The problem, here, Eric, is that we’re not all large companies with massive budgets for consulting. My agency has fewer than 100 employees, and there’s yet still a huge need for integration — as I’m sure there is within thousands of other agencies and marketing departments across the globe.

What’s necessary is a universal ASP solution which is easily scalable in terms of cost and bandwidth, and maintains an independent objectivity enough such that the DARTs, Googles, Optimosts, etc. of the world will all be willing to share their data with it. The technology itself is actually pretty simple, somebody just needs to figure out the revenue sharing and bring it to market.

What’s even more interesting about the idea of this kind of “one tech connects all” play is that it alone might have the chance to finally analyze the effect that multiple tactics have on individual consumers. If set up properly it might be able to examine the most effective marketing mix model and suggesting where the levers need to be scaled back or pulled harder in order to maximize cross-campaign efficiency from general brand/product awareness through to engagement, conversion, and retention.

Omniture had (still has) the chance to do this with Genesis, but they’re going to screw it up with a combination of terrible customer service and a focus on the short-term revenues. The cost of Genesis integration is just too high for most potential partners to be willing to swallow right now. One of our firm’s divisions (OTOLabs) is presently one of the most well-known producers of widgets and distributed media, yet they passed on Genesis without a second’s thought when they heard how much Omniture wanted to charge for the initial setup.

Had Omniture instead made Genesis integration open source but instead leaned on revenues generated from how clients apply this cross-tech data they might have had something built to last. Instead now I’m looking to see who else might be able to pull it off. Maybe Accenture comes out with a smaller do-it-yourself model.

However the very fact that Google Analytics is free proves the technology available is now starting to become robust enough to handle integration on a larger scale. GA handles 10 times (conservative estimate) as many customers as Omniture with far less problems on integration, think how well adwords works with GA. Ok we know GA isn’t perfect, it isn’t the answer, but it’s a step in the right direction.

At the moment the Omnitures of the world have yet to capitalize on the processing power of the world wide computer as Nick Carr calls it but I would say it’s simply a technical issue that needs to be fixed. That’s easy to do with a bit of investment so in the future I think that’s what we’ll see because the reward is too great to ignore. Omniture have only scratched the surface of their potential in my opinion.

Right now you’re right - integration can’t be outsourced well so it generally isn’t and businesses do it in house or hire companies like Satama to do it with Excel! I do believe this is part of the reason Accenture have made a move into this field as well because they see the integration problem as an opportunity.

However I think it’s being short sighted if they don’t think about fixing the problem at the ASP level as a service. If they fix the integration problem once they shouldn’t ever have to do it again. Just supply us with the framework to do what we need to do and support us in setting it all up and we’ll do the rest. Give us all the control, allow us the freedom to upload any dataset, allow us the ability to create our own dashboards, delivery mechanisms, predictions. Build an ASP that does all that and the worlds companies would use it. It’s a brave new world, with just a handful of mean we can start all over again! (No wait that was war of the worlds!:)

I would ping the question back do you believe that it’s better to pay consultancies thousands of euros a day to crunch data into a digestible form using excel (or something like it) when this could and should be automated by logging in somewhere? I know most businesses HATE that kind of expense but we have no option but to charge them for it.

Currently in the non-Visual site/load world that’s what we have to do. An opportunity I think.

Also, Steve I wonder a little bit about your thoughts on the “ASP-ification” of the world. I know it sounds compelling but I wonder if that will just make the integration even more heinous than it can be today? I mean, there is the potential to remove the little bit of flexibility companies have today when they own their own datamarts and have (mostly) full control over the flow.

Charlie basically highlights this noting that his ASP service doesn’t let him integrate with their own products. So what happens when your ASP is trying to talk to another ASP? I know, I know, this is what APIs are for, but I think we’re just now barely starting to see APIs in the field and based on what I’m hearing they’re really not providing the value advertised.

Anyway, thanks for pushing the conversation along. I think everyone else is in Utah and not paying attention this week but I’d love to hear your thoughts about who ultimately will be doing web analytics integrations in 5 years.

Totally doable, but why doesn’t an Omniture or someone who already has relationships with many of these data vendors already realize the market need for this and make it far simpler.

I just realized today that I can’t use Omniture Data Warehouse to pull SearchCenter campaign spend data along with site metrics like Application Rate and Average Sale. Because Data Warehouse doesn’t integrate SearchCenter engine spend data in with SiteCatalyst site activity data.

Are you kidding me? This too I have to do myself?

You’ve touched on a pain point of mine. At a recent Web Analytics Wednesday in Helsinki I was asked “What is the hardest thing about web analytics?”

I replied data integration. I constantly find myself using Excel do bring together various sources of data because other tools (like Visual Site) are not readily available.

However Omniture & Visual Site does offer a solution to this problem with Visual Load. What I would like to see the free & paid platforms develop is an easy way to integrate data (any data) into the mix. Then allow us as analysts make up our minds how we want to use that data.

I do see the ASP solution as a growth industry. As I menetioned in my response to Ian’s post as utility companies grow in size thus allowing more robust ASP solutions to come along I can see a time when all the big companies supplying BI systems, IBM, SAS, Oracle, SPSS etc will be run via ASP. This in my view is the biggest threat facing these companies today as more and more of the IT requirements are outsourced.

Nick Carr wrote a good book about this called The Big Switch, where he compares the outsourcing of electricity to the outsourcing of computing power. In the early days of Electricity everyone had their own power generators, till a subsidiary of GE invented the global network that supplied it all reliably and at a fraction of the cost.

This is what Google (and others like Amazon, eBay) are doing now, and how they can offer you an mail account with 6GB of free email space.

If I were Omniture I would be looking to compete with the BI guys by building BI systems as an ASP, add predictave tools to an ASP run version of Visual Site/Load and you’re halfway there already. More likely they will end up being bought by one of these big BI players when they start getting too competitive and force the BI company to shift to their model.

If we could buy on an outsourced basis all the tools we needed without having to invest in any IT people, hardware or software then you can bet it will happen, simple economics forces the situation. The companies that capitalize on that are the new GE’s of the world.

As for predictions on time for these kind of developments, I hate making these because things move as fast as the market requires and because the mainstream market hasn’t yet really caught up with web analytics it’s very hard. Think how many paying customers IBM and Microsoft have, that’s how many customers need to have web analytics installations before time becomes predictable. 5-10 years at least.

Cheers
Steve