The announcement that P&G was apparently turning its back on targeted Facebook ads created a storm in the marketing world, leading to some utterly crazy comments suggesting that targeting and segmentation was in some way suddenly wrong. By coincidence, I’d simultaneously published a post on the merits and benefits of shopper segmentation, and a number of readers asked of my thoughts on the P&G situation. Was this the end of targeting and segmentation as we know it? Has social media upended one of the founding principles of marketing?
Reading some of the comments in the press, you’d think so. The nonsense and hyperbole that followed this news was nothing short of astounding. One commentator said this (and I quote): “Targeting may have been an experiment, but at the end of the day a brand like Tide needs as many people to see an ad as possible”. It is, I think, the single dumbest thing I have ever heard (in marketing). Really? As many people as possible to see ‘an ad’. So current users, competitor users? Loyalists? Price Switchers? Absolute rejectors? People who live in markets where Tide isn’t available? Really? Come on. The value of the impression will vary massively depending on who the consumer is and the action that they will take as a result of the impression. The idea that ‘as many people as possible’ only applies in a world where every impression is equal, or a world where return on investment is unimportant and marketing budgets are limitless. No. Segmentation and targeting are still massively valuable. So what happened?
P&G got their targeting and segmentation wrong
P&G just got it wrong. Even P&G admit this. Marc Pritchard, Chief Financial Officer for Procter & Gamble “We targeted too much and we went too narrow.” As discussed in “The Shopper Marketing Revolution”, for a segmentation model to be valid, it must meet five criteria:
Measurable: The segment must be identifiable using criteria or characteristics that can be measured (at a justifiable cost).
Sizeable: The segment must be large enough to represent a significant sales or profit opportunity. Over-segmenting can create very clear, differentiated groupings, but if the value they represent is unlikely to deliver a return, then combining segments should be considered.
Homogeneous: All members of the segment should be the same or similar across a clear set of criteria.
Differentiated: The segment must be clearly different from the rest of the population across one or more clear, measurable criteria or characteristics.
Actionable: It must be possible to approach each segment in a cost-effective manner with a particular marketing mix and elicit a positive or desirable response from it—one that is different from the response you’d get from the general population.
It seems (by P&G’s own admission) that they didn’t get this right. They went too narrow, meaning perhaps that the segments were no longer sizeable. And the fact that Facebook charges more and more as the targeting gets tighter, it also challenged the “marketable” point, as the cost of marketing per impression increases as the target gets tighter.
For instance, P&G two years ago tried targeting ads for its Febreze air freshener at pet owners. There was little impact on sales for that effort, but when they expanded the target to include all adults, they seemed to get a response. But that doesn’t mean that all targeting is wrong: just that perhaps pet owners aren’t motivated to buy Febreze by the Facebook campaign that P&G ran. Nothing more, nothing less.
Just because you can, doesn’t mean you should
Facebook seduces marketers with the promise of better targeting – but they are selling and focusing on the three elements they can deliver best: homogeneity, differentiation and measurability. But what it appears is that P&G forgot to keep an eye on the actionability and sizeability measures. Marketers need to review the channel options they have and sense-check the segmentation makes sense.
None of these mean that targeting is a bad thing to do. None of this suggests that we should forget about segmenting consumers or shoppers. It merely means that there is a limit to the value of tighter and tighter segments. Even P&G acknowledge this. A spokesperson says “Our brands will continue to use Facebook to reach consumers, including targeting, where it makes sense”
Targeting and segmentation – measure and evaluate
Where P&G deserve credit is for trying something, for measuring and evaluating it, and for trialing and measuring different approaches.
Focus on your target, not the latest toy
So is this just a storm in a teacup? No. It’s a valuable lesson for every marketer. If P&G can get this wrong, then let’s admit it’s possible we can all make the same mistake. Targeting and segmentation are key marketing skills: as Michael Porter puts it “the key to competitive advantage”. But in the world of big data, we have the ability to take our segmentation much further than ever before. The lesson here is that may not always be a good idea.
Marketers must create segmentations which deliver against the golden rules of segmentation, rather than getting carried away with the apparent sophistication of the latest tool. Social media marketing can be brilliant. But when it comes to tighter and tighter segmentations, just because we can, doesn’t mean we should.
If you’d like to understand more about how bespoke targeting and segmentation of consumers and shoppers can drive growth of your brands, please get in touch and I’ll gladly share some case studies.