Analytics news and expert advice every Thursday.
The Real Value Of Web Analytics
OK, I’m going to go out on a limb here and say something that might not make sense at first: the most important outcome of successful web analytics (or SEO effort or landing page testing, etc.) is not a better web site. The most important outcome is a better, more functional company.
Let’s pause from the execution side of analytics this week to focus on something a lot more important. Yes, web analytics can identify a host of issues with usability, marketing, technology, information architecture, etc., and fixing these things leads to improvement in volume of visits, exposure to new visitors, increased conversion rates, increased revenue, increased profits, and on and on. These are wonderful, almost magical things. But they all completely pale in comparison to what small improvements in operational efficiency can mean to an organization. In simple terms, it is stupendous to know what your business needs to do next, but it is so much more important to take notes when your company tries (and sometimes struggles) to get it done.
Now I should probably take a small step back and say that the impact of this idea scales with the size and complexity of an organization. To say that the best outcome of web analytics on a blog is an increase in operational efficiency is crazy. Unless, of course, the writer frequently gets into arguments with himself and defends his own bad decisions. But beyond a very small operation, nearly every company doing business online has significant operational issues, and thus, significant dampening to their upside.
Think about analytics in your own company for a moment. Let’s say you’re a crack analyst (or SEO, or paid search marketer, or whatever) and you figure out a few things you think will really move the needle, or that are broken and are harming conversion. It is not unheard of for an update to a home page, a change to a form, a new small feature release, or even a repair of a broken image or link to take two weeks, a month, a quarter, even six months. And let’s say that the upside of fixing one of these things is a +0.1% increase in your conversion rate (which, let’s say, is already 2%). The day you fix the issue, you are making 5% more per visit. That is huge. But what sucks is that you didn’t have that 5% for six whole months. You might not realize it, but that may have cost you millions (or scale to something that sounds scary to your own business).
The problem with the operational issues a company experiences is that they limit literally everything, and they are almost all preventable or can be significantly improved. They don’t just hold back that 5% upside. They hold back the next project and its upside. The backlog of work means hiring more people, which means greater overhead, more process, and more complexity. And the endless meetings to bring everyone and their dog into each decision means that people aren’t free to think and innovate around new opportunities. Whatever Linchpins you do have are snapped.
What you can do
The approach depends largely on who the “you” is, but here’s the gist of it:
The first step is admitting you have a problem (don’t worry, this isn’t a 12-step program). Without placing blame, the organization has to realize that it can be a whole lot better than it is today.
The second step is creating overlap. On the web, specialization can be a big problem when web sites are built on assembly lines (product to UX to design to IT…) and people protect the borders of their particular discipline. People are going to have to learn to trust one another so small issues can be corrected without assembling the whole village. And to build that trust, people need to learn how to do a few things beyond what they’re doing today.
There are plenty of .NET developers in the world who can make something pretty. There are plenty of usability designers who can write good PHP. Overly-specialized people actually work against each other because they can ignore the context of their work. If you have a company full of single-purpose employees, you may need to rethink your strategy. And this overlap allows you to…
Get small (or act small). As small as possible. Break up into groups where you can go from brainstorm to prototype in a couple days. Let the IT guy draw a picture. Let the design guy ask questions about the code. For some organizations, this is going to be very hard, but I’ve seen big companies do it and it’s amazing.
Remember that there is such a thing as too fast. The student who does best wrote her essay a week before it was due and had time to think about it. The one who did worst was the one who stayed up writing all night before it was due. But the good student’s paper would have sucked just as much if she turned in her work immediately without that week of thought. Get done quickly, but let things marinate before you make them final.
So use your analytics, SEO, PPC and other awesome skills to come up with ways to make your site a conversion machine, but pay close attention to why it takes you so long to get there. Will the above approach work for everyone? Probably not. But the first of your competitors who gets it done is going to mop the floors with you. History has shown us that companies that can be efficient and work in concert make a mockery of those who can’t.
More ideas on taking this approach and getting well-rounded:
I’d love to link to something Seth Godin wrote, but he focuses a lot on how being a generalist isn’t a good thing. I don’t believe that he always means that literally (I think he would endorse this approach), so if you know him, ask him!
Got any good additional suggestions? Post a link to something good in the comments.
Some opinions expressed in this article may be those of a guest author and not necessarily Search Engine Land. Staff authors are listed here.