Sunday, November 09, 2003
Personalisation is expensive! Not worth it! Doesn't result in more sales! Wastes money!
A few weeks ago, Jupiter Research published a well-reported paper on "personalisation". It suggests that companies would be better served if they concentrated on the basics - making the site easy to navigate and search - rather than wasting money seeking the myth of personalisation. For instance: "Given flexible, usable navigation and search, Web site visitors will be more satisfied with their experiences and will find fewer barriers to the profitable behaviour sought by site operators. In fact, good navigation can replace personalisation in most cases." A more intriguing point was: "More than 25 percent of consumers surveyed by Jupiter said they avoided Web site customisation because of concerns that marketers would misuse the information. A similar proportion avoided registering with a Web site, for the same reasons" An exec at Broadvision responds: "Anything can be done badly and expensively. Just because there are some examples that people have invested time and money in the wrong area, that shouldn't discredit personalisation as a whole. People should do a sensible evaluation of what's going to work best for them and their customers." There are some important lessons here for anyone doing website publishing, but especially for governments. First, unless the basics are right, there is no point pursuing more advanced features. Common sense? I'd like to think so but I'm not convinced that this has been well internalised in many organisations. So many sites are poorly constructued, lack the information needed, have search engines hidden below the home page and fail to signpost content that the basics clearly aren't in place. Let me put it this way, if the first box you saw on the Amazon.com home page wasn't "search our shops", right in the middle of the screen, how long would you spend hunting for it? If the tabs for DVDs, Photo, Books etc were hidden or changed depending on where you were in the site, would you persist? I doubt that you would. Finding the products that you want is simple on Amazon - search, tabs and well structured navigation. I spent 15 minutes on it today and spent nearly £100 - saving about £35 against high street prices (and about 2-3 hours on top). What I haven't figured out is why everyone doesn't shop this way. Maybe it's because many purchases are impulse buys, urgent purchases (i.e. "I need it now"), or maybe people still don't trust the Amazon model - but if you can buy a video game for £30 instead of £40 and only wait a couple of days for delivery, wouldn't you do that 9 times out of 10? Once you're past the basics what more would you look for? I think Amazon presents a good case study still. Amazon offers a selection of products that it thinks you should buy based on what you have bought in the past. I use Amazon regularly, not just for me, but for other people that I want to get presents for. Personalisation in this instance therefore doesn't seem to do much good. For instance, today's top recommendation for me is Harry Potter 5 - on the basis that a year ago or so I bought my goddaughter the previous 4. No use to me - or at least, not at the top of the list of things I am looking for. The site gets me interested though about 2 times out of 10 - and that might be enough to trigger one purchase that I otherwise would not have made. How much does it cost Amazon to do that? I have no idea, but I can imagine that it is no longer a core area of where Amazon spend money on development. To look at another case study, how about banking? Personalisation makes a lot of sense here - how could it be any other way? After all, you only want to look at your own bank account; you likely want to make payments to the same set of people over and over again (family, the electricity co, the gas company, your stock broker etc); you might trade stocks or use a credit card from the same bank ... and that's all options you'd want to see on your start up page surely? But what of personalisation in the public sector? How about the in the offline world first? Thirty or more years ago when council houses were being built all over the UK, they were kept deliberately identical, right down to the colour of the paint on the front door. This made it simple for the council to keep them in order (put aside your political judgements on whether they did or didn't do that) - bulk purchase of paint, workmen who could move from house to house and fix things without having to worry about what had changed etc. The day that people bought their own house (starting in the early 80s I think) from the council, I bet the first thing that they did was to paint the front door - a touch of personalisation. So people want things "their own way". Today it's mobile phone ringtones, mobile phone cases and whatnot. Some degree of personalisation differentiates you from everyone else. Just as when you buy a car, unless you're in a very select market, you know that it's mass produced and rolls off an assembly line once every 15 minutes or whatever, but you also know that you've tuned the car the way you want it: a particular stereo, 18" alloy wheels, a metallic paint etc. So why shouldn't it be any different in the web world? Why should things on every page look just the way the designer wants them to, no matter how inconvenient that is for me? What if I visited a government website and I'd just paid my tax for the year? Would I want a pop up reminder saying that it was time to pay tax? No. If I had just registered for child benefit, wouldn't I want the website to talk to me about the new Child Trust Fund that was launching in a while? And, likewise, if I'd never registered for child benefit and had never looked at a page that talked about children, would I ever want to know that it was even possible to register online? I think the case for personalisation is clear. The issue is that it is only a step to take once the basics are done - once you have a highly navigable site that is content rich and has the information that people need - then, and only then, can we talk personalisation ... I'm not sure we're ready for that yet. I'm certain, though, that if it costs 4 times as much to implement and operate such a site, then we need to be absolutely sure that we're ready and that we know how it's going to be done. Oh, and then we'd do it on one site only - one that had all of the content and was proven capable of attracting high volumes of traffic because then we'd be able to tune it over many iterations to ensure that the right kind of personalisation was applied ... so as not to confuse or overwhelm people. Lastly, because governments are who they are, careful thought must be applied to what kind of data is used to do the personalisation - that is, what items, where they come from and what conclusions are drawn from those items. I read somewhere, can't remember where, that on a list of 10 companies, government's trust rating is somewhere near the bottom ... and right up the top was Tesco (or Sainsbury's or some such). So early on, personalisation is driven by what people do whilst on the site, subtly, in the background perhaps, coupled with the "people who looked at that also looked at this" algorithms. Once trust is obtained, we can ask for data and, in the longer term, seek data from elsewhere that will help the experience - for instance, if we know that Child Benefit has been applied for, but not Child Tax Credit, wouldn't it make sense to use that knowledge to make the claim? If you were given the right to anyway. So much done, so much still to do.
Posted by Alan at Sunday, November 09, 2003