How impactful is your innovation?
This summer edition is dedicated to a topic at the center of the growth strategy of virtually all our clients. For many of them, launching efficient, sustainable, and truly incremental innovation is not just a pillar of their growth algorithm, it is critical to their long-term survival. We will share here how we approach the first question that we answer when a client engages us on an innovation project: how impactful your innovation has been so far.
In “The Latest” section, we share how we have recently leveraged lean and agile primary research (yes you heard it right, I did say primary research, not artificial intelligence, or machine learning!) to add power to our dynamic models.
Make sure to read until the end so that you can meet our three latest additions to our team!
Six Steps to Measure Innovation Impact
Innovation is the lifeblood of consumer products. Brands that fail to innovate risk losing brand equity, market share, retail shelf space, and profits. Despite this, brand managers often struggle to successfully bring innovation to market in a way that succeeds in growing the overall business.
There are a myriad of factors, both internal and external, that can impact the success of innovation – understanding consumer trends, R&D investment, a clear pipeline and portfolio, manufacturing capabilities, retailer partnership, and cohesive brand strategy to name a few. But let’s set those aside for now. Those are components of a solution; first, let’s diagnose if there is a problem: do you know how impactful is your innovation?
In this newsletter, the first in a multi-part series on innovation, we will begin with six data-driven steps you can take to measure how successful your innovations are today and over time.
1. Define innovation:
It may seem silly, but this is often the most controversial step. Many analyses will start with a simple definition: innovation is a new SKU introduced within a calendar year. However, this can quickly lead to complications. For instance, not every new SKU is really an intentional “innovation.” Most new SKUs tend to be minor changes in pack size that are used for revenue growth management, or for a specific retailer’s shelf needs. For the purposes of innovation analysis, it’s best to consider them part of the core business, or to track them separately all together.
There are also different types of innovation: new sizes, new bundles, new forms, new variants (flavors, scents, ingredients, benefits, uses, applications, etc.), new sub-brands and new brands all together. There may also be classification schemes based around the intent of the innovation: sustaining vs. transformative. No matter what, you will need to take a close look at your data to make sure that you have correctly identified and appropriately classified what you want to consider innovation for the purpose of the analysis.
2. Assess incrementality:
Okay, you’ve tagged innovation SKUs in your database. Now we need to understand to what extent innovation is driving growth. Incrementality can be controversial in an organization. But at the simplest level, we are trying to understand if the sales we get from all the innovation we launched made up for the sales we lost from cannibalizing or discontinuing the existing core products.
We often like to start at the highest level possible – total manufacturer. This helps us understand if we’re robbing Peter to pay Paul between brands or categories within the business. Purchase structures, which use analytic techniques to establish the hierarchy of product attributes differentiating consumer preferences within a category, and therefore which subsegments are most substitutable for each other, can also be helpful in identifying where to draw the line for a more granular view. Depending on what you define as the total base of the business, it can be helpful to net out category trend rates first. For example, if the whole market was on track to grow 10%, and your innovation only delivered a net 5% growth, then that wasn’t successful, was it?
This analysis also becomes more helpful if completed at the channel or even customer level, to account for unique dynamics. Considering incrementality in terms of profit vs. revenue vs. volume can also be illuminating.
3. Isolate innovation drivers:
Innovation impact is not monolithic. There are a few specific KPIs that are important to decompose to understand why your innovation is succeeding or failing: the # SKUs you launch, the distribution per SKU, the velocity (volume per SKU per distribution point), the premium of the SKUs ($ per equalized volume), and the sustain rate (% of SKUs launched that still exist in the following year, and the year after that, etc.).
Deficits in any one of these factors can tank your innovation plans. We’ve found that the two most common issues brands face are efficiency – trying to execute too many low volume SKUs instead of a few productive big bets – and sustainability – investing in lots of in and out products that aren’t designed to last. The two combined can lead to a long tail of zombie poor performers unless the manufacturer has a disciplined pruning strategy: the ability to identify and quickly phase out successful innovation (spoiler alert – they usually don’t). Even if you have an intentional strategy to drive short-term newness, it is critical to understand just how much (or little) you count on these “in and out” innovations as a percent of your business each year.
4. Benchmark, benchmark, benchmark:
While these factors are helpful in explaining the impact your innovation has, they are best used as a comparative lens. Using this framework you can understand innovation’s impact between brands, between channels, between markets, and even versus competitors using Point of Sale data.
Benchmarking your brand to the rest of your portfolio or to competitors will give you a strong sense of both what’s possible – how high is high for any of these KPIs – and in terms of what’s feasible – what would be a reasonable assumption for improvement, based on what you observe in the market over multiple years. Driving growth through innovation is hard; you may find that your innovation is working pretty well after all!
5. Identify rules for success:
Your innovation database can also be used to model the probability of success for any given innovation. Of course, success itself is a term that needs to be defined – that will depend on your organization. Common measures would include things like: the ability to sustain at least one year, reaching a minimum volume threshold or market share target. Once you’ve aligned on a definition of success, you can use analytic tools to identify the commonalities of innovation programs that reach those thresholds. This may lead you to focus more on certain KPIs (for instance velocity of your innovation) or on certain attributes (for instances, new brands). These commonalities can be used to set clear guidelines and standards for your future innovation launch forecasts and investments.
6. Check forecast accuracy:
After doing all of this, you’re probably about ready for happy hour. But there is a crucial final step to determining success of innovations – and that is the innovation forecasting process. Up until now, we’ve looked at how to use historical data to measure the impact of our innovation, identify areas for improvement, and to set guidelines. But all of that won’t help much if the forecasts are completely off from how the innovations actually perform. Bad forecasts can lead to misalignment with retailer and distributor partners, out of stocks, and unproductive inventory.
Forecasting a new innovation can sometimes feel like a fool’s errand, after all it’s a known unknown, but analytic modeling techniques based on history and common program attributes can be used to get smarter. And in many cases significantly improve the margin of error. Common differentiators may be things like geography, category, innovation type, and level of investment.
We hope you’ve enjoyed part one of our series on innovation. Now that we’ve got a strong analytic toolkit for measuring innovation’s impact, next quarter we will focus on identifying platforms, or what exactly our innovations should be. Stay tuned!
Latest at Foresight
As we continue to push towards our vision of “no unanswerable questions” in marketing strategy, we have started to deploy more custom research in our projects. While we still stand behind the rich insights that triangulating existing client data sources can provide, we’ve found that adding primary research can help close data gaps, provide new learnings, and ultimately make our models more robust and actionable – at a cost that is a fraction of what it would have been a decade ago.
Here are a few examples of ways we’ve used primary research to enhance our analytics:
- Creating “shopper profiles” that are tied to commercial channels for pack price architecture decisions
- Sizing the feasible growth opportunity from our “bath tub” consumer growth lever models
- Segmenting product categories based on which occasions they are most relevant for within specific life stages
Custom panel research, particularly within one market, often does not add significant cost to a project but can unlock enormous value in the existing data sources and analytic approaches we use for any client. Let us know if you want to learn more about how primary research can support and enhance our next project together.
Welcome new team members:
We are pleased to introduce Adam Acker, Zach Ziegler, and Melissa Cooklock. Adam and Zach join our consulting team as an Associate and Business Analyst, respectively. Melissa is our new Office and Event Coordinator.
Catch up on our latest work: The Latest