Awareness, attention, recall, perception, persuasion, preference and engagement are only relevant if they are part of a process of persuasion that leads to action.

If they're not linked to action, clicks don't count. Correct brand identification is meaningless. "Noted" and "read-most" scores don't matter. Message comprehension is irrelevant. Eye-tracking studies and MRIs of brain activity are pointless. Because what we're trying to do here is get people to do something.

If it doesn't get folks to hand over cash, change their ways, pull the lever or whip out their plastic, marketing communications is an unconscionable waste of resources. That includes all marketing communications: advertising, public relations, web, mobile, events, package design, corporate philanthropy, signage and anything else that we do.

• But don't we do corporate philanthropy so people will like our companies? A whole lot of data show that people like companies to support causes or sponsor activities in the public interest, but that their appreciation has no effect on their purchase decisions.

• How about publicity? We want the public to know that our organization has been around for a century, don't we? Or that we're leaders in the field? Our innovations? Our commitment to the community? That's important, right? Actually, no. A great deal of research has found most folks don't care about an organization's past or leadership or general wonderfulness. They want to know what they'll get – their benefit – if they do business with the organization.

• But surely advertising awareness and likability count for something? If they did, Coca-Cola sales would have taken off when the cute polar bear commercials were aired. But in fact they continued sinking. And some of the most irritating campaigns ever perpetrated have been tremendously successful.

So all we have to do is measure the immediate sales impact of marketing communications? Not quite.

Direct-response marketing can do that, of course. Bob Valenti knew exactly how many Ginsu knives each commercial sold. Credit card companies can calibrate response rates to their mailers precisely. A search-click-buy pure-play online merchant knows which key words worked best and the bottom-line benefit of every transaction.

In the direct-response model, simple arithmetic is all it takes to  link consumer action to marketing.

Except for one little glitch.

Prospects don't live in hermetically sealed environments. The people who buy Ginsu knives from a commercial that airs at 2:00 A.M. Thursday may have been influenced by dozens of other spots flickering across their retinas on previous insomniac nights. Someone mailing the perfed-on application to Capital One may be responding to the mailer – and to the great "What's in your wallet?" commercials designed to soften them up for the direct mail pitch. And online merchants' customers may check ratings and reviews on social sites before buying.

The calculation gets more complicated for the transactional marketing of organizations that tie sales to promotional marketing communications: moonlight madness sales, this-week-only buy-one-get-ones, flavor of the month or other time-specific offer.

It's easy to count bodies through the door, track closing ratios and come up with a pretty accurate approximation of the sales the special promotion produced. Not so easy to determine what previous exposure to marketing communications got the product the retailer sells into the customer's decision set to begin with. Because a great special financing offer is only effective if the prospect wants the item in the first place. And this-week-only catfish sushi offer will only work on folks who have previously been persuaded to eat our finny friends without benefit of externally applied heat.  

The toughest communications-to-action relationship to calculate is in the relational mode, where brand value and the lifetime value of customers are the key metrics.

Today's Coca-Cola commercial impacts brand decisions made two years down the road, and those brand decisions usually determine brand choice for decades. Amazing though that sounds, a BrainPosse principal witnessed it in action--and learned the key factor behind the phenomenon--when he did Coca-Cola advertising. (Sorry, details are confidential. If we told you, they'd have to kill us.) Obviously the number and variety of marketing communications messages reaching the targets before the point at which they were polled, and the messages delivered afterward to bend their inclinations Coke's way, are very complex.

Buying a car is a multi-year process. If a brand doesn't make it into the consideration set at the top of the funnel there's little or no chance that it will be chosen at the business end, when the purchase decision is made.

The difficulty of tying consumer action directly to marketing communications was put into focus by Rosser Reeves. This strong advocate of accountable advertising once related: "Recently a group of marketing men, almost idly, at a luncheon table, listed thirty-seven different factors, any or all of which could cause the total sales of a brand to move down or up." He also said "We do not mean to imply that advertising is not an important factor. It is. We simply wish to make the point that big mistakes can be made if you try to judge an advertising campaign, always, by sales."

Today, marketing communications is much more than advertising. And there would be many more than 37 factors on the list. But the principle holds: there are many factors involved in the relational (or brand) sell. And while the only ultimate metric of marketing communications success or failure is action, arriving at that measurement is complex in the direct sales model, more complex in the transactional model and downright labyrinthean in the brand – or relational – model.

But that doesn't mean we shouldn't do it.

There are two good ways to tie actions to marketing communications:

The complex way. Surprisingly, we think this works best for direct sales and transactional models. The complex way is multivariable testing, in which numerous influencing factors are evaluated simultaneously to determine their relationship to consumer action. This approach has two great strengths:

• It's fast. Although a multivariate test can take months, it evaluates numerous factors. Researching the individual factors sequentially would take much longer. 

• It finds synergies between factors being tested. Conventional single-variable testing looks at each factor in isolation. So possible multiplier effects of two factors working together can't be seen in the outcomes.

There is a downside to multivariable testing. It's not something most companies can do for themselves. The methodology is highly specialized, the data collection requirements are rigorous, and the analysis is demanding.

Some web-only organizations are beginning to do in-house multivariable testing. They have enormous data gathering and analysis capabilities which give them a good start on the process. But most organizations should go to an outside firm, which is expensive.

The simple way. Find one key indicator that correlates to action.  

Fred Reichheld's Net Promoter Score revolutionized customer loyalty studies by drilling down into data to find the one question that is an eerily accurate indicator of an organization's success: "Would you recommend us to a friend or colleague?" The answers are on a scale of 1-10, with the percentage of respondents answering 1 through 6 subtracted from the percentage answering 9 or 10 to arrive at the net promoter score. The NPS is an excellent predictive metric, but it has no prescriptive component. To add that, we believe one more question is needed: "Why?" Those two questions provide all the information most companies need to determine and manage their customer loyalty.

The Coca-Cola research we mentioned earlier was uncannily accurate at predicting future actions – and at indicating ways to improve outcomes. But the survey focused on just one key factor in one demographic sub-segment.

So how do you find the simple answer? Unfortunately, that's complicated.

Fred Reichheld didn't just dream up the Net Promoter Score. He discovered it by sifting and analyzing mountains of data and finding a pattern that had not been noticed before. The elegantly simple Coke research didn't spring full-blown into the mind of Guillermo Taylor, the Coke marketer who discovered it. It was the result of seeing patterns in data over time. Years of time.

There are certainly algorithms which can detect correlations between factors and subsequent actions. They require an unbiased approach to the choice of data to be analyzed and, often, a whole lot of work formatting and/or loading the data.

Alternately, an astute marketer can pore over data until patterns emerge. This course takes a lot of time, but it has something to recommend it. A marketer who submerges herself or himself in the data that drive her or his brand acquires a deep, frequently intuitive, familiarity with a much broader range of factors which impact the brand's sales.

None of this is meant to say that awareness, attention, recall, perception, persuasion, preference and engagement don't count. They are all essential steps in the persuasion process.

But none of these factors is the ultimate objective of a rational marketing communications program. The only reason to measure them is if data analysis shows that one or more of them are key marketers of marketing communication's effectiveness at motivating action.

To find out more about marketing communications that generates action, contact BrainPosse by clicking here or calling (865) 330-0033.    

back to top I archive I home

Comment