Understanding Media Attribution to Sales

A common question I often get about media attribution is, “How do you attribute media to sales?” This query always excites my inner data nerd! My typical enthusiastic response is, “We fit a hierarchical Bayesian model that relates adstock saturation transformed media to sales,” which usually leaves everyone more confused.

So, what does that mean in plain English?

Let’s break it down.

Hierarchical means we are stacking similar data together, such as by region, and sharing insights across these groups. This allows us to draw more robust conclusions by leveraging similarities and differences across the data.

Bayesian refers to the statistical approach we use, which not only helps us learn the effect of media on sales but also measures our confidence in these effects. It provides a probabilistic framework, giving us a richer understanding of the data and its uncertainties.

Adstock captures the time-delay effect of media. Most advertising has a delayed impact on sales, and adstock helps us account for this lag.

Saturation accounts for diminishing returns. After a certain point, additional media spend has less and less of an impact. The saturation function helps us adjust for this nonlinear effect.

So, in simpler terms, a better explanation might be “We adjust media for delayed impact and diminishing returns, then relate adjusted media to sales using a model that shares learning and estimates uncertainty.”

By adjusting for these factors, our models provide a more accurate and nuanced understanding of how advertising translates into sales. This approach not only helps in pinpointing the effectiveness of different media channels but also enhances our strategic planning by understanding the varying impacts across regions.

Understanding these concepts can significantly elevate your media attribution strategy, providing actionable insights and driving more informed decision-making in your marketing efforts.

We’ve built the industry’s most innovative marketing optimization and scenario planning software:  Optimetry® that puts your internal media experts in the driver’s seat, for dynamic “what if” scenario planning and spend adjustments over the life of your campaigns.  Learn more: https://in4ins.com/optimetry/

 

Why is Retail Media ROAS so Hard to Measure?

There are a lot of signs that trade and retail media networks (RMNs) are converging. This is driven on the one hand because retail media and trade are increasingly financed from the same source: sales co-op funds. On the other hand, they are both converging on the lower funnel – the first moment of truth. Why would tactics so close to the point of purchase be so hard to measure?

To make sense of this and to understand why retail media ROAS is so hard to measure, we have to look at the simple formula: ROAS = Contribution/Media Cost. When evaluating the profitability of retail media, there are problems with both the numerator and the denominator. The cost (denominator) is often higher for the same service (programmatic or social)  if you buy it through the retailer versus buying directly. As a client-side friend put it:

“CPMs for RMNs are (basically) made up. They’ll charge you $18.00 CPMs for a Facebook buy that is completely disconnected to the actual CPMs in the auction. They just pocket all the overage and margin.”

On the other side of the table, an industry consultant reported a Walmart executive saying:

“I’m glad that Walmart connect generates such high margin income for us to offset losses in other parts of the business.”

What is happening in practice is that the RMN costs are being bundled together with other fees as a cost of doing business with the retailer. So, in truth, there are other goodies hidden in the RMN cost such as slotting fees, preferred display locations, etc.

The other problematic part of the ROAS equation is in the numerator – the contribution or lift from media. With most performance media, the drivers of contribution are audience reach and frequency, sometimes at DMA level. With RMNs the media is driving offline sales within the retailer’s marketing area. It is also focused on specific UPCs. This specificity is a two-edged sword: it makes the impact easier to derive because it is so specific, but it also reduces its scope. In fact, only the biggest retailers (e.g., Amazon, Walmart, Kroger, Target, Best Buy, Instacart) have the scale to drive enough traffic to read.

The current practice for deriving the effect of retail media is to treat it just like any other performance media variable in a marketing mix model without paying attention to details of tactic, geography, or the mix of online/offline sales channels.

With all that in mind, here are of the top considerations to improve the quality of the estimate of lift (the numerator of ROAS):

  1. If the retailer is one of the top RMNs (excluding Amazon), then it makes sense to derive the retail media impact on sales within the trade RMAs. This requires access to chain or RMA-level data.
  2. If the RMN advertising is for specific UPCs or sub-brands, it makes sense to estimate the sales impact at least at the sub-brand level. Because RMN data moves through media systems, it is not easy to obtain UPC-specific data. Text processing is usually required to infer the sub-brand from various identifying strings.
  3. Since all these sales impacts are proportional to size of retailer marketing areas and of the sub-brand, a multiplicative marketing mix model will provide much better results than an additive model, since its effects are automatically scaled. This is similar to the way that multiplicative models are more accurate for trade promotions.
  4. The impact of adstock/saturation must differ by tactic. Onsite search is lower funnel and short-acting whereas offsite media can be much longer-acting.
  5. There are many types of impacts on sales to consider, including both direct and halo. This table shows the relationship between the type of media (onsite vs offsite) and the sales impact as we move from the most direct (onsite search driving online sales) to the most attenuated (offsite media driving sales in the same high-intent lookalike target anywhere).
Media Type Direct Halo Secondary Halo
Onsite Online (DTC) Offline sales of same retailer Offline and online sales of other retailers (informational search)
Offsite/in-App Offline and online sales of all retailers

Using advanced modeling systems such as Robust Hierarchical Bayes that take geography and UPC into account, we can accurately track the effectiveness of RMNs (the lift per impression). DTC (onsite search/display driving online sales) has the most direct attribution path. Offsite targeting is usually less effective although there are exceptions. The ROAS depends on transparency in media costs – which at the present time is caught up in the vendor-buyer negotiation and escapes rationalization.

If you would like to discuss how our approach to Robust Hierarchical Bayes can help you better measure your Retail Media investment (and your marketing mix performance as a whole), schedule a call with the i4i Team.

Attribution Analysis Letting You Down? Why Not Return to MMM?

Marketing attribution analysis was great in theory. What are the things consumers see and hear that influence and prompt them to act, specifically to buy a product or service?

It would be nice if it were simple and binary: Consumers see this, they buy that.

But let’s start with the essential data and the connections brands need to make, to support decisions that impact media planning effectiveness.

Visibility into what a consumer bought: Possible in digital (if brands own their own data or can get it easily from e-commerce distribution partners.) Harder in physical stores. In both cases brands need first party data from the retailer to know what was actually purchased.

Visibility into what the consumer did to get to the point of purchase: Did they go to a retailer’s website and search for the product? Did they see a digital ad, click on it, and it took them to an e-commerce page? Did they see the product on their neighbor’s kitchen counter and make a mental note to purchase it next time they were in the grocery store?

Visibility into what advertising campaign was seen or heard before the purchase: Was it the digital ad on their favorite app? Was it the TV ad, or the OOH ad on a bus shelter? Was it an audio ad on a podcast or on the radio during their commute from the office? If it is digital media and a digital purchase with accurate UTMs embedded and an effective tracker to see what messaging converted, an advertiser may have half a shot of making a connection of the channel and the campaign to the actual transaction. But it is not a precise media planning tool, particularly if the campaign intent was brand awareness, versus activation.

Connecting all these data points to create optimal media plans

Here’s the dilemma. Attribution modeling has been described as limited, in part because it may account for one or two online channels compared to each other, but in terms of all other media and marketing exposure it can’t definitively say which is THE ONE that can be attributed to the purchase. What is more likely is a combination of factors in aggregate, in context of other behavioral conditions, drove the consumer to put the product in their virtual or physical carts.

Folks at the digital marketing agency Atomicdust back in 2021 wrote an interesting piece provocatively entitled “Marketing Attribution is Killing Common Sense”.

Still quite relevant, it contains some of the better definitions we have seen on the types of attribution (“last interaction”, “last non-direct interaction”, “linear attribution”, “multi-touch”, etc.) commonly tracked in attribution modeling. More importantly, the author basically describes how each of these models as flawed in their own way.

Why are they flawed? Two reasons, actually.

  • Attribution models are primarily limited to digital marketing, which is highly dependent on the infamous cookies that are going away due to privacy concerns.
  • These models don’t reliably accommodate the attribution of sales that are conducted in physical retail settings, in which case an advertiser may have no idea what was the last (or first) ad the shopper saw.

Frankly consumer behaviors are sufficiently complicated as to defy predictions based purely on the last or first impression of the product. Everything from seasonality and weather to time of day (particularly in the case of impulse buys at quick service restaurants or convenience stores) will be influenced by the confluence of when and how an advertising message is delivered in any given media channel.

So, if not attribution modeling, what can brands employ that is more evidenced based than “common sense”, to make better media buying decisions?

The Return to a Trusted Tool: Marketing Mix Modeling

FMCG companies have relied on Marketing Mix Modeling since the 1960s. Of course, retail options and even the media channels available to advertisers were far less complicated then. Decisions on how, where and when to plan media spend could be done at a comparatively more leisurely pace than the frenetic programmatically driven speed required in today’s digital environment.

But the beauty of MMM is its inherent data analytical sophistication, capable of incredibly complicated models with far more degrees of granularity, is that it can tackle the combination of many new kinds of data feeds at different levels of specificity (target audience, time granularity, geographic unit, type of metric, etc.) And this level of complexity requires data analytics robust enough to parse the results at a much more sophisticated level to much better media planning support.

As retail has also evolved to include far more distribution outlets (including e-commerce) and media followed suit with the introduction of digital advertising, MMM provides the perfect solution to making highly complex decisions about media buys. But you need accomplished data scientists running the models, equipped with the expertise to interrogate the algorithms with a brand’s specific objectives in mind.

  • MMM when done right, can provide far more detailed road maps for achieving ROAS (return on advertising spend) encompassing virtually endless variables that can impact spending.
  • MMM done right can provide guidelines for spending just enough in any one media channel so as not to overspend in terms of overall return on investment.
  • And we could (and will) dedicate and entire post to the unique benefit of MMM to facilitate targeting not only the many more channels in the new world of media used by people, but the situation and the journeys that got them there. Attribution never really fulfilled its promise to get a whole lot closer to this level of targeting visibility.

Brands are moving away from strictly asking themselves: “Should we spend less on TV and more in digital?”, to the much more relevant question: “What’s the role of the big-screen in my campaign and which new channels are going to replace the ubiquity of TV?” To guide the resulting media planning choices, the best MMM provider/partner will actually surface the more discrete evidence as decision support. The ideal MMM partner will reveal which of all the digital and other channels available will yield the best contributions to ROAS compared to each other, so media buying decisions can be more fine-tuned and effective.

Attribution may be letting you down. But have no fear. Our data analyst teams at in4mation insights are prepared to help you tackle media planning, delivering optimized scenarios to achieve return on your investments and ongoing growth for your brands.