Why We Must Do Something About The Impression
He isn't alone in opposing this call. Just two days ago, my esteemed colleague Tom Hespos made a similar call. The reasons both of them use to convince us we should stick with the impression differ slightly: Mr. Smith says it's because the impression offers us a glimpse of the most granular level of activity in media so 'why should we give that up?' and Mr. Hespos says it's because, so it seems, the impression is the best system we've got at our disposal and the alternatives proposed still need a great deal of refinement.
I think they are both right.
However, I feel that by anchoring ourselves to the impression, we will continue to be stuck with a system that works on compensation to publishers for delivering what is in reality a small audience unaffected by the advertising manifest in the chosen environment.
One of the big motives for needing a new currency is purely psychological. Unless the industry is willing to flounder until brand managers and marketing VPs who speak digit-ese and for whom online advertising is a de rigor come of age, we've got to find a way to make the ready-to-hand components of online advertising (i.e. impressions and creative units) comprehensible in quantifiable ways that make the online space run flush with the rest of media. The impression is only truly valuable in a unique way, distinct from offline media, if it can serve as a measure of advertising exposure to a unique audience instead of vehicle exposure to an assumed audience. Unfortunately, right now, it can do neither.
The primary tenor and thrust of what I've been arguing for is a better way to trade units of inventory that make sense to advertisers. If you want to be a missionary, you've got to speak the natives' tongue.
I also believe that the ultimate trade needs to be an "audience," or, more precisely, a universe. All us media nerds understand what's going on with an impression as it is currently being employed, but, as you know, advertisers are mostly interested in the equation (in all its forms): "X volume of media = Y number of unique individuals = $Z in projectable revenue." There are variations of consideration, accounting for repeat business and such, but I think you readers get the gist.
If the impression is to remain as a media currency online, then it will have to be tempered with a real audience, something that is NOT being done at large. Without translation to unique audience, the impression itself has no meaning. Hopeful frequency controls alone are not going to be the answer for this, though I see them as going in the right direction.
One of the models I have proposed is fixed positioning. For a property that acts a lot like print, such as Salon.com, special interest content sites, financial news sites, newspaper sites and others, the inventory could be sold much like print inventory: a fixed position that exists on a page for the life of the issue. When I buy print, I pay a CPM that is derived from a guaranteed rate base -- X number of unique individuals, in theory, will have an opportunity to see my ad. When I "turn the page" on an article, there may be another advertiser there who also has a fixed placement.
The reason I propose this model is because it has analogy in the offline world (print). It also might do something to mitigate the "what size is my audience" issue. Though others rightly point out that this is a new medium and we should be doing things "newly," I suggest that just because one CAN do something doesn't always mean one SHOULD do something (see "Frankenstein," stem-cells, disco). Eventually, entirely new models will emerge and work better, but that shouldn't stop us from trying new things with what we have now until that 'something better' comes along.
There is an issue that has seldom been mentioned in this debate, Mr. Smith being the only one I can think of in recent discussion. That is the issue of inventory availability for other advertisers. It is true that fixed positioning would put a strain on available inventory to other advertisers. I see this as a red herring, however, since advertisers don't seem to be bothered about this in other media that works on this principle. Not everyone can have positioning across from the table of contents, or a spot in the top of the pod in the first commercial break in 'ER.'
When you come right down to it, I guess the problem with 'just impressions' is that in offline media they are determined with audience as a variable, engaging a media that has traditionally operated as "viewer views, listener listens, reader reads." With online media, currently there is no attempt to do the same. I figure proposing an alternative to the status quo might wake some of the more inventive media minds from their dogmatic slumber and figure something out!