Okay, I lied when I said my next post would be evaluating Search performance. I’m interrupting my series on doing real-world web analysis (currently focusing on analyzing Internal Search) because I wanted to talk a little bit about a problem I mentioned in an earlier post that has been much on my mind recently.
In my ‘10 Things That give us all Ulcers’ post, two were specific to tagging:
1. The Plain-Vanilla Tag
Tool vendors often bring this problem on themselves and their clients by overselling the ease of putting a tag on a page. Yes, you can have measurement in an hour. Will it meet your real needs? Probably not. I see lots of companies commit to the plain-vanilla tag knowing that they will have to come back and fix it but wanting to get a deployment out as quickly as possible. Usually, I think that’s a mistake. The pressure to release numbers is always overwhelming – and whatever gets rolled out is immediately in-play. That means the organization starts to use and react to the numbers – almost always before they’ve been adequately tested.
2. The Tag as Software-Development Project
There aren’t two sides to every web measurement coin – but it’s surprisingly easy to either under or over-do your tagging effort. At the opposite end of the Plain-Vanilla tag spectrum is the tendency to treat the tag like it must be a fully-engineered software development project. It’s this tendency that sometimes causes business managers to just throw their hands in the air and scream – ‘Let’s just roll the damn thing out!’ A tag is simply not as complicated as even a very basic software development effort. It has no GUI, the number of options is paltry and the amount of code is about 1/1000 that of even the smallest software developments. IT organizations that haven’t ever implemented tags and don’t really understand the technology often give Business Units wildly inflated estimates of the time and effort involved. If you’re seeing big-ticket numbers around tagging, your best solution is to work with your vendor to train and hand-hold IT (we do this too – but for this particular service the vendor will be just as good). A little bit of training will almost always bring on the "AHA" moment where the IT guy says – "Is that all there is to this?"
A big part of the problem here is the way many vendors have chosen to implement the tag – making it responsible for pushing lots of information that really doesn’t belong there (I basically include everything in this category except the URL and data like Internal Search Term, Customer Characteristics and Shopping Cart detail that is otherwise unobtainable). A tag – the ideal tag – should be the very minimal version suggested by the "Plain Vanilla Tag." Except in most vendors’ products that isn't ideal at all.
It’s funny, because when we first got started implementing web analytic tools, this was a very big deal to me. And I wrote quite a bit about it. But as you do more implementations, two things happen. First, you get pretty good at them and they no longer seem very difficult. Second, you stop thinking about how things should be and tend to accept them the way they are.
What’s brought me back to thinking about this problem is the sheer number of complaints I’ve heard in conversation about how difficult companies find the tagging/implementation process to be. This kind of makes sense, because most people who implement a tag are doing their first one, not their twentieth or thirtieth.
That, and the fact that I’ve been working a little bit recently with a tagging implementation that is much closer to the ideal.
The solution I have in mind comes from Unica – a company that acquired and has upgraded and re-released the old NetTracker solution as NetInsight. They’ve implemented tagging just about exactly the way I would have done it.
How does it work?
It’s pretty simple. The basic tag passes all of the standard info that every tool collects. It passes the entire URL back to the system. In the system, the Administrative Interface lets you refine the URL by dropping parameters you don’t want – and then attaching a Page Name to the result.
Let’s start with why this is a big advantage. Many tagging systems give the implementer five basic choices: use the Page Title, use the URL, put a hard-coded Page Name in every page, use Javascript to create an ad hoc page name, or use a Page Name generated by a CMS and embedded in the URL or Page. Of these, only the last is usually acceptable. And if it isn’t available then you are in for some heavy sledding.
Page Title is a terrible choice – because if your SEO efforts haven’t made Page Titles useless yet, they almost certainly will in the very near future. I hate this – and I blame Google and the other engines for encouraging lousy UI practices. But it is what it is, and Page Title simply isn’t useful.
URL is cursed for other reasons – including lack of readability and the likelihood of parameter values that don’t mean anything differentiating pages.
Hard-coding page names is just plain awful. And using Javascript to build a Page Name is often impractical and always both laborious and dangerous.
With Unica’s method, you get to strip out parms you don’t want and then attach measurement page names. The mechanism for parameter identification is Regular Expressions. Anyone who’s ever used REGEX will attest that they are a bear – but simple expression matching isn’t too hard. And you can hardly blame Unica for implementing what amounts to a standard in text pattern matching.
What makes the method really valuable is that everything else that can reasonably be done from knowing the distinct page is done in the Administrative Interface – not in the tag. So you build Campaigns, Content groups, Scenario / funnel steps and segments in the GUI not in the tag.
I’ve written before on the issue of Content Groups (page hierarchy) – and how unspeakably clumsy it is to embed them in the tag. To me, this is the biggest win in the whole idea. I’ve also written frequently about how useful tagless campaigns are as an analytic tool (a feature HBX shares – and which we use constantly in analysis) and not just as a way to track real marketing efforts.
Unica has also done a nice job handling cases where you really do need to pass additional information. Custom variables related to some internal state (and not derivable from the URL) can be passed as a simple Name/Value pair. For a programmer, that’s very natural. It minimizes what you have to learn – and it provides maximum flexibility in terms of passing what you want to the software.
What about page interactions after a page loads? This is very common now – and while most tools support it, the Unica approach is unusually clean and understandable. They’ve implemented a special "Event" Tag that you can use. It’s a single call that you differentiate by passing an Event Name and – again – sets of Name/Value pairs. Providing a clean mechanism for passing interactions and describing them flexibly is going to be increasingly important as organizations implement interactive technologies. And the Unica implementation should be a model.
In addition, the implementation provides some special Name/Value pairs that are genuinely useful. Among these are a flag to suppress counting an interaction as a Page View, a flag to override the passed URL and set your own and a flag to populate a User dimension.
If you’ve spent much time building tags, you’ll immediately see the value in each of these.
Most vendors have made at least some progress in moving measurement functionality out of tags and into the interface. I love the strides that Omniture and WSS have made, for example, with funnels and segmentation. But tagging remains much more onerous than is really necessary. And placing items like hierarchy in the tag often ends up essentially eliminating important functionality from the solution. In addition, vendors seem to be struggling to provide implementers with a simple, clean, logical and consistent tagging method for handling page interactions.
To my mind, Unica’s implementation is all of that – and raises the bar on what vendors should be expected to deliver in making tagging easier
wow nice!c
Posted by: otuyi | August 03, 2007 at 02:31 AM