Marketing and Research Consulting for a Brave New World
Subscribe via RSS

Marketers are caught between the need to drive short term ROI (which drives them to use promotional advertising) and the belief that too much short term focus could mortgage their brand’s future.

Unfortunately, current marketing research and analytics tools rarely provide a “full effects” view of advertising.

  • Return on Ad spend (ROAS) studies and marketing mix modeling focus on short term sales lift and miss longer term brand building effects.
  • Typical brand and campaign trackers focus on longer term funnel metrics but they often are flawed, leading marketers to false conclusions and mediocre ROIs (for reasons I will describe).

Why is a “Full effects” measurement approach so hard to create? Because a key concept is missing from all research I have been asked to review!

What’s missing is this. Think consumer journeys. All potential users of your product or service can be segmented into two groups: actives and dormants. All category users will become active shoppers at some point in the future but the great majority are dormant at a given point in time. This creates some very interesting math…

  • Nearly all short term sales will come from this “small” segment of actives…perhaps 5-20% of category buyers (depending on the category purchase cycle)…almost by definition. Even allowing for some impulse buying from dormants, over 80% of weekly or monthly sales comes from less than 20% of category users! Furthermore, the “actives” segment is where nearly all short-term sales response to advertising comes from!
  • Dormants. The complement group, dormant consumers are 80-95% of category users. They are unlikely to buy this month, but will certainly buy in the future, Because they are dormant now, changes in their brand opinions can only pay off down the road, when they become active. So this segment is where long term effects come from.

This leads to a simple but powerful new insight: understanding short vs. long term advertising effects starts with consumer segmentation based on journey stage. Once you realize this, it will change everything…how you think of the full effect of advertising, how you set media plans, how you create and deploy creative assets and how you measure campaign effects.

Start by mapping advertising goals to the appropriate segment: short term goals map to actives and long term goals map to dormants.

Media strategy for actives. Precisely target actives to drive up ROAS by creating a segment of actives at scale, regularly refreshed in your DMP. This enables promotional advertising via programmatic targeting. And really high ROAS should be the goal for actives…forget all else. I co-authored a study on three CPG campaigns that proves that effective targeting of those likely to buy in the near future (actives) doubles (at a minimum) ROAS. Many believe creative is 70-80% of the story for ad effectiveness.  I do NOT believe this is true when targeting actives; to drive short term ROAS, media science is at least as important as creative.

Now one more important point: if only 10% of category buyers are active but they are much more responsive to advertising, don’t you want to spend proportionately more against actives?  I’m not suggesting that 100% of media goes against this segment (because we want long term effects as well) but perhaps half of your media budget should be targeted to them. Working with client media plans, their agencies and DMPs, I have not seen this, but I council them that the conclusion is inescapable. Reduce media placement based on demos and redeploy towards programmatic, mobile, advanced segment targeting, and search.

Media strategy for dormants. The only reason to not put 100% of addressable ad spend against actives is if you believe that changing brand opinions among dormants will pay off down the road. However, I suspect that no one really knows because brand trackers do not measure this (although they could). And the half-life of this effect could vary substantially by category. For example, the opinion of dormants relating to brands of smart phones probably stick with the consumer when they activate.  The opinion of a TV manufacturer among dormants is probably less important.  They might not buy for 10 years and by then technology will change. For those situations, in a digital age there is an alternative brand building strategy; target actives and build brand predisposition “on the fly”, as actives search, visit comparison sites, etc. In either case, for brand building, creative matters more, teaching consumers new things and creating new feelings towards your brand.

A full effects model would take all of this into account and guide you to the right media allocation between segments and creative asset approaches.

Adjusting your research approaches. The last few trackers I designed for clients were based on a full effects model by segmenting actives and dormants and something amazing happened.  A new key metric emerged that reflected both brand equity and was also predictive of upcoming sales. As a result, management actually paid attention to their tracker again…because the short-term connection of the metric to sales was clear and meaningful.

­Dormants vs. actives segmentation…so simple and yet such a powerful new pathway towards understanding and properly estimating the full effects of your advertising.

Tags: , , , , , , ,


8 Responses to “Why you are getting the wrong answer on advertising effectiveness”

  1. I like it. Good basic thinking and a great practical advance that recognizes and accommodates the diversity of products that marketing is called upon to support. From CPG (frequently purchased) to residences, cars, mattresses, etc. (infrequently, but very widely, purchaseds), an explicit split of respondents along the “journey” will bring manifest benefits – and delights – to the studious research analyst. BRAVO, Joel.

  2. Stan Sthanunatan

    how does this fit with Ehrenberg Institutes focus on driving reach?

    • Joel Rubinson

      Stan, it actually does not harm reach for a CPG marketer because this is about timing ad impressions since the same category users cycle through being active and dormant. There are other recommendations I might make that are more in conflict with EBI and would be happy to debunk some of their thinking for you.

  3. Allen Bach

    Good stuff Joel, thanks for sharing

  4. Phil Bonanno

    Good, straight-forward approach. I use a 3 lane system:
    – Core, what you call active – you dont need to spend much to get a great return, esp from a mobile-first & progammatic PoV (indeed w/ addressable audience)
    – promiscuous, active but not loyal (these are a fish or cut bait group, as many of these customers are not profitable)
    – persuadable, what you call dormant – I recommend investing heavily here (I dont agree that this has to be long-term benefit here, many in this segment can be activated with relevant comms)

    there’s a fourth which is Intractable, audiences that will never buy (or even consider) your offering (really important for film releases and government elections as an example)

  5. Jo-Ann Osipow

    Like many winning ideas, the active/dormant view seems self-evident, should have been done a long time ago! I’m guessing it will quickly become a key tool for effective marketing across the customer journey.