Brass Tacks Design

The much-awaited EyeTrack07 report is here. For an industry much in need of help, it's much ado about nothing. Let's begin with Poynter's number one finding:

"People in our study selected what they wanted to read, of course, and then…they read a lot! We were amazed by these numbers."

The most amazing thing is the banality of this finding. Did we really need a study to tell us that heavy readers read most of the stories they select?

Moreover, what is the practical application of this information? How can an editor boost the readership of his or her newspaper knowing that heavy readers tend to read most of the stories they choose to read? How did Poynter hope to help us sell more more newspapers by studying the habits of people who already have the habit?

In contrast, consider API's Newspaper Next, which delivered practical information newspapers can act upon today to strengthen their franchises.

To be fair, Poynter never promised jaw-dropping findings. Nor did it promise to deliver practical information that would help editors sell more papers. But this raises an even larger question for Poynter: If this research was not designed to help editors sell more newspapers or boost the bottom line, what was the point?

Let's see if we can find it.

The key word in the sentence that leads Poynter's annoucement is "selected." I don't believe EyeTrack07 provides any data that speaks to how or why test subjects made selections – which is what editors really want to know.

Until we know why people choose to read what they read, we really don't have effective tools for boosting readership.

EyeTrack07 goes on to note that alternative story forms drew 15 percent more attention than narratives. But these forms are merely a secondary driver of readership and comprehension, and miss the more important – and primary – driver, which is story selection. Based on its online descriptions, Poynter's upcoming EyeTrack workshops in April and August do not address story selection, even though it is tied to their number one finding.

There's no denying that it's still all about the content. Same as it ever was.

A week after releasing these findings – and with some prodding from noted newspaper analysts Alan Mutter and Mark Potts – Poynter is admitting that content is king. But you won't find any mention of content in the Eyetrack07 materials released so far. The king is conspicuous by his absence – perhaps becaue he is wearing no clothes.

In contrast, research conducted by Brass Tacks Design indicates that content selection is the number one driver of readership, and that relevant content about pocketbook issues and health/personal safety trumps all other kinds of stories, regardless of form.

Eyetrack07 does not include any consideration or evaluation of these content-based issues. It's limited to what people look at rather than why they read.

Now we must consider methodology: No evaluation of EyeTrack07 would be complete without consideration of the Hawthorne effect. From Wikipedia:

Hawthorne effect refers to the phenomenon that when people are observed in a study, their behavior or performance temporarily changes.

In other words, the very act of measuring something changes the thing being measured. What's this got to do with EyeTrack07? Get a load of this gizmo:



Any scientist worth his salt will tell you that EyeTrack07's findings cannot say anything conclusive about newspaper readership, or the use of online sites, on an empirical level. This is not to say that the Eyetrack data isn't accurate. But all we can only say for certain is that EyeTrack07 tells us how people read newspapers and use online sites when they are being observed while wearing a cyborg-like contraption on their heads.

It's been 16 years since the first EyeTrack study. Hasn't this technology progressed to the point where people can be measured without being adorned with such an unnatural device?

More from EyeTrack07:

"A big surprise was that a much larger percentage of story text was read online than in print. On average, online readers read 77 percent of what they chose to read. Broadsheet readers read an average of 62 percent."

Now let's do the math. Take the typical 20-inch story. Eyetrack07 tells us that online users will read about 15 inches (77 percent) of that story, while users of print will read about 12 inches (62 percent) of the same story. Are we supposed to get excited about three inches?

A spread of 15 percent is not a lot – certainly not enough to boost online revenue to the point where it takes over from print. Online revenue remains in single digits as a portion of total newspaper revenue. This "big surprise" isn't gonna change that.

Steven Rattner, writing in The Wall Street Journal, and Warren Buffett seem to agree:

"Because readers will never linger over news online the way they did on the printed page, the advertising profits of yesteryear are gone for good."

Despite Eyetrack's claim to the contrary, this research does nothing to "dispel the myth of short-attention spans" because the deep reading "phenomenon" only occurs after a story is chosen for reading. Furthermore, readers were told they had 90 minutes to read the paper – who spends 90 minutes reading a daily paper? This unreasonable length of time could have created the deep reading phenomenon, rather than merely recorded it.

Most importantly, this research seems to ignore what happens before a story is selected for reading, when common sense dictates that users must be scanning and browsing on a very shallow level in order to locate stories of interest.

But wait a minute…Wasn't heavy use of headlines and blurbs online the major finding of the last Eyetrack study? Now Poynter is saying that the web is all about deep reading. Is this study contradicting that one? If so, then which one is right?

Actually, it's a moot point because both studies are based on the current online models for content, design and advertising that have failed to generate the kind of revenue newspaper companies need to survive.

Data is meaningless if it isn't useful.