Building a Dynamic Pricing capability (in under 90 days)

A step-by-step guide for designing and implementing a dynamic, automated price-setting capability based on relative price position

Problem

Many executives in charge of pricing for mid-market retailers and wholesalers often struggle with manual, ineffective pricing processes that leave significant profits on the table and relegate capable staff to rote pricing tasks. These executives inherited pricing solutions that don’t quickly scale and rely heavily on expensive manual processes. The lack of insight and automation means retailers and wholesalers cannot effectively address ever-changing competitive pricing threats, leading to significant margin over-investment (priced too low) or a market share loss (priced too high).

When most people think of dynamic, automated pricing, words like “optimization,” “ deep learning, “ and “ black box “ emerge. Fortunately, it doesn’t have to be that complicated. You can build simple yet effective dynamic pricing solutions in 3 to 4 months that achieve up to 80% of its maximum potential Gross Profit dollars. You can design and implement it using practical methods and accessible technologies that empower your teams to take complete control without expensive 3rd party support.

A popular way to implement dynamic pricing capabilities is to develop optimal Competitive Price Indexes (CPI) for each Product Group, Pricing Category, and CPI Elasticity Group.

A CPI is a measure of our price position relative to key competitors. For example, a CPI of 105 means we are 5% more expensive, while a CPI of 95 indicates we are 5% cheaper.

Similarly, CPI Elasticity tells us how our unit sales change in response to our CPI change. For example, at a CPI Elasticity of -2, if our CPI goes from 100 (at parity with competition) to 102 (we are 2% more expensive), we expect our unit sales to decline by -4% (since -2 x 2% = 4%). In other words, think of CPI Elasticity as “Relative Price Sensitivity” (relative to relevant competitor prices).

The execution of dynamic pricing in this setting is straightforward. As the weighted basket of relevant competitor prices changes weekly/daily/intra-day, so does your price based on the optimal Competitive Price Index. It’s a highly effective method and quick to implement for Year 1 impact. It saves you thousands of hours of manual pricing labor. Implementing a CPI-based dynamic pricing capability allows you to reap fast Gross Margin $ wins for the first couple of years as you search for a more robust, long-cycle (and significantly more expensive) Price Strategy and Optimization solution (e.g., a price management software solution, implementing Value-Based Pricing, etc.).

Approach

Step 1: Understand the Problem

Understand the outcomes you want to drive with your dynamic pricing solution for each Product Group, and become very familiar with the competitive landscape. If you need to enhance your competitive price data assets, now is the time to invest in more robust competitor price scraping capabilities.

Meet with each of your stakeholders (especially Merchandising/Category and Sales leaders), and understand each Product Category’s goals: does the Category exist to maximize Unit Share, Sales Revenue, or Gross Margin dollars?

Lastly, learn from the points of failure from what your company or consultants have done before.

Step 2: Model Relative Price Sensitivities

BuildCompetitive Price Index Elasticity (CPI Elasticity) models for each product or product cluster within a Product Group. If you have enough transactional data for a product, you can build product-level elasticity models. However, the chances are that you will need more transactional data for ~ 50% of your SKUs. In this case, clustering your products based on attributes, purchase patterns, customer concentration, and other relevant metrics makes sense. Once done with the clustering exercise, you can build the price elasticity models based on a combination of products and product clusters.

Understanding your products’ price sensitivity to your relative price position will enable you to group products into CPI Elasticity Groups. It will be the foundation for your optimal CPI Index.

Step 3: Establish Pricing Categories

Within each Product Category, establish relevant Pricing Categories. While there are more out there, the four Pricing Categories we typically use are derived through a combination of analysis and heuristics (through conversation and alignment with Commercial leaders):

1. Anchor Items: the numbers vary, but typically ~10% of total product assortment, driving 25–50% of sales revenue (high velocity, high price)
2. Value-Perception Items: ~ 15% of SKU assortment driving 15–30% of sales revenue. These are high-velocity, low-priced items that are commonly known and popular with customers.
3. Assortment-Perception Items: selected categories or sub-categories of products for which your retailer or wholesaler is primarily known. ~10% of your assortment, driving 10–25% of sales revenue.
4. Background items: the long tail of your product assortment, ~ 65%, driving 35–50% of sales revenue.

Below is a quick visual of how these Pricing Categories play out in terms of assortment mix, sales revenue, velocity, and price elasticity.

Step 4: Architect the Solution

Build your Dynamic Price Optimization Matrix (DPOM) with the following building blocks and adjustment factors. Baseline optimal Competitive Pride Indexes will be established by Product Category / Pricing Group and CPI Elasticity Groups. Basic descriptive analytics or regression models are good enough to accomplish this by observing at what CPI intervals were Category Goals maximized for each Product Category / Pricing Group/CPI Elasticity group combination. The rest of the building blocks are adjustment factors that are applied sequentially:

Optimal CPI starting point by:

1. Product Category
2. Pricing Groups
3. CPI Elasticity Group
4. Adjustment factors applied iteratively:

• Category Goal
• Competitive Density and Intensity
• Seasonality
• Pricing Sentiment
• Online Cart Abandonment
• Premium Differentiation

Below is a sample illustration of the above methodology to create our DPOM:

Here’s what this may look like in practice:

Suppose we are setting the optimal Competitive Price Index for our High Price Elasticity products within a particular Product Category for our Wholesale or Retail business. Starting from our baseline CPIs for each Pricing Category (99, 100, 101, 102), we sequentially apply our Adjustment Factors. These adjustment factors are derived from basic diagnostic analytics and managerial heuristics (internal experts in Sales, Category Management, Merchandising, etc.). In our example, we end up with a CPI of 99.8 (i.e., 0.2% below weighted or average competitor prices). Note that this CPI applies to a product that is:

1. Highly price sensitive to movements in our relative price position.
2. Its role is to drive Gross Profit $ for the company.
3. The store or DC we are setting prices for has a Medium competitive density.
4. We are pricing during typical months (not during the high season)
5. Customers have complained about high prices for this product (negative Pricing Sentiment)
6. Cart Abandonment rates are high (B2C or B2B customers add the product to the cart but often delete it).
7. Lastly, since we provide some differentiated services relative to our competition, we charge a 75-basis point premium.

Step 5: Align on Execution Process

Collaborate with your stakeholders to build the automated price execution process. Typically, we have seen dynamic pricing setups with a manual review step for Merchants, Category Managers, or Pricing Managers for Anchor Items. In contrast, full automation for the other three Pricing Categories. It ensures that Commercial managers still have skin in the game. They can set prices for their most essential and visible product segments or accept the Dynamic Pricing engine recommendations. Meanwhile, they are getting a much-needed break from setting prices on the other 90% of the product assortment. To ensure seamless, automated integration of recommended prices with your pricing execution engine, your best friends will be the CIO and his organization.

Step 6: Build your Minimum Viable Analytics Solution

Build a Minimum Viable Analytics Solution (MVAS) and review it with all your key stakeholders and end-users. Incorporate critical pieces of feedback and turn around in 2–3 weeks a Beta solution ready for piloting.

Select the appropriate test and control markets (could be groups of stores for Retail or Distribution Centers for Wholesale) and conduct a 6–8 week test of your Dynamic Pricing Solution.

Step 7: Align with Stakeholders

Create alignment throughout the project with your stakeholders through iterative roadshows demonstrating work in progress and incorporating their feedback. These roadshows will typically occur in four stages, culminating with the final solution launch and ongoing post-launch updates afterward:

1. MVAS review and feedback session
2. Beta solution review and feedback session
3. Pilot design (A/B testing) and rollout plans
4. Pilot review and full-scale launch plans

Once your solution launches full scale, you will lead periodic updates woven into the business’s regular operating rhythm. You’ll need to remain hawkish about evaluating business results on an ongoing basis and adjust your Dynamic Pricing Architecture when required.

Step 8: Launch Full-Scale

Launch the full-scale solution with a clear, easy-to-understand Dynamic Pricing Guideline (think of it as an online resource). As a valuable daily resource, you will hand this guideline over to Pricing, Finance, and Commercial teams (Sales, Supply Chain, Category Management, and Merchandising). Having a living, breathing document that is always up to date and easily consumed by critical stakeholders will build trust with the Dynamic Pricing solution.

Step 9: Assess and Augment

Assess the business impact and adjust the Dynamic Price Optimization Matrix architecture as needed. At least annually, but ideally, semi-annually, update your price elasticity models or Pricing Categories as sales and competitive trends change.

Measurable Outcomes

Deploying a CPI-based dynamic, automated pricing for your everyday pricing execution can improve your Revenues and Gross Profit $ by up to 5% and 15%. For most companies, a 1–3% Gross Profit $ increase in Year 1 is more realistic.

It can enable you to increase profits without sacrificing your market share position. It can also help you and other Commercial leaders understand the impact of price change proposals on unit volume, market share, and profit.

Finally, it can yield substantial labor cost savings by not spending thousands of hours doing manual price setting and execution. You can eventually reallocate your existing staff to work on more strategic priorities instead of chasing manual, repetitive tasks.

Remember, simplicity often wins, and this is an excellent example of that.