Two things are noticeable – first, with the ever improving email platforms now in use, companies are getting to grips with creating relevant, timely and engaging emails. Second, these emails rarely live up to their promise once you click through.

I showed an Aer Lingus email, nicely put together and full of useful info and links just four or five days before I was due to fly with them. A good example of picking up a relationship with an ongoing dialogue that immediately connects you to a conversation with the brand. My expectations were indeed raised, especially as they used my name, my reference, they seemed to be talking to me.
Sadly, clicking through brought me to a top level generic homepage. Not only did they not remember me, but they even advertised a fare to where I was going at a price lower than I had paid!

The message for brands is, if your departmental structure, systems, etc. don’t really support this sort of personalised interaction, then don’t set the expectation that you can.

Marketing budgets continue to grow, IT spend on platforms continues to grow and spend on Web Analytics internal staff continues to grow (econsultancy Online Measurement Report, published yesterday).

All good signs.  On the flip side, different messages came through at an event in London yesterday (Figaro Digital Analyse, Optimise, Realise).  Nick Pandya from IMRG shared latest figures with us – all relentlessly positive except one – site conversion rates have continued to fall throughout 2009, just as they did throughout 2008.  I asked Nick why he thought this was so. Some recessionary effect plus a proliferation of sites that mean we browse more before we commit. I agree with both of those, and add a third, the biggest of all – most sites don’t handle arrivals well and don’t make it easy for visitors. Poor experience, poor conversion rates.

Repeated speakers gave the impression that once your analytics were sorted out, you could move straight to Multivariate Testing -  no need to talk to the customer or understand why things were happening. Whatever happened to the voice of the customer, actually listening to real people and how they experience your site? 

I’m being a tiny bit unfair. Ok, so the event was sponsored by web analytics vendor WebTrends – you would expect a vendor of site analytics and now MV Testing to say you need, well, those things.  Logan Tod took the same line – there was a slide which mentioned user experience in very small letters but really only addressed data analytics – again, we forgive them for concentrating on what they sell, and the presentation was otherwise useful. 

Most disappointing of all, the lone client towed the line, apparently not too concerned with what customers might have to do with it. A business analyst at Visit London for four years, he described the efforts to get all of the basics in order, cleaning up data sources and integration, generating meaningful analytics reports and insight. He said that last year he had spent most of his budget on getting a Single Customer View. I don’t think this was SCV in the traditional data warehousing sense, but a concentrated effort to move from aggregated site data to full tracking of individuals (where they came from, what they did) in order to properly link activities to results.   

He described how they had previously used too much guess work but now their decisions could be underpinned by decent data. This year his budget would focus on MVT. Our hearts dropped..what about understanding why customers do what they do. We’ve done all that work to properly understand ‘what’ is happening and now, without understanding ‘why’, we’re off to do MV Testing.

In the questions that followed, a member of the audience asked the one I wanted to ask – “You’ve found out what is happening, how do you find out ‘why’ customers are behaving this way?”  The answers, tragically, was “perception and common sense”.  So, after all that effort to remove guess work, we’re back to guess work (albeit two guesses at a time!).

I’ve had many debates with MVT vendors (Google Web Optimiser, Maxymiser) and we’ve discussed the danger of using MVT as some sort of replacement for elements of site design, but it’s key we don’t lose track of the role of each part. The foundation is quality analytics (so well done Visit London) – this will tell you what is happening. Then you need customer input (feedback, surveys, focus groups and best of all, customer experience testing) – this will tell you why it is happening. This is the key input for web designers and architects to do their stuff. And once we’ve done all that, we have a great new tool – MVT software to try out more subtle changes in text, visuals, layout – this is the process of refining the site. 
With these three keys stages continuously in operation, site experience and conversion rates can be continually optimised as part of day to day business.

Another good session from Neil Argent and his team at Figaro, and useful updates from Webtrends and Logan Tod on the progress and application of analytics.

Yet any web user will tell you that most sites are a poor experience for the prospective customer. Typical moans are difficulty in finding what you want, confusing navigation, and that it is simply not easy. Despite this, paltry sums are spent on making sure that sites are easy to use and deliver completed sales.

How is this so?

In most cases it comes down to that age old division between Marketing and IT. In the Blue corner we have IT, responsable for big budgets, big projects, big technical builds with back end systems and technology and wires and all that stuff. Not a huge amount of customer knowledge being applied at that end.

In the Red corner we have Marketing, with great big budgets for driving Direct Response, otherwise known as Traffic. Lots of expertise here in concepts, advertising themes, campaigns – things that get the punters in the door (where they find the IT bloke under the desk sorting out cabling).

Somewhere in between, there is the online experience, and no-one really owns it. Marketers fire traffic at it, IT build a robust platform with disaster recovery and, er, fat pipes. A junior analyst runs a weekly report that says there really are quite a lot of page impressions. Saddest of all, customers wander about on it, unable to find what they are looking for, struggling with the payment screens and generally finding it easier to go elsewhere.

A positive development in 2009 was a growing trend for marketers to task agencies with actual sales rather than just traffic (CPE – Marketing v Sales) but the bigger challenge is within the companies themselves. There is a crushing need to align marketing, sales and service (excuse the use of these old fashioned, pre-digital terms…). I worked with a big UK company last year that saw a big drop in conversions when they launched a new site. I was working with the Marketing bods – they weren’t too happy but it had no impact on their marketing spend – millions of pounds comtinued to deliver traffic to a site that couldn’t turn it into sales.

A good starting point will be an alignment in the measures of success for the different departments. Imagine those that drive the traffic, those that persuade that traffic to transact, and those that enable the whole thing, all pulling in the same direction and all rewarded for the same success.

There’s a whole process of audience research, persona and scenario creation, customer journey analysis et al before you come up with a site design that is likely to serve your customers well and deliver good conversion rates. And with existing sites, customer experience testing laid over web traffic analytics gives a powerful “why” over the “what” of the analytics.

The danger with multivariate tools is that much of this is forgotten – if it’s easy to create different treatments and see which is best, why not just try all sorts. Play with the copy, change the colours, and see which works best.

This is akin to lining up your cannon, firing out two random shells and then celebrating the news that one was closer to the target than the other. You could do this for years without getting near your target. Better to remember the basic disciplines of website success – understand your target audience, the journeys they take from interest to sale, and how they experience your brand offering online. Get the basic offering right, then use multivariate testing to tweak the offering for continuous improvement.

There is a valuable role for multivariate testing, but uninformed site design is not it.

So far so good.

But when advertisers place media budget with an agency, the agency control is only in delivering the best possible marketing effort, driving volumes of qualified prospects to the website. After that, it is the responsibility of the website to convert the prospect into a sale, be that a purchase, a download or subscribing to a newsletter. And that is in the hands of the website owner, not the marketers.

If the CPE model is to work, advertisers must address the Sale, not just the Marketing. And that means a shift in mindset – when consumers are confident that clicking on an ad will take them to a place that is a good experience, engaging, easy to use, maybe we’ll see clickthrough rates finally start to rise again.

Meanwhile, CPE pits Marketing against Sales and that is not a recipe for driving performance.

All this tracking effort has been racked up a notch with the advent of behaviourally targeted online advertising. This week a Federal Judge in Seattle ruled that IP addresses are not personal information based on the premise that the IP address identifies the computer not the person.  Not so the EU Commission, who decreed that an IP address is “personal information” and have instructed search engines to “expunge users IP addresses as soon as possible.”  IP tracking is already used for tracking Internet behaviour by Google as well as AudienceScience (formerly Revenue Science) who provide behavioural targeting for several large UK media publishing groups.

The UK has historically always been more relaxed and passive about privacy matters.  As a result we have surveillance cameras everywhere we go. But in Europe the pressure for action is already greater than in either the UK or the US.  The European Commission is investigating the UK Government’s role in behavioural targeting tests by BT and Phorm in 2006.  In the US, a group of over 5,000 companies that includes Google, Microsoft, Yahoo and Disney has just created a report called “Self-Regulatory Principles for Online Behavioral Advertising,” attempting to head off any aggressive Federal regulation there.

There is clearly consumer confusion at the very idea of privacy – millions have Nectar and Tesco Clubcards, but there has been resistance to the National ID scheme. So for behavioural targeting, what do we need to beware of if we are not be spooked by it? Understanding behaviour is part of the story, but when it is used inappropriately or in a way which doesn’t add value to the message, yet exposes the machinery behind the ad, we get the worst of both worlds. If we understand properly both the consumer needs and the context in which the ad is displayed, then we can give consumers real value (or at least a perception of value) without invading their space.

This post is also published as  Guest Comment in this months The Leveller magazine, published by i-level

http://tinyurl.com/theleveller

At first glance this is a bit depressing – how do you address an issue that isn’t fully under your control..? However, this isn’t the whole story – a proper analysis of the language shoppers use shows a different picture – not so much that Delivery Cost is a barrier, more that it annoys shoppers. There’s plenty we can do about this…

In many sites the delivery costs are hidden until the very last stage of payment. In other cases the default method of delivery is not the cheapest and the shopper has to (a) notice this, and (b) specifically choose a cheaper method if they want the cheapest deal.

One of the other top barriers was that of Trust in a site – not exactly unrelated – if we behave in a trustworthy manner, be upfront with our pricing and the options available, then even the Cost of Delivery can be made less annoying, even if it won’t go away completely….