Own web analytics for startups – Part I

Part I – Where to start, first thoughts aloud

“…, but I like the way you think”

From a joke

The internet is full of information about everything, but it isn’t very easy to find a ‘step by step’ story unfolding with time and solving some problem. It is understandable as when you are trying to achieve something you just do it and there is no time to record it. You don’t keep a log of what you’ve done. A log, which could become a source for really interesting blog post series or even a book, will not be created.

It’s good to know that “Time on page” is not at all what you thought it was. It often doesn’t solve your practical task and just acts as a piece of knowledge, a fact. I always liked stories, that starts by describing the problem and bring you to a logical point, and on the way, describing all of the nuances of the solution found and the mistakes that were made.

You rarely ever see log books, more often glossy magazines. A good example – the new book from 37signals ‘Rework’, it is full of good advice which can be put into practice for your own business, but this isn’t a book of development, this is a book of advice, even though advice based on real experience of authors. ‘Dreaming in code’ could have been a book like that if it was more descriptive, and it was written by an insider (a member of the team).
The only blog I can mention as an example, and which is thematically linked to our story, is Ash Mauryas’ blog. If you came across such insightful blogs or books, especially the ones related to internet and business please share them with us.

So, this series of articles is about web analytics – about our understanding and use of it for our product. You will find our thoughts, actions, decisions as well as achievements and from start to finish. If you start reading in August 2010, you have to be patient because this isn’t our most important task, but we couldn’t find an immediate solution to our problem so we will be solving it as it comes, leaving notes. We hope that this helps someone and we anxiously await your advice, counter-arguments and links.

Here are the examples of articles that are related to our topic, but as I already said – torn out of a glossy magazine. Yes we can get some good ideas, confirmation for our thoughts, but nothing more.

The first 6 steps to homegrowing basic startup analytics
Omg* I’m just a startup, I can’t do those fancy analytics!

Our story will begin a little detoured. You will see that jumping straight into web analytics won’t work – the narration will lose practicality.
Looking at different web analytics systems and reading blogs on this subject I found a book ‘Complete Web Monitoring’, which was a complete waste except for one thing. The second chapter is called:

# Chapter 2 What Business Are You In?
1. Media Sites
2. Transactional Sites
3. Collaboration Sites
4. Software-as-a-Service Applications

If you don’t have the book – the same information is located in the authors’ blog. The point is that there are a few types of web businesses, and to swim smoothly you have to understand differences. Starting from this healthy thought the book goes on to talk about everything in the world, probably to let the reader filter out what they need for their business. On the other hand, we will only be talking about analytics, and about only one type of business (see a table shown below), getting an understanding and applying it step-by-step.

So, after a sombre and unorganized acquaintance with web analytics systems and reading books and blogs, web businesses classification idea became an aha point where thoughts lost fuzziness and lined up near a starting point shown below.

In the columns we have the types of businesses, and in the rows we have the systems that we might need to let visitors make a conscious decision about becoming clients, and help existing clients – remain clients for as long as possible. Do we want to know how visitors and clients use these systems? Yes, of course. Will it be more comfortable to use a single analytics system for all these systems? Sure!

Now it will become clear why we took a little detour. It is crucial to understand what we want to analyze, which systems and how. So the question for now is – what and why, not how to get a heat-map or engagement time pages.
Don’t agree, or want to add? Feel free to let us know.

We are all old enough to understand that every categorisation is conditional. We are in the first column 100%, but if you’re unlucky create your own column and write us what is different about it. It is very easy to be in a few columns at once, SitePoint for example. It has a forum, articles, job announcements and books.

In short, about our categories.

A – web applications. Because this is a business that we ourselves are in, there will be more detailed explanation later.

B – a place for people to interact, as those systems primary use. Here you have forums, and social networks and communities like stackoverflow.com.

C – commerce and markets. The best examples are Amazon and eBay. Amazon became also a marketplace in addition to being an e-commerce place when they allow other traders to sell same item side by side with them.

D – news, reports, overviews, opinions, gossip and rumours.

Now we have everything ready for detailed ‘cell’ analysis in a next post.


Print this post | Home

Post a comment