The data comes from TiVo’s StopWatch, available since February of this year, which measures second-by-second viewership from a random sample of 20,000 TiVo units. Two months of data were released (April and May) and the results, while preliminary, are telling.

TiVo

It seems that commercials at what Steinberg calls “opposite ends of the spectrum” did well against the fast forward button. One is “bare-bones direct response” and the other is “high-production-value movie ads.” The data on movie ads are tenuous, in my opinion, given the small amount of those ads on the list of least fast-forwarded. On the direct response side, however, the data are more convincing.

In the month of May, for example, three of the top five least fast-forwarded spots were DRTV ads. I’ll spare you the somewhat-less-than-kind words Steinberg used to describe direct response ads.

Todd Juenger, from TiVo Audience Research, says that the information is still too new to draw strong, sustainable conclusions. In fact, you can’t even break down the data by demographics. Still directionally it’s suggestive. Mr. Juenger suggests that direct response ads do well “because they are easily understood.” He also points to the relevance of the message as an important consideration, although I struggle to link lack of message relevance to fast-forwarding when the sample is random.

Based on this, some agency media buyers expect agencies to offer more bare bones information in advertising, suggesting an approach closer to that of text-based online paid search ads.

It will be interesting to see what the future releases of the TiVo data show, but for now at least, my interest is piqued.

While essentially a research-related article, I found it valuable for marketers in general to consider what Neveril writes in the context of marketing communications. He writes that it is common in his field to find dissonance between what people tell you about themselves (and what they do) and what they actually do.

He proposes that in order to cut through all the “clutter” of a description of what a person thinks about the brand or their desire and intent to purchase it, for example, we can discover the real person by focusing on actual behavior. “Actions do speak louder than words,” he writes.

The article goes on to recommend three practical approaches to gathering behavior data, defined by the elapsed time between the behavior and its recording. The best approach, he concludes, it recording the behavior as it happens, which is usually associated with ethnographic research. The reason it’s the best approach is because the researcher doesn’t need to rely on the consumer’s memory, but rather the data are recorded in real time.

In the end, Neveril reminds us that past behavior remains the best (albeit imperfect) predictor of future behavior. He writes “if you want to know how your customers will behave in the future, don’t ask for more of their opinions; just take a look at their behavior.”

I submit that direct marketers have known and leveraged this for generations. In direct marketing, it doesn’t much matter if a consumer claims to have that elusive “purchase intent” or if they are in the “consideration stage” of the purchase process. What does matter is if the consumer takes a measurable action.

Only when the consumer exhibits a behavior can a direct marketer safely measure and predict future actions. Think of it this way. Instead of asking a bunch of consumers if they’re interested in Brand X, let’s measure how many respond to an offer for more information about Brand X. In the first case, we’d know what consumers said about their interest in the brand. In the second case, we’d know consumers actively got off their butts and acted by saying “hey, I’m interested, please send me more information.”

I would argue that more valuable is the second instance where there is actual hard (quantifiable) data about hand-raisers. That’s one of the beauties of direct marketing. There’s little need to do research in the traditional sense of the word. We just test in market with real environmental variables in place. What works we keep, what doesn’t we don’t.

Paraphrasing Neveril, to know what’s going to happen, just do what direct marketers do and look at what’s happened before to determine what will happen in the future. When that’s not available, don’t ask consumers what they think they’d do. Make them an offer and see what they actually do.