WASA launch: tough politics

Firstly, here is my list of disclaimers before I even start saying what I have to say:

1. I work in the media and publishing industry, the so-called “old-boys club” serviced by the Online Publishers Association
2. I work closely with Matthew Buckland, the current chair of the OPA
3. I know almost everyone involved in this situation and some of the relationships are less cordial than others
4. No vitriol in this post is intended personally

So here it is: this week a new online stats auditing company launched calling itself the Website Association of South Africa, WASA for short. The initial list of sites in its rankings are impressive, and that list exists because the founders of WASA used their own OPA accounts to extract the current dataset to populate the site. This was done without the explicit permission of the OPA or Nielsen Netratings, the auditing firm that collects the data, and the information is not on public record because the OPA releases their stats quarterly, not monthly, to the public.

Andy Hadfield posted a blog entry about WASA and an interview with Jonathan Smit, one of the WASA owners. Before the interview starts, Hadfield says:

My take on the OPA for those interested. At first glance, disliked them. In fact, launched various petitions under the guise of my previous company. It took to the market like a bloated old boys club, forcing sites to pay to be members of an association with no discernable value besides “audited” statistics that, theoretically, advertisers should trust.

That may have changed, I’m not sure. Like so many other things, once a product leaves a bad taste in your mouth, you rarely return to graze.

Here starts a history lesson for those who like to shoot from the hip: The value of the OPA is precisely in the 3rd party audited statistics provided to advertisers. When the OPA was formed, every web site owner had a different set of data with different variables to show advertisers. Some would be talking about hits, others about page impressions, others about user sessions and others about unique users. This was in the aftermath of the dotcom bubble bursting, when advertiser confidence was at an all-time low. It was before Google Analytics and the Web 2 explosion.

In order to collect credible audited statistics at the time, the OPA contracted Nielsen Netratings for this service. They are an impartial 3rd party who audits not only the collated statistics but the measurement algorithms themselves, and this information is then handed over to the members and advertisers. The impact of this system was significant: now an advertiser had a single source of information for the sites they planned to advertise on, with a coherent and consistent set of metrics by which to compare the various member sites.

Currently OPA membership costs R20,000 annually – that’s R1,666 monthly to have these stats counted and audited.

In the interview, Smit says of their relationship with the OPA:

We see ourselves as working hand in hand with the OPA. We have already met with them and our meetings have been very positive.

Regardless of the truth in this statement, who is actually going to believe Smit when one of the other partners, Rudolf Muller of MyADSL fame, has criticised the OPA at almost every single conference I have been to. There is no mistaking the hostility, regardless of how its being spun now. Time and time again I have heard Muller mention the OPA membership fees being a problem and I don’t disagree with him: in the time of Google Analytics it is ridiculous to pay R1,666 a month for statistics alone. The difference is that the OPA don’t just provide audited statistics that have been presented to them, as WASA plan to – they also provide a audited measurement system and a collation of results across the member base.

I posted a snarky comment on Hadfield’s post about it being unethical to import a dataset from a competitor for launch, though I was talking about Amatomu and Technorati, and this evoked a response from Smit:

Unethical would be if we “sucked down” the OPA / Nielsen’s data and pretended that it was our own. We however, clearly state that the site is an OPA site as well as the statistics source as Nielsen’s Netratings (As we do for Google Analytics).

Now this is precisely where I take issue with the how rather than the what of WASA. For the record, I support WASA and think its a great idea and that it will fill a much-needed gap in the industry. However, there are two conditions under which Smit’s argument could hold firm:

1. All publishers listed on the WASA site had given their stats to WASA
2. The stats, as they appear there, are currently available to the public rather than to members of the OPA

Neither of these conditions are true. I know of at least two publishers who have not given their stats to WASA but have found themselves listed there.

Smit corrected me by pointing out that WASA uses a 3rd party firm Buys Inc to audit their system, and I am grateful for that.

The problem, however, is going to be the following: the OPA are perfectly within their rights not to allow WASA owners to use their OPA accounts to populate a rival system for launch. WASA can’t take those stats down now without some embarrassment, but what they should have done is approached each publisher, explained what they were about to do and asked the companies to submit their OPA stats to WASA for listing on the site. As I said before, the problem is not what happened, but how it happened.

My weather prediction for the next week: hot with no sun.