Web Analytics in the web 2.0 world? What implications for Web 3.0?
E-Consultancy’s Richard Maven recently interviewed Bryan Eisenberg about why people buy online. Bryan touched on a number of great points about stuff I see every day with clients this side of the water. We don’t do persuasion architecture as defined by Bryan and the folks at future now, but we do a lot of similar things at SATAMA. Combining qualitative (Bryan talked about one subset of this – Internal search keywords used) with and quantitative data is a must if you want to find out customer motivations in order to cater for them on your website.
Web analytics alone can only pinpoint areas which need attention but qualitative data like usability studies, user surveys, heuristic analysis (Bryan gave an excellent example about Overstock and the three questions PA uses to do such studies), purchase abandonment (at point of sale, pre-sale and post sale), interviews and much more raises stuff you couldn’t dream of getting from web analytics data alone.
But that is not what I wanted to talk about. Bryan pinpointed what I think is a huge failing in most web analytics tools today and I would be interested in their (the vendors) opinions. I quote;
It˘‚¬„˘s not really about reports ˘‚¬€ś it˘‚¬„˘s more of a holistic view. That˘‚¬„˘s where a lot of analytics fails today. Most analytics tools today are fancy hit counters. That˘‚¬„˘s why they are having issues around Web 2.0 and Ajax.
I couldn’t agree more. Now I am not blaming vendors for the past. In part I am as much to blame for not shouting loudly enough that rich media applications and the social media explosion has not been thought about in tool design.
I even would go as far to say that measuring social media is overhyped. Why is it needed? how do they fit in with your KPI’s? What are the business reasons behind knowing that ‘x amount of people’ talked about you publicly online? The question “so what?” comes to mind. Ask me if I think what they are talking about is important and I’d agree 100%, that’s qualitative data I can use to back up my quantitative data. However I’m drifting off the point, I’ll cover social media in another post.
So now I am SHOUTING it publicly. What happens when we go mobile? Web 3.0 is coming and it’s coming fast. We’ve known this at SATAMA for some time and maybe haven’t shouted loudly enough. Eric Peterson was kind enough to point this out in his visit to us last week. I’ll use his example (and I hope he doesn’t mind). How do we measure the mobile phone? How can we push relevant ads at people who ask for them when they need them in real time? if I search on my phone at Google for a cafe, I want the cafe’s in the vicinity to appear (the cafes in the surrounding two blocks) not just the cafe’s in the city I am in. Would I even use Google? Would it be Google Maps? Would companies like Nokia or Samsung make something better than Google Maps that means I never visit the cafe’s website but find the cafe anayway? How does the Cafe owner know where I came from? How does web analytics fit into this picture? What does it all mean? Who am I? 
Ok so I got a bit carried away. Web 2.0 rich media especially is one area which was not addressed well by the vendors. How do we plan for Web 3.0? when we can track positions of people geographically by their phone? what are the implications for web analytics? Opinions?
Hi Steve,
You say that you would be interested in my opinion (a.k.a. the vendor) – which, as you know, I am more than happy to elaborate on.
AND you state that you agree 100% with Bryan (which I really respect a lot) on the following quote:
>>It’s not really about reports – it’s more of a holistic view. That’s where a lot of analytics fails today. Most analytics tools today are fancy hit counters. That’s why they are having issues around Web 2.0 and Ajax.
I think your post is super interesting and smart – but I think you and Bryan might mix up the elements in question in this dialogue between Web 1.0 Analytics and Web 2.0 Analytics. That being:
- Data Collection
- Data Representation
- Data Interpretation (creating insight)
So to begin with. I absolutely agree that it is not really about reports. (I have made several public statements about how important is to understand the difference between reporting and analysis). But to say that we (the analytics vendors) are having issues is a bit off in my humble opinion. I think “WE” should be defined as vendor (me) + consultants (you Steve) + practitioners (our clients).
That said. The data collection part is doable and everyone of the enterprise vendors today operate with external data sources (e.g. qualitative data) AND on an event model (being able to collect ANY web 2.0 events). All the vendors can segment, filter and present this data any way a client wants. SO the challenge is for somebody (and I still see that as an evolution and combined effort from vendor + consultant + practitioner) to come up with a “standard” for
A) WHAT data is it that we are expected to collect (from a best practice point of view)
B) HOW are we supposed to report on this (for best possible practitioner understanding)
C) and finally how are we suppose to CONCLUDE (success, failure and impact measurement) on the presented data
Everyone of the above questions are technically possible and most of it solvable from the tools today. But WE (still vendor + consultant + practitioner) have to decide on this together.
I am only talking about web 2.0 here… web3.0 ..hmmm can this wait?
May I draw your attention to the fact that Google JUST LAUNCHED “Google gadget ads” which includes the opportunity to Receive site-by-site interaction reports tracking actions. This is actually Google taking a decision on A, B and C above and they created about 60 metrics. Perhaps we should just adapt and build on those?
LINK:
http://www.google.com/adwords/gadgetads/tutorial.html#interactionlist
Cheers Steve.
Dennis R. Mortensen, COO at IndexTools
My Web Analytics Blog