John Wyllie is the global web analytics manager at GE Money – a role in which he advises and directs the analytics activities of the company’s numerous international businesses.
We asked him a few questions about how analytics can best be managed within organisations, and how the financial services industry stacks up against other sectors.
How is web analytics managed within GE Money?
At GE Money, we have about 50 businesses in 30 countries worldwide. I don’t work for any of those individual businesses; I work for the global HQ. That global function provides thought leadership, best practice and support across all these different businesses.
There is a global e-commerce team within HQ that has experts in CMS, online marketing, affiliate marketing, usability, design and analytics. That’s me. The idea is the global team supports all the businesses worldwide to make sure they are getting the most out of their activities online.
I’ve been with GE Money for around 18 months now. The year one challenge was to get Omniture deployed across all our websites worldwide. It was a technical challenge. Now that we have achieved that, we are concentrating on using that data to make a difference. There’s no point having analytics unless you are using the data to improve your business.
How are you going about achieving that?
GE Money has an analytics research team with research centres in Canada, India, central Europe and China.
In conjunction with the centre in India, I have a team of analysts that helps with the implementation and analytical side. Their job is to do the analysis and provide recommendations. We also have a couple of consultants in Canada, and myself in London, so we can support our individual businesses 24/7. That’s important as we are a global organisation.
Web analytics isn’t just about reports – it’s about insights and recommendations.
Has the skills shortage been a problem for you?
Yes, it is hard. Recruiting offshore was a challenge – finding people with experience of Omniture was difficult.
When I recruited the team, I assumed it would be difficult to find an experienced Omniture analyst, or experienced web analyst in general, without paying them large amounts of money. So what was important for me was looking for someone who was passionate about the web and was internet savvy.
I also recruited a mixture of people from a technical background – developers and so on – and people from a business data analyst background. That has worked well.
Do you use Google Analytics at all?
No. One or two businesses may have it from a legacy implementation. People say Google Analytics is free and it is true – it is free.
But for me, that cost is irrelevant because where we really want to spend our money is the people and the talent that will drive value, as well as the technical skills to get analytics implemented.
Would you agree with Avinash Kaushik’s assertion that you should spend £9 on analytics staff for every £1 you spend on analytics technology?
Oh, at least. Without a doubt. A lot of people say Google Analytics is free and provides a lot of functionality. I don’t care.
Where is the financial services industry on the analytics learning curve compared to other sectors?
Financial services are generally a bit behind other verticals. They have tended to be a bit slower. I worked with Lastminute.com and Channel 4, and from what I have seen in this industry I would say financial services is a bit further behind.
There is probably more of an appetite for numbers but they are not as internet savvy.
What other sources of data are you looking to integrate with your analytics findings?
One thing we have done recently is integrate our DoubleClick DART for Advertisers (DFA) data. We use DoubleClick’s DFA tool to display ads on other websites and we have integrated the data into our SiteCatalyst tool so we have a full end-to-end view.
That went live a week or two ago so I can’t tell you yet exactly how that has gone. We’re still getting to grips with what it has given us. But we do know that visitors touch different campaigns. Using Omniture, we use campaign stacking to see if people that start off banners use search as well, for example.
Also, the plan is to integrate our offline data using transaction ID functionality. Currently, you can make an application online and in some markets you can get the decision online. But we want ultimately to see whether they have gone to book, whether they have been profitable and all that data is all offline. We want to integrate offline profitability and things like that.
The challenge with that tends to lie in the offline legacy systems and making sure they can be integrated. They vary from market to market.
Can you tell us a bit about how you are using participation as a key metric?
With Omniture, you track key success events, such as a sale or download. For us, it is an application.
With Omniture, there is this participation feature where you can see how other parts of your website participate in those events. We use participation to see which pages and content areas influence those success events on our site.
Are you doing anything in the realm of content optimisation yet?
Not at the moment. But we are looking at changing our CMS platform soon, and the new platform will allow dynamic content based on visitor profiles. That is something we will be looking at.
For instance, there’s the classic example of when someone comes to you searching for a credit card, you make sure that page shows a credit card, not a mortgage. Very simple stuff – but stuff that is often very effective.
Are you looking at widgets at all, and how they can be measured?
People do get excited about that sort of stuff. But generally, in banking, we want to make sure we are offering a good banking experience and making sure people can service their accounts online.
So widgets and things like that aren’t a huge priority for us right now.
Web Analytics Buyer’s Guide