# What do small businesses need from web analytics?

If you read this blog regularly, you may know that my wife runs a online retail business in the UK called mirrormirror (since Valentine’s day is coming up, and I know that the overwhelming majority of my readers are male, this may be a good time to go over there and earn some brownie points with your loved one; but that’s not the point of this post). I help with various aspects of the business in my free time – specifically, anything involving numbers or HTML tags.

One of the reasons I’m happy to help out (as I sit hacking away at HTML in the wee small hours) is because mirrormirror is slap-bang in the middle of the target market for our ‘Gatineau’ web analytics service. So whenever I’m thinking about whether a product feature would be useful for our audience, I think not only, “would mirrormirror benefit from this?” but also “could mirrormirror benefit from this?”

The latter question is a very useful filter. On several occasions since arriving here, I’ve had conversations along the lines of:

Me: “So how are our customers going to instrument their sites?”
Interlocutor: “They’ll get their IT department to do it.”
Me: “These people don’t have IT departments, dummy.”

So I thought I’d share some of the rules-of-thumb that I use when thinking about small businesses (I used to work for one as my day-job, too, so I feel moderately qualified to generalize) that are guiding my thinking about the feature set for Gatineau.

1. No money
Small businesses have no money. Well, ok, maybe they do have some money, but it’s already spoken for several times over. Spending money on measuring stuff comes at the bottom of a long list which is headed by items like paying the rent, employees’ salaries, paying other suppliers of absolutely essential stuff (e.g. stock), taxes, and so on. Towards the bottom of the list are ‘marketing’ and ‘making changes to the website’. Measuring the effectiveness of the marketing, or the changes to the website, comes even lower.

That’s why, of course, a free web analytics tool is so appealing. But actually, having no money is the least of our worries. No, the real problem for small businesses is no time.

2. No time
Small businesses have no time. It’s basically the same problem as money – what time there is is carved up between a number of important things like finding customers, selling them stuff, delivering the stuff to them, and doing admin.

But time is more difficult to carve up than money. If you say to a small business owner, “this will only cost you $50 a month”, they may be able to slice off a sliver of their earnings for that month and give you$50. But it’s not so simple to say, “this will only take 5 minutes of your time a month”, because you can’t carve time up that small, unless the thing you’re asking the person to do for 5 minutes a month is so darn simple that they can remember how to do it in the intervening 30 days, 23 hours and 55 minutes.

Unfortunately, web analytics doesn’t fall into this category of tasks – it doesn’t even come close. So actually, having no time is not the real problem. No, the real problem is no expertise.

3. No expertise
My wife is (as I am fond of reminding her) extremely lucky – she lives with one of the world’s foremost web analytics authorities, so she benefits from a level of expertise in this field that others can only dream of. But most small businesses are not that lucky. It’s very hard to maintain perspective on just how hard (or at least, unfamiliar) web analytics is to someone who doesn’t think about it every day.

The lack of time is, of course, a root cause of the lack of expertise, but another thing that we data geeks need to bear in mind is that there are some people in the world who (whisper it) don’t find this stuff as fascinating as we do. So whilst my wife is keen to learn how people clicked through the latest e-mail campaign (and will prod me with a stick until I laboriously add all the relevant tracking tags to the e-mail HTML), she’s less interested in the growth of Firefox vs Internet Explorer – at least, until it becomes apparent that there’s a Firefox-only bug with the site, when the fact that 20% of the site’s visitors are using it suddenly becomes terribly important.

So what does this mean for any aspiring multinational corporations who are hoping to give web analytics away for free to this market? To me, it means the following:

Keep it outcome-focused
Poverty-stricken, rushed-off-their-feet, stupid small business owners need technology to deliver them answers in a context that they can understand. So web analytics needs to mesh with the real-world concerns and needs that these people have. Top amongst these is understanding how marketing dollars could be better spent. Because so many small businesses now use paid search advertising, there’s a growing awareness of some of the ‘traffic’ concepts around this activity, such as impressions and clicks (though we should not assume too much). Web analytics for small businesses should grow out of this context, rather than provide their own unique context and expect small businesses to learn that context and, more importantly, remember the context month-to-month.

Make it easy
I don’t just mean the tool, here. I’m referring to the whole process of signing up, getting set up, looking at the data and drawing conclusions. Technology will help in some areas – for example, by automating instrumentation – but the rest is a people issue. Not only does the system have to be easy, it has to stay easy – you need to assume that the customer will climb the product learning curve almost every time they use the product. If that learning curve is steep, or even contains just one counter-intuitive step, the customer will get pissed off climbing it every time they sit down in front of the tool, and they’ll stop using it.

Provide people
“Web analytics needs people” is such an old saw now that I even hesitate to mention it, but in the context of small businesses it means access to expertise in other, approachable, organizations who specialize in web analytics but don’t say they specialize in web analytics – or at least, don’t define themselves this way. Small businesses (and many big ones) aren’t interested in web analytics – they’re interested in selling more stuff online, paying less for advertising, or selling more ads with their content. Web analytics is just a means to an end. The sorts of people that small businesses need access to are people who understand how to use web analytics to help their customer achieve their goals. These people need to be almost as suspicious about web analytics as the customers they’re serving, and understand that it only provides a piece of the picture.

Google has done a great job (whether by luck or by judgement, I don’t know) in building an ecosystem of small consultancies who offer services around its product. Given the enthusiastic response I’ve had from this community to the news of Gatineau’s impending availability, I’m hopeful we can do the same.

That’s probably enough ranting for one blog post. In future posts, I hope to give you some more concrete examples of the measurement challenges that we’ve faced at mirrormirror. If you deal regularly with small businesses for web analytics, I’d love to hear from you – pl

### 2 thoughts on “What do small businesses need from web analytics?”

1. I’m not sure that post was entirely complimentary.
And if I live with one of the world’s foremost web analytics authorities, why the heck aren’t we selling more stuff?
Love and kisses
xxxxxxxxx

2. Couldn’t agree more. I see plenty of sites where the owner has stuck Google Analytics on there because it’s free and then scarcely looks at it. They don’t know what to do with it, they don’t have time to learn and they don’t have the resources to pay someone else to do either of those.
My stock response is that Google did them a disservice when they named their product ‘analytics’ instead of ‘reporting’. They would be better off with something like Clicktracks because the interface has a go at pointing people in the direction of WHAT to look at, not just providing the tools and leaving them to work it out. Few take up the suggestion because it a) costs money [even if it’s not much] b) doesn’t have the brand recognition of Google.
Gatineau looks set to deal with those objections. But if what you’re hinting at in this post is true, and the interface itself can encourage sensible analysis and not just reporting, then you will be dealing with fundamental issue which affects users of the top-end solutions as well.
Tim