 |
| |
June
2002 |
|
| Online
Reach/Frequency and GRPs - An Update On Progress |
By David L.
Smith
President and Media Director, Mediasmith,
Inc. |
|
| The column will be devoted to the
state of online reach & frequency development for
the Web. Regular updates are anticipated as developments
occur, so check back to see what is going on. It is
amazing to many that, in this medium’s 7th year, "the
most measurable of all media" does not have a basic
metric like reach & frequency. But then, we just got
agreement on the first impression definition last year.
Can we achieve the goal of having useable reach &
frequency models for the Web? Absolutely. It may be
sooner than you think. Discussion of the process that is
currently going on and the various players will be dealt
with later in this article. But first, some definitions,
background and history. |
|
 | |
|
|
|
|
|
Definitions In work recently performed
as a part of the ARF Online Reach & Frequency committee
(full disclosure: I chair this committee. I write this article
as an independent writer but will report out on the progress
of the committee where that progress is a matter of record),
Roger Baron of FCB provided the following definitions:
Textbook Definition
Of Reach In Traditional Media: The number of different
persons or homes exposed to a specific media vehicle or
schedule at least once. Usually measured over a specific
period of time, e.g., four weeks. Also known as cume,
cumulative, unduplicated, or net audience. (Glossary: Sissors
Advertising Media Planning, 5th edition).
The ARF Online R&F Committee Definition of Internet
Reach: The number of different persons exposed at least
once to an Internet advertising message over a specific period
of time, usually four weeks, but preferably starting with a
one-week period.
Frequency: The average number of times a (person has
been exposed to) an advertising message over a specific period
of time, usually four weeks, but preferably starting with a
one week period.
Reach is generally couched in terms of a percentage of the
universe. For traditional media, this universe is generally
total U.S. population for the target audience in question
(unless it is a local market effort, in which case, the local
market DMA or metro area target audience population). Thus, if
a campaign has a 60% reach vs. men 18-34, that means that 60%
of men 18-34 in the U.S. are exposed to the campaign at least
one time.
Return
to top of page
Background and History It should be
noted here that the Internet audience measurement companies
have used the term reach to characterize the cume potential of
a medium, not the number of people reached by a campaign. This
is unacceptable in the media world and their numbers should
not mislead you. This is akin to saying that CNN has 75% reach
because 75% of the U.S. can get CNN in their home. A campaign
on CNN might only reach 10-15% of your target. Similarly, you
might see info that says that Yahoo! has 45% reach. This is
their cume potential. When we get the actual R/F data, a major
campaign will probably be in the single or low double
percentage points.
The movement to obtain reach/frequency tools that can be
used is gaining momentum. The IAB, ARF and other bodies (including
Rick Parkhill’s iMedia Connection)
are all talking about moving the ball ahead on this
topic. The goal is to be able to have a desktop tool
wherein: a) a planner or buyer can estimate the differing
R/F impact of various campaigns, e.g., what happens
if you buy a bunch of small sites vs. one big site,
and b) the ability to combine the Internet R/F data
with other media to determine the increment added to
a traditional media campaign with an Internet effort.
The big issue: The base data for reach and frequency
from other media comes from the survey-based suppliers. For
print, sources like MRI and Simmons make their data
available to third-party processing companies like Telmar and IMS/MediaPlan. Advertising
agencies, clients and the media use these systems for print
and broadcast reach and frequency and multi-media reach and
frequency combinations. MRI also has their proprietary
cross-tabbing and estimating system, MEMRI. Data for broadcast
comes from a Nielsen audience cume study, performed every five
years or so. This is also made available for R/F analysis
through Telmar and IMS/MediaPlan as well as their own
proprietary system. |
|
|
|
The important thing to
note here is that, to a large degree, vehicle audience
is assumed to be the same as commercial audience. While
we know that this is not technically the case, there is
an OTS (opportunity to see) for the whole vehicle
audience and at least empirical evidence that most in
the vehicle audience have been exposed to the message.
|
| |
|
The Web however is different.
When a banner or other type of ad is served, it is rarely
exposed to the full audience of the site or even the full
audience of the page served. As a result, there is no
relationship between the site’s audience and the audience for
the advertising. It is for this reason that modeling of site
survey data does not produce logical reach and frequency
information. Leslie Wood of LWR has spent a great deal of time
analyzing both JMM and Nielsen//NetRatings
data. According to Leslie, the survey-based data (which the
ARF refers to as User Centric Measurement or UCM) shows
potential that overstates the reach that is actually achieved
by the advertiser. There are several possible reasons for
this:
- Users may by habit only be going to certain parts of a
site that they are used to, and even a large sample of users
might not view all pages on a large site, thus the potential
cannot be achieved,
- It is common for the major sites to sell out much of the
inventory on many of their top pages to long term exclusive
"business development" or "strategic alliance" deals. The
result of this is that the ROS (run of site) advertiser
cannot get a true dispersion across the site. This is akin
to buying RON (run of network) on NBC and not getting the
big reach prime-time spot. As such, no workable solution has
been put forward based on survey data.
Advertising on the Web is "trafficked" through Web servers.
Many agencies use a third party server such as DoubleClick’s
DART or Atlas DMT’s offering to post their ads. (The ARF
refers to this data as Server Centric Measurement or SCM.) As
such, the third party server has information on which pages an
ad actually appeared. They just don’t have the audience data.
So an alternative solution was put forth to marry server data
(SCM), which has advertiser’s campaign performance
information, with the survey data (UCM) that has audience
information. This alternative does not sit comfortably with
researchers who prefer their data from a single source but it
is one alternative that is being explored.
The use of GRPs in planning is a related topic to the reach
and frequency issue. While at Tribal DDB, Tim McHale presented
a compelling argument for charting GRP or TRP numbers for the
Internet on the traditional media flow chart. This GRP/TRP
concept is sound and should be seriously considered by all.
Return
to top of page
ARF Meeting The recent ARF Online R/F
Committee meeting referred to above saw presentations from a
number of vendors who will be offering some aspect of the
reach and frequency solution.
The basic postulates put forward at the ARF meeting had to
do with whether SCM, UCM or a combination of the two made the
most sense. Presentations were made by Leslie Wood (background
on the subject), JMM (beta product), Telmar (needs fuller data
set from other vendors), Atlas DMT (beta product), comScore,
Diameter/Doubleclick (post analysis product but no planning
product at this time), and IMS/MediaPlan (needs fuller data
set from other vendors). It should be noted that
Nielsen//NetRatings is also working on a product but did not
present at the ARF. Stay tuned for news from them.
After the presentations, a vote was taken relative to UCM,
SCM or a combination of the two. The agreement to use the
combination (UCM and SCM) as a solution was nearly unanimous.
This gave the industry direction towards which to work. It was
also discussed that further examination of data sets needs to
take place in order to determine the duplication between sites
and campaigns. Atlas and DoubleClick provided very different
perspectives on this aspect. It is possible that their
computations are being performed differently. An ARF
subcommittee will report back on this and other analyses.
In addition, the ARF is working on a white paper, which
will be issued on this topic. |
|
|
|
In the end, it will be the
research and third party server companies who come up
with solutions, not the ARF. These solutions should help
with both pre-buy
and post-buy. |
| |
|
| For pre-buy, we need GRPs (at
least as broad as Nielsen TV demos, but hopefully deeper, more
like MRI and SMRB data.) Then, we need pre-buy reach and
frequency data. For all demographics. In order to have a large
enough sample to take into account the many different ways
that the Web is purchased, a cume study similar to the one
that Nielsen does as the basis for TV R/F is probably in
order. DoubleClick is looking at the viability of doing this
study and they are the most likely candidate. The results of
such a study could theoretically be used with all survey
sources (JMM, Nielsen, comScore).
We also need the ability to put this together with
traditional media R/Fs to look at combined reach and
frequencies on 1+ and 3+ bases, frequency distribution, etc.
We need this for the US at first and then for other countries.
Jupiter Media Metrix, Nielsen//NetRatings and (hopefully)
comScore’s NetScore should be able to do this. Of course, they
will need to define universes for each demo they wish to
report on (on a country by country basis, starting with the
US). Once we have this, we will need them to break out user
data by country. This will give us composition by demo. We
apply this to the impression schedules that are proposed to us
and then we’ll have GRP measurements, whether they report them
or not (impressions/universe x 100=# GRPs) with real GRP
estimates of schedules. It is a short step from this to reach
and frequency, either from the individual companies or from
third party interfaces like Telmar and IMS. We can then use
the Methringham formula or other mathematical models to
combine online media weight with traditional media reach and
frequency.
The benefits to this will be enormous and immediate.
- Buyers will see what the relationship is of the schedule
they are buying to the site as a whole. I have a sneaking
feeling that we are underbuying a lot of sites. Except with
business development deals, where sites are probably
overbought. Would be nice to know, wouldn’t it?
- Planners and media management will quickly see that they
might be allocating 4000 GRPs to television and 40 to
Interactive annually, and adjust their budgets to something
more realistic. Importantly, though, it will also help
reveal an opportunity for advertisers to reinforce their
message against the light TV viewers.
- Sites will have more information on how to sell a
schedule that has the best chance of optimizing their
audience through reach/frequency analyses. This could result
in bigger buys on their site.
- Advertisers will understand what the Web adds to their
schedule. For the big advertisers, it will not add much, if
any, reach. But, through demographic and geographic
analysis, it may help to balance out impressions against key
groups—those light viewers mentioned above.
Return
to top of page
Next Steps The dialogue within the
industry is heating up. In the last several weeks, some new
alliances have been struck. The ARF committee is moving ahead
on setting parameters and industry guidance. The various
vendors are working on their products. We will keep this page
on the site up to date on progress. So, check back for news.
We expect a report on the new alliances some time in June.
Return
to top of page
|
|
|
David L. Smith, President
and Media Director of Mediasmith, Inc. in San Francisco
is a nationally known expert in the areas of new media
application, media strategy and media planning. A
thirty-seven year veteran in the advertising media
management arena, Smith has a major involvement in
national committee work to establish and refine
standards in metrics, business practices and financial
issues for Interactive advertising with organizations
such as the AAAA's, IAB, OPA and the ARF. He currently
chairs the Online Reach & Frequency Committee for
the ARF. |
| |
|
| |
| |
