Evidence-based website management
The Web is the perfect environment in which to make management decisions based on evidence and facts, rather than emotion and opinion.
The educated guess and the gut instinct are-like the action hero-vastly overrated. In human psychology there is a compelling attraction to the idea that the gut instinct is the seat of wisdom.
Listening to your heart and your gut may be nice romantic concepts, but for good decision-making the world is moving inexorably towards cool logic and hard evidence.
“Increasingly, doctors seeking to provide their patients with the best possible care are exploring what is known as evidence-based medicine–a hard, cold, empirical look at what works, what doesn’t and how to distinguish between the two,” wrote Christine Gorman in TIME in February 2007.
“Few people deny that the trend in medicine is increasingly to be guided, if not governed, by the data–an idea that is spreading to other fields as well,” Gorman continued. “Evidence-based practice is now being taught in nursing, general education and even philanthropy, thanks to the influence of the Bill and Melinda Gates Foundation, a results-based group if ever there was one. You could see even the political fights over global warming as the birth pangs of the new practice of evidence-based policy.”
“If the editors of a magazine-NEWSWEEK, for instance-want to know what interests their readers, their resources are limited,” wrote Jerry Adler in September 2007. “They can count cover sales, but that only tells them about one story a week. They can convene a focus group, but that’s a cumbersome and costly way to assess the tastes of 3 million subscribers.
“Online, by contrast, that information is available for the asking-not just the numbers of readers, but how long they spent with a given story and what else they read. So as journalism increasingly migrates to the Web, the job of figuring out what readers want becomes almost automatic-thereby raising the question, how much do we really need editors, anyway?”
The Web is the greatest ecosystem of content that the world has ever known. We can, with increasing precision, know what content gets people to act and what content doesn’t. The length of time people spend on the page is just a basic measure. Here are some others:
- How many links have there been to the content. This is the ultimate measure as the link is the gold standard of the Web.
- Where did the customer go once they read the content? Did they have a positive or negative reaction?
- How has the content been rated by customers?
- Has the content been blogged about? Did it get a conversation going?
Opinion, emotion and gut instinct are dangerous things when it comes to managing websites. Invariably they lead to creating websites that are organization-centric and full of vanity publishing. These sorts of decisions are compounded further when senior managers get involved, who often have no deep experience of Web, thus making their opinions even more likely to be wrong.
We need evidence that clearly shows how content helps customers complete common tasks. The cry of web management must be: ‘Show me the data’.

Nick Usborne says:
Added on September 23rd, 2007 at 5:07 pmGerry, hi
Your advice is sound, as always.
But I am still a firm advocate of common-sense and intuition.
Smart business decisions are based on data, metrics etc.
Remarkable business decisions are made from the gut.
When I see metrics-based decisions being made, and process-driven sites being created, I see a growth in the sameness and mediocrity of everything.
When I see something transformational, I know it was derived from an intuitive decision…not from some safe, metrics-based strategy.
Show me the data? Sure. Bore me to death.
BTW - When I went shopping this morning I parked my Toyota Corolla next to a new Volvo. When I cam out with my shopping cart, I put my key in the trunk of the Volvo by mistake. How come? Same color, same shape. Metric-driven mediocrity.
Best wishes,
Nick
Victoria says:
Added on September 23rd, 2007 at 5:59 pmI so thoroughly disagree, although I see the point being made and I think the idea of all sorts of fragments building up to evidence and the role of web 2.0 in that as perfectly valid.
Lets take a different situation - the outdoor ladies pond at Hampstead Heath in London by way of comparable illustration. This is, without doubt, the most beautiful place in London for women, a sanctuary open to all, in which a few hardcore ladies are willing to swim all year round. The Corporation of London which maintains the ponds has tried, over the years to make a case for value for money based on socalled evidence. Value being bums in water. But this entirely misses the point of the role of the ponds as a dream, even for those who don’t swim every day or through cold snaps; as a place to be social for older ladies who might otherwise be isolated at home through the winter; as a place of celebration and community, even if it’s shuddering past someone and giving them a comradely glance as you totter on frozen feet back to the changing rooms and they stride, flipflopped, to the steps.
There’s no evidence here - just a special transformational experience open to all which a lucy few are able and willing to take the plunge in.
And I feel the same about so called evidence on the web. I’ve written papers myself and blogged, and borrowed a little in the process - I’ve certainly been quite superficial in references I’ve borrowed at times, not read the whole book, summarised inaccurately. And these borrowings and opinions, however sloppy, quickly get borrowed in their turn - there’s a mix of hard challenge, supercilious challenge and superficial borrowing that goes on that can propogate half-baked things which shouldn’t constitute evidence. I do it too. I’m not setting myself above this. There’s neither enough considered challenge, nor enough disicipline about what gets borrowed and linked and how for me yet to consider this an evidence-based process. It could be. But it’s not yet.
Graham says:
Added on September 24th, 2007 at 8:33 amI noted the drive toward evidence based publishing and this is to be applauded for businesses, the people the article is targetted at. I note also Victoria’s comments and think that fortunately the entry point to publishing the web is very low so it can support both approaches and many more simultaneously. What struck me most however was that I took the survey on car buyimng carewords to look at this carework idea and I was horrifed.
The survey seems to have misssed the last thirty years of market research. How can I possibly pick five words from a list of what 50 alphbetically arranged phrases or words. If they were grouped and presented by area - looks, perfomance, cost etc I might have stood a chance. As it was I registered a protest by just rating the first five on the first page then the first three on the second page.
Unless this is a cunning plan to show people how not to do things it is a poor reflection on Gerry’s editorial skills to allow his column and website to be linked to such a dreadful parody of a survey.
Interestingly it is a classic example of a single thing threatening the integrity of a brand - the very thing an editor must preserve……
Alan Charlesworth says:
Added on September 24th, 2007 at 8:57 amHey Gerry / Nick - you are my most read, most quoted/quotable pair when it comes to web content : no bickering please.
Surely there is a before and after aspect to this?
Nick – totally agree, some of the best sites are those written from the heart, I can even live with [just] bad spelling/grammar when the writer is giving over their experience/enthusiasm/knowledge of the subject - as is often the case in small business [isn’t that their USP over big business].
That said, there is no reason why that content should not be monitored to see if is working - as suggested by Gerry. And from a marketing stand-point, the ultimate answer lies in what individual users are looking for to meet their needs.
My example of this would be in sport. I’m in the UK so I’ll use football [soccer]. The club web sites are [normally] written by an employee. They are factual, well presented, well written. But they are ‘dry’ – a lack of soul. The ‘fanzine’ type site, on the other hand, is written by a fan. Not professionally trained, not part of the club’s corporate-communications team – but they put their heart, their opinion, their belief into their content. And of course, like most supporters, if I want team news, players’ injuries, re-arranged fixtures etc – I go to the official site. If I want an opinion on who is playing well, who had a stinker on Saturday, and which new players we should sign – I go to my favourite supporter’s blog.
But then the blogger isn’t looking to meet any marketing objectives or raise income through advertising.
Obviously, this is not an issue that can be fully addressed in a blog entry and a couple of replies.
Gerry McGovern (blog author) says:
Added on September 24th, 2007 at 11:07 amGood points. (Hello Nick; long time no speak–hope all is well.) Alan you write about the football fanzines and official sites, and you’re absolutely right. But now if we look at the BBC where you see a journalist’s story and then see it rated by and commented on by the fans, you see evidence of whether the journalist is actually connecting.
And, of course, the bloggers are out to get linked and get comments because that’s great evidence that they’re respected and getting noticed. If you’re a blogger and for the last six months nobody has linked to you and nobody has commented, and you only have 2 visitors a day, that says something.
I know of major media sites now that feed loads of short pieces to the Web. If they get good comments, passed on, read a lot, then the editor will develop the story further and it will be a major candidate for the print version. I know other media sites that take minute-by-minute feeds from the major search engines, thus knowing what the hot topics are.
I think the problem with the gut is that it’s often organization and me-centric. By building up hard evidence of what the customer really wants, and what they really connect with, and by delivering that in a passionate way, we get the best of both worlds.
Gerry McGovern (blog author) says:
Added on September 24th, 2007 at 11:13 amGraham,
Sorry you didn’t like the survey. It’s been tested over a six-year period and it works very well. I have numerous clients who have changed their web strategy and content because of the results, and have seen substantial, measurable results.
It is deliberately meant to somewhat overpower, so that the customer will choose quickly and go with their gut instinct. By the way, with this “Buying a car” survey, 4 tasks are getting a massive 29 percent of the vote out of 88 tasks in total. And no, they are by no means the first four in the list. (Your entry was deleted because it was corrupting the data.)
Alan Charlesworth says:
Added on September 24th, 2007 at 12:30 pmGerry – re football, spot on: I also read the reports from ‘independent’ sources - BBC, newspapers etc - for a view that falls someway between the ‘official’ and the ‘unofficial’.
Re ‘organization and me-centric’ – is this not a wider issue of CEOs/MDs not understanding, or valuing, this thing we call ‘marketing’?
Hey … I teach the subject in a university business school and half my [non-marketing] academic colleagues think that marketing is a non-subject, or simply a very small element of what makes up business management [and research in industry backs up that this view is common-place at the top of many organizations].
I’ll bet you don’t have me-centric discussions [arguments] with clients who run market-oriented organizations – they will already get it.
Sadly, that is part of a much bigger picture – and way out of the scope of this particular blog.
Nick Usborne says:
Added on September 24th, 2007 at 12:31 pmCharles, don’t worry. Gerry and I don’t bicker, and we agree on 99% of issues. Victoria, I love your use of the Hampstead Heath Ponds as an example. And I agree about the “borrowed” evidence.
Let me say a couple more things about evidence.
Recently, over the course of 12 months or so, I had the opportunity to immerse myself in a vast quantity of data and metrics, all drawn from online testing. Some fascinating stuff, much of it useful. But the other thing I discovered was that while the test data was sound, people’s analysis and interpretation of the data was not. I saw huge, unwarranted assumptions being made, time and time again. What test data actually “proves” is usually very, vary narrow. However, the publishers of this data are often tempted to make assumptions and broaden the scope of their findings, way beyond what the test actually demonstrates.
This “evidence” is then distributed across the web and taken as gospel. It is quoted and requoted. And all too often, it’s nonsense.
The more I speak to those who conduct testing and research online, the more I hear about their growing misgivings about the true validity of much of what they publish.
This is a problem, because marketers are lazy. Instead of thinking for themselves, they grab the latest published “evidence” and use it to support their next marketing presentation to their managers.
To create exceptional products, services and web sites you need a little more courage than that.
By all means, gather together all the evidence you can. But be critical of it. Don’t assume it is all accurate. Do your own thinking.
And yes, listen to your gut.
Nick
Gerry McGovern (blog author) says:
Added on September 24th, 2007 at 12:37 pmAlan, it’s interesting, isn’t it–marketing and content are often seen as minor aspects of whatever needs to be done. And I think one reason is because of lack of evidence of measurable results. It’s a really hard nut to crack this evidence thing, but here’s an example.
I worked with one B2B client that sold very expensive products. At the bottom of their product pages they had a link such like:
“For more information contact us”
They changed the link to:
“Request a demo”
“Speak to an engineer”
Qualified enquiries from the website rose from 100 a month to 250 a month. That’s pretty powerful evidence.
Gerry McGovern (blog author) says:
Added on September 24th, 2007 at 12:44 pmNick, Excellent point, and of course data/evidence in the wrong hands can be used to prove anything. My problem with the gut is that it is often organization-centric.
I often give a test of words to customers and then give the same test to product/marketing managers. What the marketing managers think is really important to the customers and what the customers actually vote for is often not very well aligned.
Gut instinct is great once it is well trained. I think it was Jack Welch who said that gut instinct was the results of constant and repeated actions.
Nick Usborne says:
Added on September 24th, 2007 at 1:03 pmGerry, you’re a 100% right.
And I know exactly what you mean when a “gut” decision is simply a safe, organization-centric decision.
But that’s not the kind of “gut” decision I’m talking about. I’m not talking about the floppy, limp, lazy approach, with a complete absence of critical thinking.
What I favour is a much more disciplined, courageous approach, which assesses the evidence, but also listens carefully to the voice of intuition.
Regarding being company-centric or visitor centric…that continues to be a struggle. And yes, there is a mountain of evidence to show that the visitor-centric approach delivers better results.
So I’ll say it again, for the thousandth time:
**The language of the Web is the language of individuals, not that of corporations.**
Nick
Bob Johnson says:
Added on September 24th, 2007 at 1:24 pmA debate that has no end… Of course, interpretation of data is an essential element after its been collected and in that interpretation, “gut” instinct will play a role since the data does not always give a definitive path to follow.
That said, what scares me are people who are so self-confident that they know the right path that they feel no need to look at any data at all… particularly anything that might conflict with long-held cherished beliefs.
The best people to work with are those with a reasonably open mind, who do want to see the data before making their next marketing move, including how to present website content to various audiences. I’m convinced that a rather large number of people don’t like data because the reality it often introduces makes decision-making more of a challenge and sometimes contradicts earlier decisions.
Not long ago I was working on a Carewords project with a university client that had just removed the words “distance learning” from the website from a belief that those words were academic jargon not used by the public… and replaced them with “online degree programs.” A quick use of Wordtracker showed that while “online degree programs” was often used in searches, “distance learning” was used more than three times as often. And when the client’s potential students completed the Carewords survey, “distance learning” was at the top of the response list.
The conclusion… use both terms in the marketing effort. A conclusion that, without the Wordtracker and survey evidence, might never have been taken.
Data can’t replace informed judgement. But making decisions without data really is pretty silly, particularly when the data, as in a Carewords survey, isn’t that difficult to get.
Jennifer Lyle says:
Added on September 24th, 2007 at 1:26 pmFascinating, fascinating. I recommend that we get away from the either/or nature of the discussion because,like life, marketing is not an either/or topic. Good marketing rquires good judgment, which is a combination of educated instinct (known as gut)and a highly cultivated undrstanding of the use of numbers (so quickly they can betray us if we’re not careful, as a couple of of writers have mentioned). Both are needed. One is no good without the other.
I view Web marketing as the ultimate ability to carry out proper direct marketing techniques. Proper direct marketing techniques are based on the “simple” scientific method that we all learned in high school (or the equivalent, depending on which country we live in): continual testing that changes only one variable at a time. Sounds simple, but isn’t.
General marketing receives short shrift in the boardroom & direct marketing receives even less. Turn to an experienced direct marketer to create your Web programs; read the literature by the great direct marketers. You’ll be amazed. Nope, I’m no longer a member of the Direct Marketing Association. This is just what I believe.
Thanks for giving me something to think about on a Monday morning.
Alan Charlesworth says:
Added on September 24th, 2007 at 2:07 pmNick : re ‘marketers are lazy’ – I’m sure you are right, but I would add that they are too lazy to learn the ‘science’ of interpreting data. I’ll go further and suggest it is a forgotten skill in marketing education – often because it involves an element of maths and that is seen difficult by students so they either opt out of it or fail it. Sadly such is the nature of the funding of UK Higher Education that subjects that produce ‘fails’ are withdrawn because fail=no funding [and btw, don’t worry, you’re not the first to call me Charles!].
Bob : as the ‘distance learning’ story is close to home, it’s a great example in more than one way.
Jennifer : re ‘great direct marketers’ – Claude Hopkins’ Scientific Advertising is compulsory reading on my e-marketing modules. Ignore the blatant sexism and the brands/organizations that are long gone, and I think the book is a guide to Internet marketing.
Now, I really must go and convert some more business students to the importance of good marketing.
Andy Baxter says:
Added on September 27th, 2007 at 9:34 amInteresting two snippets from this exchange:
“We need evidence that clearly shows how content helps customers complete common tasks. The cry of web management must be: ‘Show me the data’.”
and
“(Your entry was deleted because it was corrupting the data.)”
How do you know the data you’re seeing hasn’t been messed about with? On what basis do you recognise the corrupting data to delete? Gut-instinct? Deviation from the formulaic?
Gerry McGovern (blog author) says:
Added on September 29th, 2007 at 7:08 pmAndy, good question about the data. Of course, we need quality data. However, the main point I was trying to get across is that the Web gives us this wonderful environment in which we can test what the customer thinks, what they respond to, what they really like.
Often what we think the customer wants and what the customer actually wants can be two very different things. The Web allows us to get the data that clearly shows what is getting a response and what isn’t.
Neil McPherson says:
Added on October 2nd, 2007 at 8:06 amOne of the scariest details on my daily Web site stats report is the size of the daily crowd of 0-30 secs. viewers. When it comes down to it, changing that statistic is what the offerings of Gerry and Nick (especially) are about, I believe. But is that all there is?
“Creativity is the defeat of habit by originality,” Arthur Koestler is supposed to have said. according to Alan Fletcher who quoted Koestler in his own superb book: The Art of Looking Sideways (Phaidon, p.99).
Koestler’s observation, when applied to the discussion here over recent days, reminds me of the phrase “a map to fit the territory”. It was popular in management theory during the 1990’s. Both Gerry and Nick appear caught on the flypaper of what is nowadays dished up as necessity by the major search engine management: necessity to shoot for a new (lower) kind of common denominator.
One web writer counselled me recently: “While you and I might love language correctly used and think it’s appalling that people of average and above intelligence feel compelled to use words like, for instance, killer as an adjective to describe copy and customize (with a zed) when “modify” might suffice. It’s because the major search engines rate those expressions highly. They are killer words. So we must look the other way and pass on. Oh, and exclamation marks are now OK!”
The crusaders for “Keywords” and ”Carewords” have dug trenches in the heart of creativity on the web.
Nick and Gerry are convinced about their approach to site copy writing. There is evidence for both ways and for the overlap -not that any of it is backed by the sort of provenance that some might hope for. But we do see evidence of a formidable collection of experienced (maybe even hot-housed) observations, opinions and surmises.
What they (these ever-so-slightly priestly figures both) do for me is to provide me with “a map to fit the territory”. Nothing more. Admittedly, that certainly helps untangle the mess that stands for “the wonder of the web” for now. For those small mercies they should be thanked. But not too much! In the meantime it appears that it is up to a new generation - a dedicated priesthood of writers - to take on the venerable task of the ancient monk scribes - and set about saving the culture.