Breaking News: The Online Industry Has Comparable Measurement Systems
|
Written By Reprise Media | July 15, 2005 | Share This
|
|
Sometimes I see something that I just don’t get, like this announcement that Andy Sernovitz has co-founded WOMMA (Word of Mouth Marketing Association), an organization dedicated to measuring and codifying word of mouth marketing.
The very notion of measuring word of mouth in a well-structured manner outside of brand recall studies and focus groups (both of […]
Sometimes I see something that I just don’t get, like this announcement that Andy Sernovitz has co-founded WOMMA (Word of Mouth Marketing Association), an organization dedicated to measuring and codifying word of mouth marketing.
The very notion of measuring word of mouth in a well-structured manner outside of brand recall studies and focus groups (both of which are fundamentally flawed) just seems.. well… fundamentally flawed.
But then I saw a statement from P&G’s Manager of Interactive Marketing and Innovation that the interactive industry needs to develop its own measurement system comparable to Nielsen ratings. And it got my attention.
What’s Holding Back Interactive?
McConnell told his audience Monday at Ad:Tech in Chicago that to truly compete with television, the interactive industry should develop its own measurement system, comparable to the Nielsen ratings.
McConnell also attributed the minute proportion of ad budgets spent on Web advertising (3.7 percent last year, according to the IAB) to this very fact. Now if Ted were an executive at one of the big traditional ad agencies I’d really be baffled by the statement. But still, this is a thought leader at one of the biggest brand marketers in the world, and his very job (title) relies on his ability to grasp the similarities and differences between new media and traditional.
Speaking on behalf not of search engine marketing, but of the interactive field overall, I’m going to have to take exception to Ted’s statements and provide some clarification:
The interactive industry does have systems of syndicated research and audience measurement that are comparable to traditional, though most are far more involved, illuminating and actionable from the Nielsen Ratings.
Because let’s face it, Nielsens aren’t really meant for the marketer, they’re meant for the media vendors and their ability to shape CPM rate cards based on audience viewership in addition to deciding which television content becomes syndicated. More on this later.
Present-Day Interactive Measurement Tools
I’ve worked with a handful of interactive measurement systems over the years, and below is a basic breakdown of some of the tools that have provided insight and empowerment to my media planners.
1. Hitwise - Hitwise is a data resource service that monitors consumer behaviors across 7.5M people within the US (via ISPs) and another 2.5M users domestic via panel-based companies. Worldwide they span 25M people across 500,000 websites.
Very often clients ask how they compare to their competitors. Hitwise compiles third party data for this very purpose, to track visit count, number of pages viewed and average time per visit. Because their information is sourced by focus groups and ISPs, it’s entirely possible that at some point they’ll be able to provide market-based conversion information as well.
While ancillary benefits include seeing exactly where your competitors get their traffic from - by site, engine or keyword, Hitwise used properly can provide proportional audience measurement (they don’t extrapolate their figures to the full web audience, as does Nielsen).
2. comScore - As the title suggests, this product is all about audience measurement of interactive elements. Particular products of the acquisition include Media Metrix, AiM, and netScore. According to the words of comScore President and CEO, Magid Abraham, Ph.D,
“Media Metrix, having created the industry, is a recognized standard in Internet Audience Measurement…It has a well established reputation for the most complete and comprehensive audience measurement for digital media, and a business orientation focused on superior customer service.”
Today comScore captures a comprehensive view of surfing and buying behavior of more than 2 million participants in an extremely cost-effective manner. While comScore provides services to garner audience permission to confidentially monitor their online activities, they get to measure every site visited, page viewed, ad seen, promotion used, product or service bought, and price paid.
To further recognize the limitless boundaries and technology-based challenges of the Internet, comScore built out:
- The industry’s largest work sample
- Unique college and university samplings
- Global coverage, with Internet users under measurement in 200 countries
- Measurement of AOL proprietary usage and other digital applications
- A relationship with Arbitron for considering not just the largest DMAs, but also samplings at the local level
3. Nielsen/NetRatings - This data research company offers a variety of products not just for traditional broadcast, but for internet audience measurement and analysis as well, and is thought to cover 70 percent of the world’s Internet usage. Here’s two of their more powerful products for media planning:
a. Nielsen//NetRatings @Plan
This resource reveals details of the online population – capturing everything from basic demographics to in-depth lifestyle and preference data. With more than 1,700 profile points, @Plan offers the industry’s most comprehensive profiling of adsupported Web sites available.
- Helps advertisers Determine the most efficient way to reach defined target audiences on the Web.
- Empowers publishers to contrast their demographic, lifestyle and product preference profile information across publisher-based websites.
- Aids ecommerce marketers maximize their multi-channel sales efforts by profiling the overlap between online, catalog and in-store audiences.
- Helps determine target audiences and calculate the size of those targets online.
b. Netwatch
It’s difficult to ascertain the current statistical reach of ACNielsen’s NetWatch product, though most recent estimates claim domain over 16 countries and surveys of 149 thousand individuals. Specifically Netwatch captures data relating to Internet usage and using habits:
- Total number of people using internet
- Profile of internet users
- Products ownership/services usage and lifestyle of internet users
- Recency and frequency of using internet
- Time spent using internet
- Activities on the internet
- Money spent on accessing/using internet each month
A Practical Application of Audience Measurement
One recent study by Nielsen/NetRatings spoke to the engine’s audience overlaps. Apparently 58 percent of Google users use a second search engine, the lowest of the top three search engines, while 70 percent of MSN users and 71 percent of Yahoo users turn to another of the top three. In terms of overall market-share, Google led the pack with nearly half of all searches, followed by Yahoo at 21 percent and 13 percent for MSN Search.
Altogether, out of 60 search engines surveyed, the top three garnered over 80 percent of all Web searches. Let’s all agree that this is useful information in the planning process.
Shortcomings of Nielsens
So we’ve presented a variety of interactive audience and behavior measurement tools in our opinion far more useful than what Nielsen provides in broadcast. Now on the chance that Mr. McConnell was simply comparing Interactive to Traditional and commenting how Interactive doesn’t present one unified system for audience measurement, like traditional has the Nielsens, we should first explain why broadcast only has this singular unified system in place:
False Idols: The Golden Calf
Initially there were multiple sources of data research, which obviously carried conflicts from samples, methods of data measurement, and wreaked havoc throughout the industry. Eventually television executives embraced the idea of offering one company a monopoly over the ratings business, simply to whitewash the complexities. So even if this current service provides inaccurate numbers (which it often does), those numbers have become standardized and accepted for purposes of media negotiation. So let’s not get too caught up in the notion of one unified system of audience measurement.
That aside, the Nielsens ratings system is far from a perfect model, and the least of that has to do with their statistical viability. While statisticians have the technical proficiency to prove statistical significance over an audience, even using miniscule sample sizes, they can also undermine that data-sampling a dozen different ways.
In writing this article I was curious to hear just how wide the Nielsen sample size is (I’ve heard it’s in the thousands, but wanted to confirm this). Here’s the closest answer I found: “By 2006 when all elements of the expansion plan are complete, the effective size of Nielsen’s National People Meter Sample will nearly double from 5,000 homes to approximately 10,000 homes”. In an ironic juxtaposition, Nielsen’s parent company VNU Media Measurement & Information Group employs nearly that many people (38,000), relatively speaking.
Flawed By Design
And now we come to the real problems with the Nielsen system:
- 1. It’s an electronic device people let into their homes, for the sake of monitoring their viewing (and recreational) habits. I don’t know about you, but I’d never be a Nielsen family. With all the hullabaloo and conjecture about online users dumping 40% of their cookie data due to concerns of privacy, who wants to be recognized as a habitual viewer of Beauty and the Geek?
- 2. The Nielsen set-top box can’t distinguish individual household members (just as cookies can’t). This especially concerns me in those houses where the teenage girl is watching 60 Minutes while Dad is watching the OC (actually, a lot concerns me about that scenario).
- 3. The networks completely manipulate how this data is used: Nielsens are applied to CPM rate cards and viewership data based on timely pools of data. March Sweeps? Please… The networks also stack their programming 4x a year to validate artificially inflated audiences, market-share and rate cards. This is nothing more than TV gerrymandering, an abuse of the system which undermines the whole CPM pricing model.
So back to Mr. McConnell’s initial remark, we don’t see the need for one unified measurement system. We also believe the interactive industry has pioneered various research and measurement tools far more credible and useful than those in broadcast. Ultimately we’d challenge any suggestion that Interactive spending comprising only 3.7% of overall ad-spend has anything to do with what P&G’s spokesman has claimed.
We’d actually like to suggest something completely contrary, that this miniscule proportion of ad-spend has a whole lot more to do with major brand marketers comfort and understanding of this new medium, its unique attributes and competitive advantages, rather than the compilation and availability of institutionalized research…
Randy Schwartz is Director of Strategic Development at Reprise Media.
Topics: Advertising: Online |
Great POV on the TV model, which is undeniably broken.
While the TV industry artificially inflates itself, don’t the media buyers and planners support that broken model as well? The fact of the matter is that so-called big mass market TV buys are an incredibly efficient way to spend markters’ dollars, and therefore an incredibly efficient way for agencies to scrape commission dollars off the top. Fragmentation of media - including online - destroys that model, so there will be reluctance. While ultimately it is in a media buyer/planners best interest to truly invest clients’ dollars, isn’t there a conflict in that certain media buys have higher margins than others - at least with today’s agency pricing models?
You also mention NMR’s inability to track users, just like cookies. In addition, you should also note simultaneous media usage. The current model is such that media are purchased as if there is 100 percent attention on any given medium. How many online ads are viewed while the television is on - or vice versa? And why is so much research dedicated to impression and reach, versus change in attitude or achievement of a business result?
Finally, I must call you on your quotes about WOMMA - because you are incorrect and misinformed. What that organization is doing is amazing. Is word of mouth difficult to measure? Absolutely. Can it be purchased like an ad buy? No, but that is precisely what makes word of mouth so powerful, and it works in tandem with all the other controlled media channels. Btw, NOP Roper and many other major research companies that study consumer attitudes find that the relative influence of word of mouth has become much greater in recent years. Why? While all the push-model cram-it-down people’s throats advertising is experiencing some backlash. (For the record, I believe in advertising and it is a hallmark of our democracy. But relative and qualitative impact and importance is changing.) You should note that with word of mouth, people are in control. But word of mouth can be inluenced. Take a company like BuzzMetrics. The billions of real, naturally occuring conversations they’re track!
ing and analyzing have increadible value for brands who want to understand how word of mouth works for them. How their push-model marketing is impacting natural discussion and the attitudes of customers. How and why certain strategies became viral and others don’t. Why people talk about certain brands and not others. Do me a favor, please call Andy Sernovitz and get yourself educated on WOMMA and the terrific work they’re doing before you write them off.
Don’t forget Compete, Inc. - http://www.compete.com
They also provide market-based conversion information as well.
Message to Max-
We appreciate the well thought-out response, contributions like this flat-out define the community we’re attempting to foster with Searchviews.
As for the specifics, I agree there is a critical divide between media because the different media do carry different agency commissions. But since we’re talking about ways to offer (unified) audience measurement, we might as well give those technologies the benefit of the doubt and expect to enter a new era where those traditional agencies are forced to offer accountability for their spend, *and their returns*. That should level the playing field and prohibit a shop from buying traditional where the venue pays their commission as opposed to other media where the client does (and a smaller commission at that).
Totally agree that research should be dedicated to change in attitude or achievement over impression and reach. The latter is simply accountability of spend, doesn’t really provide much to business metrics.
Finally, I wasn’t suggesting that word of mouth is any less critical than our perceived media, I merely suggested how much more elusive it is and difficult it’ll be to manage within codified research methods.