computation of gradients.

When the neural network has learned and adapted and is deemed good enough by the person responsible for training it, it can be used for predictions.

The strength of the neural network is definitely the capacity to process and as well as the capability to be built in complex manners where they can take into account many different factors. This makes them good for automating price optimization if there are vast amounts of suitable data available. However, this is usually not the case and reinforcement learning can be more suitable to use in price optimization.

Want to read more about how AI will transform pricing?

Machine learning vs Reinforcement learning

Reinforcement learning is an area of machine learning where the machine learning models are continuously trained by making decisions in an environment and learning from the result. In reinforcement learning, the software agent adapts to make decisions often in uncertain environments and often faces a game-like scenario, where the outcome of a decision needs to be predicted in order for the “correct” decision to be made.

Reinforcement learning differs from supervised learning in that it does not require the labeling of inputs and outputs, nor does its suboptimal actions need to be corrected by the programmer or data scientist.

Rather, the model corrects these itself by adjusting the probability of trying the suboptimal action in the future according to the reward it receives from the action taken. In this sense, reinforcement learning reminds you a bit of the teaching technique where one is allowed to make a mistake in order to learn for the future instead of the teacher stepping in and correcting the situation immediately when the mistake is made.

The focus of reinforcement learning

The focus here lies in finding the balance between exploration (i.e. uncharted territory of the environment state) and exploitation (choosing the option which is thought to be optimal at the moment).

Reinforcement learning suits well for tasks that revolve around long-term vs. short term reward trade-offs, since the computer is able to make predictions for the near and far future based on what it has learnt.

The overall goal is to make the software agent to act as optimally as possible in the environment, minimizing regret (i.e. the difference in the performance of a trained software agent against an optimal agent) and maximizing the cumulative reward.

In reinforcement learning, we can use a couple of different methods making it a powerful tool in certain use cases. Samples are used to optimize performance, whereas function approximation is used to deal with large environments. This means that reinforcement learning can be very useful when we are dealing with a large environment with the following conditions.

Analytical solutions for the environment are not available and only a model of the environment is
known
Only a simulation model of the environment is given
Interacting with the environment is the only way to collect information about it

The conditions above perfectly illustrate why reinforcement learning is the ideal form of AI for price optimization. We often have (or at least have the possibility to gather) a lot of data and the environment (in this case, the market for the product or service) is large, but no clear theory of why certain things happen e.g. why a product sells better at a slightly lower price than a more reduced price.

In pricing, the best way to collect data is also to interact with the environment, in this case by adjusting prices, and learning from it. Furthermore, the market of products or services is often not static and can change over time as the environment changes (e.g. new technologies are developed and consumers adapt to these, changing their consumption preferences), which means that reinforcement learning allows the software agents to adjust their behavior accordingly with the changed environment.

Now that we’ve covered the basics of why reinforcement learning is optimal when you want to optimize prices, it’s time to dig a little deeper into the technical aspects of how reinforcement learning works.

The environment for reinforcement learning is typically stated in the form of a Markov decision process (MDP). The process starts with the software agent taking an action in an environment. This action is interpreted into a reward and a representation of the state of the environment. These are then fed back to the software agent. When the result is fed back, it allows for learning for the future as the agent becomes “aware” of the result of the decision it made. To support the explanation above, we can take a look at the following technical explanation of the process:

(courtesy of: Wikipedia)

A reinforcement learning agent interacts with its environment in discrete time steps.
At each time the agent receives an observation ot, which typically includes the reward rt.
The agent then chooses an action at from the set of available actions. This action is subsequently
sent to the environment. The environment moves to a new state st+1 and the reward rt+1 associated with the transition (st, at, st+1) is determined.
The goal of a reinforcement learning agent is to collect as much reward as possible.
The agent can choose any action as a function of the history, possibly randomly.
When the agent’s performance is compared to that of an agent that acts optimally, the difference
in performance gives rise to the notion of regret.
In order to act near optimally, the agent must reason about the long term consequences of its
actions (i.e., maximize future income), although the immediate reward associated with this might
be negative.

Implications on price and profit optimization

Now you are probably asking yourself; “how does this apply to price optimization?” Well, it’s really quite straightforward:

The environment is the current market where the product is being sold
Actions are changes in price made by the AI process
Changes in the environment’s state are the volumes sold when the price is a certain amount.
Reward is the amount of profit received by the AI process from the price change it made
If the environment changes due to external factors such as the COVID-19 pandemic, when the agent is performing an action, the change in the environment state and reward received can be different than previously in a situation and the agent learns from this, adjusting its behavior accordingly.

FAQ

How does machine learning specifically aid in addressing the challenges of pricing and profit optimization in dynamic market environments, such as those influenced by the COVID-19 pandemic, compared to traditional rule-based approaches?
- By processing vast amounts of data in real-time, machine learning algorithms can detect patterns and trends that humans may overlook. This allows businesses to react rapidly to changes in market demand, such as those seen during the COVID-19 pandemic, by adjusting pricing strategies accordingly. Unlike rule-based approaches, which rely on predetermined guidelines, machine learning algorithms continuously learn from new data and adapt their pricing strategies to optimize profits in evolving market conditions.
While the blog post discusses the technical aspects and advantages of using neural networks and reinforcement learning in price optimization, what potential limitations or challenges might arise when implementing these advanced AI techniques in real-world pricing strategies?
- One potential challenge is the requirement for large volumes of high-quality data to train machine learning models effectively. Obtaining and managing such data sets can be resource-intensive. Additionally, the complexity of neural networks and reinforcement learning algorithms requires specialized expertise in data science and computational techniques, which may not be readily available within all organizations. That’s why the use of easy pricing software is a good solution for retailers who don’t want to spend too much time thinking about it.

Learn more about e-commerce pricing

All things e-commerce pricing & price optimization right in your inbox. No spam.

The post Using machine learning to calculate price-elasticity appeared first on Sniffie.

What is price elasticity? A complete guide for your e-commerce business pricing

Jonas Timonen — Thu, 01 Apr 2021 05:48:00 +0000

When it comes to pricing and choosing the right price for a product, one of the first things you will stumble upon is the discussion of supply and demand. And why shouldn’t you? It’s the pillar that almost all pricing related theories are built around. However, you don’t need to dig very deep into the concept before you start to hear talk of elasticity, price elasticity and price elasticity of demand. Even the myths about price elasticity.

Price elasticity measures the way consumers, individuals or producers respond to changes in price with change in demand or supply of a product. Understanding price elasticity in pricing is important. You can do profit optimization easily in eCommerce with profit optimization algorithms.

Want to know the price elasticity of your products?

With Sniffie’s free Price Elasticity calculator you can easily calculate the elasticity of your prices and use the data to make informed and profitable repricing decisions.

Price elasticity can be divided into two different types

1. Price elasticity of demand

Price elasticity of demand refers to the degree to which the effective desire for something changes as the price changes.

2. Price elasticity of supply

Price elasticity of supply is a measure of how sensitive our quantity supplied is to a percentage change in price.

Let’s take a simple example to illustrate price elasticity. Think of a product like insulin. The drug is vital for people suffering from diabetes and having to live without it is not an option for millions of people around the globe.

In essence, changing the price of insulin is not going to reduce its demand and thus volumes sold. Of course we know that at some point no one will be able to afford the product any more but for a long time the price will be inelastic meaning that the price goes up and the demand stays the same.

Now take 1 euro for instance. If you offer someone the chance to buy 1 euro from you for 1 euro no one is going to do that (unless the buyer is a coin collector and your coin happens to be way nicer than his, in that case someone might go for it but other than that, it’s a no go).

Now what would happen if you sold 1 euro for 85 cents? Someone would probably already buy it. Lower the price further and more people would buy until you reach a point where the demand evens out. Still, the euro is a good example of an elastic price.

The elasticity of price is affected by both internal and external factors, in this case factors relating to
the product and other factors. The most important internal factors affecting price elasticity are:

Degree of necessity
Being time-bound
Accessibility of substitutes

The outside factors affecting price elasticity will be covered with a more detailed approach later in this article and in our eBook about AI in pricing. So let’s look more closely at the internal factors and use our example of insulin to illustrate.

The degree of necessity with a product like insulin is of course very high. The higher the need, the more inelastic the price is. People won’t stop needing insulin just because the price went up. If we look at something like fuel, it’s still considered inelastic but less so than insulin. People need fuel to commute to work for example but at some point when the prices rise people are going to find other means of transportation. Now if you look at something like indoor plants, they are a necessity for very few which means their price will be elastic and the demand will drop if the prices rise.

Preferences and buying behaviors may change over time which leads us to our second factor; price elasticity being time-bound. Let’s think of insulin again. Are people’s preferences or buying behaviors when it comes to insulin likely to change over time? Well no, not really. Unless you are cured, you’ll be buying insulin for the rest of your life. Fuel is a good example in this context as people have become more aware of the effects of global warming and some are already committed to driving electric cars or cycling to work.

Another example of how demand can change over time is the seasonal products related to weather. Bobsleighs are a lot more likely to be extremely elastic in price during summer but during a winter with heavy snowfall you might already be able to hike up your prices quite a bit before it will show on your sales volumes.

COVID-19 also provided a great example when people started hoarding toilet paper and hand sanitizer. These products suddenly had such a huge demand that some retailers were able to triple the price and still sell the products.

The last factor relates to the accessibility of substitutes. Now this is quite logical isn’t it? The more substitutes that are available, the easier it will be for a consumer to buy a substitute instead of paying more. If you have only one pharmaceutical company producing and selling insulin, they can hike up the prices a lot before people will stop buying but if you have ten producers on the market for this kind of drug, you are most likely all going to be pricing your products nearly exactly the same.

We’ve all been to a pharmacy and been asked if we want to switch out whatever the doctor ordered for a cheaper substitute from another manufacturer and quite a lot of us have at one point said yes.

Download Ebook AI in Pricing

Learn how to use AI to improve your pricing profitability.

What if the product seems inelastic?

Sometimes a product can seem inelastic and typically in those situations we are either too cheap (always selling a lot) or too expensive (never selling anything). Because of this we have not achieved the elastic curve and the only situation where the product’s whole elastic curve is truly inelastic is when the market is at 0 (i.e. no products are sold to anyone). This is what we mentioned earlier, even though insulin is vital, at some point it will be too expensive for anyone to buy.

What if the product seems elastic?

If the product seems elastic it can mean a couple different things. First it can mean that the price points that have been tested have all been in the “sensitive area” meaning that the changes have all led to change in demand. Another thing that can lead to a product seeming like it’s elastic is simply by chance.

The dynamic nature of price elasticity

When reading the text above, one might get the feeling that a product’s price is either elastic, inelastic or something in between. This however isn’t really the case. Price elasticity actually has much more of a dynamic nature that is present in two different ways:

Product lifecycle and price elasticity

It is vital to understand that price elasticity changes during product lifecycle. Products price elasticity changes because product maturity, market competition and other economic factors – such as inflation – change the value people perceive in your products.

Therefore the size of the elasticity or inelasticity depends on the price point based on how people perceive the price.

If we start by looking at the first statement; price elasticity changes during the product life cycle. This of course means that during the different stages of a products life cycle the product price can be either completely elastic, completely inelastic or anything in between.

Examples of product lifecycle and price elasticity

A new product introduced to the market surrounded by hype can have a huge demand, people don’t care about the price, they have to have it. At this stage the product’s price is nearing inelasticity.

The other extreme is when the product is at the end of its lifecycle, there is a lot of competition and your product only gets sold when the price is just right. In this situation the price is very elastic.

The value in understanding price elasticity is very much dependent on understanding that the price elasticity is so dynamic. Of course it would be nice if we could study the price elasticity once, decide on something and then use the same hypothesis when pricing the product in the future but unfortunately that is a bit too optimistic. Lucky for us, we have computers who can help us follow the elasticity of a product’s price and suggest how we should capitalize on the knowledge.

Understanding the amount of price elasticity

The other factor to take into consideration when it comes to the dynamic nature of price elasticity is that the amount of elasticity or in-elasticity varies depending on the price point.

If you look at price elasticity as a curve that is made up of price points and then connected to form a continuous line, it is easier to understand why the dynamic nature is present all the time. Now let’s move one step forward and start imagining price elasticity as something that there can be more and less of. Instead of thinking of it as a value that just describes the price’s elasticity, also think of it as a scale in between inelastic and elastic. Since the price evolves and price elasticity is calculated between two price points it means that the value defining the amount of elasticity is different depending on the price points used for the calculation. A product can be almost inelastic at some points and elastic at another, depending on the price.

Outside factors affecting price elasticity

As mentioned earlier, price elasticity is based on internal factors ( degree or necessity, being timebound, accessibility of substitutes) and outside factors. Now it’s time to look at the outside factorswhich can be boiled down to:

1. Market size
2. Market competition
3. Customer characteristics
4. Events in the world

Market size and price elasticity

If we start with market size, it is quite obvious that the amount of customers and the amount of actors affects the pricing and therefore the price elasticity. If you have a market of 100 people vs. 100 000 the difference between the customers is already something that is quite noticeable. In a smaller market you might have to think about capturing everyone on the market to be a success whereas on a large market that is rarely the case. The size of the market and its effect on price elasticity mostly boils down to how consumers differ and the more consumers there are on the market, the more different individuals there are who you have to get to buy your product.

Competition and price elasticity

Competition affects price elasticity by adding the element of having to consider at what price consumers are likely to buy from a competitor instead of from you. The more competitors, the more you have to focus on where your pricing strategy should be in comparison to your competition and the overall flexibility of setting prices tends to diminish.

Demographics affecting price elasticity

As mentioned above, customer characteristics are a big influencer. No two people are alike and the more potential customers your market consists of, the more different kinds of demands are you able to/required to meet and sell to. Changes in people’s way of thinking is one way that really affects price elasticity. We all know that being environmentally friendly is becoming a bigger thing and businesses selling eco-friendly products can see a more inelastic price compared to their not so-eco-friendly competitors. This is all due to the customers way of thinking about what is important in a product. Another good example is how certain brands have a luxury-image in people’s minds and customers are therefore willing to pay more since they have a stronger belief that a raise in price makes the product more desirable.

There is no more powerful example right now of how events in the world affect price elasticity than the COVID-19 pandemic. In a matter of weeks almost the entire world had to shut down, people retreated into their homes and prepared to stay there for a long time. This led to products that you normally buy once in a while without thinking being the most sought after and the demand skyrocketed when people started hoarding things like toilet paper, hand sanitizer, hygiene products and canned foods.

Before the pandemic, hand sanitizer was sold in moderation by pharmacies and grocery stores but as soon as the pandemic was a reality, suddenly every store wanted to carry something similar to hand sanitizer and while the hospitals were first in line for all produced products, small companies switched focus and started producing and selling hand sanitizer at a huge markup.

There is no doubt that price elasticity is as much affected by internal as external factors, making it challenging to estimate. Luckily AI and especially reinforcement learning has proven quite successful at generating forecasts in the most sensitive area, something we will discuss more in our eBook about AI in pricing.

How eCommerce Managers can utilize price elasticity

Segment your customers by demographics, interests, buying behaviors, income levels and their perception of your product – then use this data to come up with effective pricing and promotional strategies.

You also need to take into account how competitors are pricing products and how your potential customers perceive these products.

As a business, your end goal is to differentiate your brand in the market – this will happen when you can set the prices that your target audience identifies with.

For instance, Lamborghini (a highly price elastic brand) would have a much higher price point, but for the Ibuprofen (a medicine – relatively price inelastic), the price point would be considerably lower.

However, pricing strategies alone will not get you there – you need to conduct focus groups, surveys and employ other research techniques to help you find out your target audience’s perceptions, demographics, buying behaviors and interests pertaining to your market.

Supplement qualitative with quantitative research to make sure your pricing and promotional strategies are successful and so is your business. Let Sniffie help you find the right data and make sense of it.

Get in touch with us today to learn more.

Reference cases

Examples on how we deliver value

Nordic Wellness and Beauty Store

500.000 € more in profit in the first year alone.

AUTOMATION

Sexual Health Store

1032% ROI in just 2,5 months.

AI CASE

The post What is price elasticity? A complete guide for your e-commerce business pricing appeared first on Sniffie.

Price Elasticity – Sniffie

How to forecast discount campaigns with forecast module

What is Sniffie’s Forecasting Module?

How to Use Sniffie’s Forecasting Module

1. Select Your Product

2. Open the price simulator

3. Adjust pricing, cost and stock

Planning a Discount Campaign Using Sniffie’s Price Simulator

1. Establish Preliminary Discount

2. Adjust Discount Levels

3. Assess Stock Levels

4. Modify Stock Levels

5. Adjust Costs

Conclusion

Dynamic ABC Analysis: Powering Your Product Portfolio Management

Dynamic ABC analysis

A Deeper Dive into Dynamic ABC Analysis

Why machine learning and AI are the future of product pricing?

How to use Dynamic ABC Analysis in different pricing workflows

1. Filtering

2. Analytics and dashboard

3. Automate your pricing strategies

A Tool for Every Team

Pricing Automation for a Consumer Electronics Retailer

Streamlined pricing processes with pricing automation

5/5

Kari Kauhanen

Lumise

Need a Pricing Tool for Consumer Electronics

Case background

The challenge

Kari Kauhanen

CEO, Lumise

Try out our Pricing Automation Solution!

Why are consumer price index searches soaring in Google?

What is a Consumer Price Index?

How does it help companies to understand the environment they are in?

1. Identifying trends:

2. Analyzing customer behavior:

3. Using inflation expectations:

4. Combining with other factors:

How can you use Consumer Price Index in pricing?

Understand how your Cost of Goods Sold might change

Use inflation figures periodically

Understand how your competitors are affected by inflation

Make clear estimation of overall consumer demand and how your customers react

Understand your brand positioning and brand value

How to modify your distribution and promotion strategy in an inflation

FAQ

Most frequent questions and answers​

How to calculate a perfect sale price after the Omnibus Directive came into force?

The background for the sale price simulation

Show your Omnibus price automatically

Determining the customer profile first

1. Extremely price sensitive customers

2. Very price sensitive customers

3. Price sensitive customers

4. Price affects very little to these customers

5. Price has no effect to these customers

Use Our Free Price Elasticity Calculator

Specifying the sale price before the sale

Selecting the sale campaign discount %

Returning to normal pricing after sale campaign

Running the sale price simulation

Lighter pink

Green

Purple

Ensuring Price Display Compliance

Determine the market for Light Wooden designer chair

Setting the 30 day price before the sale

Finalising the sale price and sale discount %

Returning to normal pricing after campaign

The outcome of the sale scenarios

Use our sale price planning tool

Visual difference on the product pages

Analyzing the results to understand why the difference is so high

Use our FREE tools from Resource Library

Should you optimize individual SKUs or bigger optimizable product groups?

What is an optimizable unit?

Are you in need of profit optimization?

Most frequent questions and answers

Want to know the price elasticity of your products?