| Last
week, Tom Hespos took on the problem of smaller websites and the
reach and frequency debate. It turned out to be one of the best read
columns of the week and one of the most popular discussion topics on
the Spin Board. He wrote, "Some weeks ago, I weighed in with my
opinion on how we should gather the data necessary to bring reach
and frequency to the web. My opinion was that actual visitation data
should be used, as media properties have ready access to it, and
because panel data can be somewhat sketchy. In looking at the issue
a bit closer, I'm inclined to solidify my position that there's no
substitute for actual visitation data."
In response, one member wrote, "Tom, you're absolutely correct. The Internet is the one true medium that will allow us to track these sort of numbers as well as individually identify who is visiting frequently (with a margin of error that is certainly able to figured out). Why on earth would we want a panel system? It wouldn't work for two reasons: 1. There are a helluva lot more sites out there than there are even channels on satellite; and 2. The Internet has a global reach in which anyone at anytime from anywhere can bounce thru a site." The Spin Board exploded into a heated debate. What do you think? Join in! |
Monday, November 18, 2002
Web Reach and Frequency: Game Over or More Study?
By David L. Smith It seems like, everywhere I go, I am asked about Web Reach & Frequency. Now, I have certainly set myself up for this, having written about it for several years, by my involvement with the ARF Online R&F Standards Committee and participation in industry meetings and through various speaking engagements. Not that I mind the questions. This issue is near and dear to me and I can give you an hour on it if we could find the time. But in this case, I'll try to limit it to a page or two.The biggest issue on most people's minds? Which vendor(s) should I be using? Some have asked me if we just couldn't declare a winner and move on. Well, with all due respect to the fine work that has been done this year, there is still a lot to do before the current and future solutions are ready for prime time. That said, after over five years of apparently no progress, the development this year has been nothing short of terrific. This is a great example of competition at work towards a common industry goal, with standards emerging and other questions yet to be either asked or answered.
Here are just some of the issues:
In the first place, picking a vendor is not easy. There are three different types of companies offering solutions of some sort. These include the all media research platforms used by agencies, clients and publishers for planning purposes like Telmar and IMS; the third party ad servers like DoubleClick, Atlas DMT and Bluestreak. (The ARF calls these companies SCM or server centric measurement); and the survey companies like Nielsen NetRatings and comScore Media Metrix (The ARF calls these companies UCM or user centric measurement). If you looked at the vendors three months ago, you would have seen some alliances that seemed like the path and choices were clear. On one side was WebRF on IMS with NetRatings survey data and a promised future product overlaying DoubleClick paired duplication data. On the other side was Atlas DMT, which was incorporating comScore survey data and talking with Telmar. Nice little package, where everyone had a seat, unless you were Bluestreak. All that does not matter, as there is a serious version of three-dimensional musical chairs going on right now. One thing is clear; it is far from "game over".
Another issue is the ARF position that UCM and SCM data should be combined for the penultimate system. Not everyone agrees with this. The reality is that this theory has not been either proved or disproved. Is seems logical. After all, the UCMs have been misusing the term "reach" for over five years. Reach is simply not a media vehicle term. It is an ad campaign term. As agencies, we are interested in reporting out the reach of our campaigns to the client. The coverage of a given medium or vehicle is certainly of interest to all, but is not the key in determining brand success. The UCMs can provide us with some important data points, including demographics, daypart usage, cume potential curves, etc. And what they cannot provide, which is actual paired duplication data from specific campaigns for reporting and future modeling purposes can be provided by the SCMs. As such, the combination of these two sources seems logical. The how of this is a whole other article.
What needs to be done at this point? Continue to push the vendors for more data releases, test the options, and be vocal about your needs. This is a unique time in any marketplace. There are so many threads going on at once and so much flux. All stakeholders seem to be looking for information and ready to give their opinions about what they need.
I'll give a few examples of what can be expected in the next few months. This is not in any specific order as ship dates for software and research are always more than a little soft.
- Atlas DMT is expected to release a desktop version of their software and data in the next few weeks. As they have used comScore data before, this is still expected to be the case. The inclusion of a significant comScore update is looked for soon. And, don't be surprised if they make some deal with NetRatings to give you a choice of UCMs for their software.
- DoubleClick is working on a release of NetRatings data through IMS on their MediaVisor system. Release tbd.
- Telmar's WebPlan is ready to use and seems to offer some fixes to releases from other vendors but it needs data. You can provide your own, or look forward to the pending release of new comScore Media Metrix data that is expected be the largest respondent database available for use in R/F programs. And more respondents means that more sites and greater granularity of data can be expected.
- New comScore Media Metrix is about to be included in both Atlas and Telmar releases (see above). CMM has a rudimentary desktop system available now and seems willing to work with a variety of vendors in what could be characterized as an open systems approach to data.
- WebRF+ (IMS with NetRatings survey data) is available now if you have your own third party server data cube. It is expected to incorporate DoubleClick data at some time in the future that has not been defined publicly.
- BlueStreak is reported to be releasing their first system this quarter.
- And don't count NetRatings out. While their product has been out for some time, they are not expected to rest on their laurels. Look for more announcements from them, including the possibility of including some @Plan data.
What more could we ask for? There ARE a few things I would like left under the tree at Christmas.
- I would like to see a single vendor who can provide us with what Jim Meskauskas would call a "sperm to worm solution". That would include a system that we could use strategically in combination with other media to help us determine whether Interactive should be part of the mix and if so, how much; a planning and pre-buy system (most of those above are in this category); a system for tracking actual R/F delivery as the buy is running, making it possible to tweak if goals are not being met; and a post analysis tool, which like the pre-buy system would be on a US population basis. After all, the proof of the pudding in the planning and pre buy systems will be whether they hold up in a post-buy analysis.
- The industry would be well served by the third party ad servers acting like the UCMs. Release of a data cube that is the equivalent of Nielsen Media Research TV Cume study by all third party server companies would make the choices really robust.
- It would be nice to be able to buy somebody's product without having to pay so many different vendors and scaled so the little guys can afford it too. I know that everyone has to get their cut on this, but it would be nice for the end solution to be packaged in such a way that I knew what the bottom line price was before I committed.
- And lastly, it would be nice to get the ARF Pilot study done and the white paper out.
Wait a minute. The last two are on my plate! Time to get back to work.
David L. Smith is President and CEO of Mediasmith, Inc.
