|
According
to a 2006 ANA survey of its members, 86% of respondents
were "very interested" in having commercial
ratingssimilar to Nielsen's TV program ratings,
a measure of how many viewers in America watched a TV
commercial when it airedavailable, while 13% said
they were "somewhat interested." When asked
if members would like to see commercial ratings become
the currency by which TV time is bought and sold, 44%
were very interested and 40% were somewhat interested.
Thankfully for these advertisers who aren't so willing
to go on spending $70 billion annually on TV advertising
unless they know, at the very least, people are watching
the ad, a new day is dawning....albeit pretty slowly.
Nielsen
recently rolled out a weekly report of average commercial
ratings, which show how many households participating
in the Nielsen panel on average stay tuned in minute-to-minute
during an advertising break during a particular program
whether it was viewed live or via a DVR up to a week
after the original broadcast. Advertisers, their agencies,
and broadcast networks had been tapping their feetor
in the case of broadcast networks, pacing the floorsholding
back on TV buys for the coming 2007-2008 TV season,
sniping over how to count TiVo and other post-live show
views, engagement metrics, and more, all in anticipation
of the big Nielsen release day.
Unfortunately,
for many, if not most, advertisers, Nielsen's much-anticipated
average commercial ratings turned out to be more like
getting underwear and socks for Christmaskind
of disappointing. "This is a step in the right
direction, but we're not quite there," lamented
Steve Kalb, the director of broadcast media at MediaHub,
the media buying arm of ad agency Mullen. What did marketers
really want to find under the tree? "Measurements
of how audiences view TV programs and ads on a second-by-second
basis," according to Ad Age. "While
the step towards providing an average rating for all
commercials in a television program is helpful, advertisers
want even more granular data," explains a white
paper from the ANA. "They want brand-specific commercial
ratings that can answer the question, 'How many people
actually had the opportunity to see my spot?'"
Of
course, Nielsen isn't the only game in town. There's
also that whole engagement thingthat ubiquitous
and as yet undefined hot advertising topicthat,
at least when it comes to TV, has become a sort of proxy
for how many people actually pay attention to commercials.
As Steve Tipps, senior vice president at Copernicus,
explains, "If programs are unequal in providing
viewers who stay tuned to embedded ads, then a measure
capturing that performance difference is essential to
accurately assessing the value of placing advertising
in a given program." In TVland, "engagement"
and "involvement" are used interchangeably
and as yet, there's no universally accepted tool for
measuring its effects on TV advertising response.
To
be sure, there are a growing number of tools on the
market. For example, IAG Research offers program AND
commercial ratings based on "engagement" scores.
The company has folks come into its headquarters every
night to watch everythingprograms and commercialson
the broadcast networks and shows on 20 different cable
networks. These folks then pull together a detailed
"quiz" that tests how much someone paid attention
that's given to an online panel of 5,500 people the
next day. Media industry pundit Jack Myers also offers
"Myers Emotional Connections" research which
assesses the performance of broadcast and cable networks
and programming in terms of "viewer engagement"
and "advertising impact." According to Marketing
Evaluations, creators of the Q ratings for TV programs,
some of Nielsen's highest rated shows are much less
engaging than their ratings would indicate. The company
found programs such as Desperate Housewives, a top-ranked
show by Nielsen standards, may not be all that engaging.
Though these tools and others have received a fair share
of positive attentionIAG in particularthey've
also encountered a good share of criticism, primarily
that they don't offer firm evidence that the way they
calculate engagement/involvement actually measures it
or is related to advertising response.
As
our readers might imagine, we've been hard at work looking
for the best performing measure and have made some interesting
discoveries in the process. Drawing from past academic
studies, our own research in this area, and our understanding
of current media industry practices for measuring involvement/engagement,
we looked at many potential indicators of involvement.
The indicators fell into three broad categories:
- Behavioral
(e.g., number of people in the room; type and size
of TV/monitor; level and number of distractions during
viewing; I watch the program every week; I always
plan to watch the program; I record the program if
I am going to miss it)
- Attitudinal
(e.g., the extent to which a program provides entertainment
value and is enjoyable; provides new relevant information;
includes recognizable characters with whom viewers
identify; reflect viewers' personal beliefs and values)
- Summary
(e.g., poor to excellent rating; comparison to previous
episodes; likelihood to view again or recommend; overall,
I paid very close attention to the commercials during
this program)
We
found the best measure of involvement included a mix
of all three types of indicators and confirmed that
viewers highly involved in a program are far more likely
to recall the advertising.
We
also discovered that, in addition to influencing the
attention paid to an ad, involvement had a halo effect
on the attitudes and behavior towards a brand. If a
viewer was aware of an ad, the more involved he or she
was with a program, the stronger the perception of the
advertised brand as superior, a good value, and overall
positive impression. Importantly, highly involved, ad
aware viewers were also more likely to purchase the
brand the next time they were in the market for a product
or service in the category. Our analysis took into consideration
the possible effects of demographics and product usage
on perceptions, but found only microscopic levels of
influence from these potentially mitigating factors.
In other words, program involvement is driving the bus
here.
TV
shows, as one might expect, differ in their mix of low
and high involvement buyers and that difference is indicative
of likely advertising performance. More popular shows
(a.k.a., shows ranked highly on the Nielsen program
rating scale) are not necessarily more involving, nor
are lower ranked shows indicative of a high concentration
of disinterested viewers. We confirmed that involvement
is not a function of age, gender, income, or some other
demographic factorthe concentration of high/low
involved viewers for a particular program was not influenced
by demographics. We also discovered that, in contrast
to the prevailing conventional wisdom among media buyers,
involvement is not a function of day part when it comes
to "niche" programsi.e., programs with
less than a 1% show rating, barely a blip on the Nielsen
scale. As it turns out, viewers watch programs with
just as much interest and vigor any time of day, not
just during prime time. What this means for advertisers
and their agencies is if they rely on Nielsen ratings
and conventional wisdom alone, they could be passing
up some very good TV buys.
Just
how critical is involvement to advertising performance?
According to our findings, the vehicle influences effectiveness
at least as much as the commercial itself. Program involvement
accounts for half of the attention paid to an ad and,
as we mentioned before, casts a long, positive shadow
on persuasion among those aware of the ad. The other
half is explained by a combination of the ad creative
and the extent to which the characteristics of an intended
user of the advertised product/service correspond to
the characteristics of viewers of a particular program.
As
for Nielsen's program rating scale....when compared
to the predictive ability of different measures of involvementours
and othersNielsen is woefully lacking. While it
certainly reflects the total numbers of viewers of a
particular program, it is an inaccurate predictor of
the total number of viewers who can recall an ad. In
fact, our analysis revealed Nielsen ratings account
for only about 60% of the difference in ad aware viewers
across different programs. Evaluating ad rates according
to a Nielsen-based cost per thousand (CPM) basis will
not yield the most lucrative media buys.
Now,
Google has partnered with Echostar, the #2 US direct
broadcast satellite TV provider and owner of DISH Networks,
offering the in-demand second-by-second commercial ratings
for buys made on 120 cable TV networks through Google's
ad auction system, an importantand inherently
limitingcaveat. But clearly a brand-specific (as
opposed to a program average) second-by-second measure
is getting closer to reality. Until then, a program-involvement
adjusted version of Nielsen ratings and CPMIs, cost
per thousand involved, instead of CPMs is a far more
economically effective and efficient means with which
to purchase TV airtime. Second-by-second ratings are
still an imperfect system; like the existing Nielsen
programs ratings, advertisers will know how many viewers
had the TV set on when a specific commercial aired,
but they'll still be left wondering how many actually
paid attention. The long and the short of this is that
involvement measures will remain as equally important
and relevant predictor of ad recall and persuasion as
it is today.
Back
to top.
|