Are you measuring your shopper marketing programs? Do you even have the capability? “On the surface, most companies want to measure this investment, and they should, but efforts to do so seem to be limited,” says Mike Anthony, CEO of Hong Kong-based consultancy Engage, adding that to some extent companies are almost happy to keep this as “the elephant in the room.”
The massive amount of spend should be reason enough to measure. Not to mention that, as shopper marketing has become a larger part of the marketing mix, it comes under increased scrutiny. Measuring can show accountability on what you spend, justify shopper marketers receiving a bigger piece of the marketing budget, and validate requests for increased resources. “In the new world of zero-based budgeting, the accountable marketers are succeeding because they are justifying their budgets to management,” says Rick Abens, CEO and founder, Foresight ROI.
But for many marketers, measurement is not easy. In a recent GfK survey, more than half of respondents identified measuring ROI as the biggest barrier in the shopper marketing discipline, says Sarah Gleason, GfK senior vice president.
Anthony, Abens, Gleason and other industry leaders recently discussed the topic of measurement with Shopper Marketing. Here is an edited virtual roundtable discussion:
In addition to justifying spend, why is it important to be able to measure shopper marketing?
LILY LEV-GLICK: To get all the partners involved on the same page as to what the specified, identified measurable criteria is for success.
ABENS: To create the most impactful marketing programs by knowing what works.
ANTHONY: Because shopper marketing is key to brand growth. If shoppers don’t buy more, consumers can’t consume more.
MARLO CRUZ-SANDS: Shopper marketing should be influencing and/or changing purchase behavior. You must know if it is and by how much.
Do you find that CPG companies want, or expect, measurement of their programs?
LIZ CRAWFORD: Many shopper programs benefit brands that are too small to warrant the expense of elaborate measurement. So, simple measures – like lift and percent display – are often used as a gauge of success.
LEV-GLICK: I think they do, but they are kind of skeptical that they will ever be able to do it well.
TIM O’CONNOR: They do. The challenge is to do so accurately, and that requires attribution and significant broader understanding of what engagement and activation efforts actually drove what behavior. CPG firms that get too obsessed with this can get lost in the weeds really quickly or buried in data with little result.
What do you get asked the most from CPG companies about measurement?
ANTHONY: To be honest, it doesn’t get raised as much as it should.
ABENS: How do you separate shopper marketing effects from trade promotion effects?
LEV-GLICK: How do we optimize a program we’re about to launch? Can you predict how this is going to perform?
Is it even possible to accurately measure shopper marketing?
GLEASON: This is the $64,000 question. Many companies have built their businesses on models for measuring shopper marketing programs. And many companies investing in shopper marketing programs have partnered with the measurement suppliers finding ways to measure shopper marketing. Yet there still remains some level of skepticism in the industry as to the credibility of some of these available measurement approaches.
ANTHONY: Yes. The starting point is to set meaningful objectives.
ABENS: Yes, using an econometric model specifically designed to measure shopper marketing tactics. The important thing is to isolate the effects of each shopper marketing tactic from all other demand-building effects like pricing, distribution, merchandising, brand advertising, competitive effects and external factors.
Is ROI the only measurement? What about return on objectives, on marketing objectives and on relationship?
LEV-GLICK: Those are all really great but still kind of esoteric in a way. I haven’t seen hard evidence of measurement. People are so driven by bottom-line dollars.
O’CONNOR: More qualitative measures are of course relevant but need to be honestly prioritized relative to ROI and market realities and not used to justify misspent resources.
GLEASON: We should think very broadly about types of measurement. ROI and sales lift is clearly important. Equally important is the impact of your shopper marketing efforts on long-term brand equity and the strength of relationships with retailer partners, although measuring these can be all the more challenging.
How do you measure shopper marketing?
ABENS: The first stage we call the cost model, which quantifies and predicts how the marketing actions or spending converts to marketplace reach. Then we use an econometric predictive model to estimate the sales lift from the marketing reach for each tactic along the path to purchase and for all of the other demand-driving variables like brand advertising, seasonality, merchandising, distribution, competitive effects, weather and the economy, as well as others.
GLEASON: Our preferred approach is through the use of retailer loyalty card data. This is most effective when we can isolate the timing and location of shopper marketing activity or shoppers that are either exposed to or respond to programming. In addition, we have emerging solutions to measure in-store conversion, both at a macro, total-store level and a micro, category-specific level.
How do you go beyond transaction data? What about shopper behavior?
CRUZ-SANDS: Transaction data is the ultimate reflection of shopper behavior. Where a consumer puts his/her money tells marketers what they really want to know.
CRAWFORD: Increasingly, marketers are turning to video tracking and heat mapping to look at shoppers’ behavior in aisle. Online, we can look at their digital footprints. Looking at “tracks” that shoppers have left behind will become increasingly important to understanding effective messaging at point of sale.
LEV-GLICK: Transaction data wedded with the shopper data – that’s when you see the whole picture.
GLEASON: Insights into the shopper behavior associated with the transaction will provide direction as to whether the program worked and in what way. It is the why behind the what.
What role do predictive methods play?
LEV-GLICK: Predictive tools give us a heads up – if you go this way, this is what you can expect to happen. It gives you a chance to avert an issue. Arguments I hear are “I don’t have the money, I don’t have the time.” But there’s always an existing tool or methodology that can be used. It doesn’t need to be a million-dollar project, and it’s better than doing nothing.
CRAWFORD: The role they play will increase many fold in the next few years.
GLEASON: Always helpful but only as good as the assumptions we apply.
ABENS: To me, everything in measurement is about being predictive because it’s too late to do anything about the past. Measurement must help us make effective decisions that will affect future outcomes.
How has shopper marketing measurement evolved during the last 5 or 10 years?
LEV-GLICK: From not even thinking about how to measure to “I do need to measure, and measure what matters for my culture.”
ABENS: As the amount of investment in shopper marketing has increased, the stakes have gotten higher. When the budgets reach a certain threshold, management notices and starts asking what we are getting from this investment.
CRAWFORD: There are mobile surveys and passive data collection techniques that are emerging now. These new tools are still in their infancy but will grow increasingly valuable.
ANTHONY: Very little. There are plenty of tools and loads more data, but most of this focuses on the commercial aspects and not enough on the behavioral aspect.
What else should shopper marketers think about when it comes to measurement?
GLEASON: How as an industry can we get to a common set of metrics and what role do retailers play in helping to provide the data that can help manufacturers better justify the programming that benefits them both.
CRAWFORD: What kinds of passive data collection are we ignoring? Where can we see behaviors and attitudes in digital footprints?
ANTHONY: Too much discussion flounders on the cost of measurement. The real question should be, what is the cost of not measuring?