For CPG brand suppliers, presenting retail partners with new pricing and promotional strategies can be complex. Here dunnhumby’s head of insight sales, Moray Duncan, reveals how to tackle key forecasting challenges.
As anyone who has ever sat across the table from a retail customer can confirm, there’s a big difference between brilliant theory and practical reality – particularly when it comes to pricing.
Proposals might look good on paper, but as a supplier, explaining why there is a plan to raise base prices or downweight a SKU often requires more than just a robust forecasting model. It demands proof that the path chosen is the right one.
For most large consumer packaged goods (CPG) brands, finding that evidence is a task that usually sits with their net revenue management (NRM) teams. As they start planning their growth models for the two or three years ahead, those teams will take a long, hard look at the possibilities that exist around pricing and promotions, and then present their recommendations to the wider organisation. Typically, that’s where things start to get complicated.
Key forcasting challenges
Today, a vast number of people are acutely invested in a CPG’s pricing strategies. In addition to the revenue specialists mentioned above, there are plenty of others to add to the list. From commercial teams to category teams – and even brand teams on occasion – price (and promotions) are an increasingly cross-discipline concern. Then, there’s the retailer itself to consider too.
This is where that need for hard evidence comes in.
Not only do CPGs need to be able to demonstrate to their retail partners that they’ve chosen the right way forward, but they also need to be confident in that decision internally. That’s particularly true for commercial teams, who often have the unenviable job of pitching any modifications in their strategy to their retail partners.
As important as it is, though, proof can also be very hard to come by. Most of the forecasting models used by NRM teams today require a number of assumptions in order to get from A to B, which means that hard evidence is usually lacking. Any proposals might look good on paper, but it can be hard to convince a retailer of their validity when all they have to go on is an assertion that “the model works”.
Evidence isn’t the only element of forecasting that’s tricky, of course. The sheer number of possibilities that exist make it an inevitably complex task, with a web of interconnected dynamics ensuring that even small alterations to one aspect of a forecast can have dramatic consequences elsewhere.
What would happen if CPG brand supplier scrapped a half-price offer and replaced it with 3 for 2, for instance? What if they kept that offer in place, but reduced the frequency down from four promo periods to three? What if they left their promotional mechanisms alone, but cut the number of items in a multipack by 10%? What if they took three grams out of every product as standard? The permutations are endless.
All of this leads to three major problems for CPGs:
- Forecasting is slow, laborious and expensive
With so many possibilities to consider, and so many different combinations to try, finding the best possible mix of price, promotions, and product is a laborious process that requires a lot of manual effort and extensive iteration. Inevitably, that can represent a significant cost to the business, particularly when senior expertise is required to run that process effectively.
- The potential for error is enormous
The more assumptions that go into a forecasting model, the wider the corridor for error becomes. That’s before issues like bias and human error come into play, which only heighten the risk that the ultimate results end up skewed. While any inaccuracies will have consequences for a brand’s own projections, they’re likely to have serious implications for their relationship with their retail partners too.
- Convincing retailers is harder than it needs to be
As discussed above, many commercial teams find themselves in the uncomfortable position of making proposals based on (well-informed) assumptions rather than actual science.
Getting ahead of the curve
Together, these issues helped to inform the development of Revenue Growth Planner, a self-service forecasting tool that gives brands an understanding of the likely response to changes in their pricing and promotional architectures.
Underpinned by three years of transaction data from a UK retailer with over 40 million transactions a week, Revenue Growth Planner also makes use of a powerful machine learning model to generate fast and accurate insights. In combination with deep, basket-level data, that algorithm brings a new level of scientific rigour to the forecasting process.
Using Revenue Growth Planner, brands can tackle their key forecasting challenges head on:
- It’s fast and time-efficient, giving them quick answers on what price changes could mean for sales and profitability.
- It’s accurate and reliable, capable of handling cross effects and tracking the impact of changes without the risk of double counting.
- It brings confidence to your forecasts, employing real-world data from billions of shopper baskets and enabling you to test and compare different scenarios on the fly.
As a result, Revenue Growth Planner gives CPG brands the ability to make better, faster decisions – and ones that are based on hard science rather than gut feel or received wisdom. Not only does that help to drive more strategic and productive conversations with retailers, it does the same internally too.
Whether it’s category teams looking to get ahead of the curve on upcoming consumer trends, commercial teams who are navigating an increasingly complex supply chain, or NRM teams trying to pinpoint the most profitable future position, Revenue Growth Planner brings clarity and consistency to the conversation. With everyone on the same page, and with accurate and reliable data, everyone can be confident in what the right way forward really is.
