Guest post: Don’t leave data on the table

Without web traffic data, most online marketers feel like they are flying blind. Tools like Google Analytics make it fast and convenient to monitor page views, goals, and even information on ad spend and customer acquisition. However, if you stop there, you’re not flying blind, but you are keeping one eye closed.

Too many businesses, especially small businesses, overlook the most valuable source of information they have: the back-end database. Virtually every ecommerce site, social network, SaaS web application, widget, ad network, online game, and most other websites have a database. It is the workhorse that stores messages, logins, profiles, transactions, relationships, behavior, preferences, and everything else that is necessary to run an online business. The database is the historical record, for every user and customer you have ever had.

If you are not yet convinced that the database is a critical tool for online marketers, then think about this question: What is the maximum you would pay to acquire a customer? Your web traffic analytics tool might tell you what you are paying now, or what you have paid in the past, but it cannot address the question of the most you should be willing to pay. The answer to that question is your expected customer lifetime value (CLV). There is more than one way to calculate CLV, but each requires an analysis of the historical record of your customers’ activity.

Sure the database stores all this great information, but what do you do with it? The number of analyses you can run is endless, and they vary based on business model and data set. Some of the more interesting ones include customer lifetime value, repeat purchase probability and cohort analysis. All these metrics can be segmented by referral source, initial activities (i.e. characteristics of first the transaction or free trial), geography, and demographic information. The number of potential permutations can seem overwhelming, but you can narrow down the analyses if you think carefully about the questions you want to answer. [thinks editor’s note: For more information on how to do that, check out these 7 keys to successful web metrics for one approach and our series on how to build the right metric for your marketing for another. Back to you, Jake.]

So how do you actually get the data and run the analysis? If you or a member of your team knows SQL (Structured Query Language, the lingua franca of databases), then you can use it to extract data for analysis. From there, most people use a spreadsheet like Excel to do the nitty-gritty work. However, if you have a large data set, want your analyses to be continually updated, or just don’t want to commit the time, energy, and expertise to do it in house, then it might make sense to use a specialized tool to extract and analyze the data. After all, you’re not tracking your page views by reading server logs, are you?

Here are five tips that will help you get the most out of your database:

1. Store the source: When a new user or customer registers on your website, store the associated referrer, coupon code, and/or campaign in your database. This allows you to calculate how much revenue, including repeat purchases, is generated by different sources. Plus, it’s easy to add this to the database (our site, www.rjmetrics.com, does it with less than 10 lines of php code).
2. Don’t overwrite data: It’s tempting to store information that changes often in a table that tracks the current state of the world (is the user logged in right now? Yes or No) but you lose data on historical activity if you don’t keep a record of each action. As Yogi Berra could have said, “Once you lose data, it’s gone.”
3. Don’t slow down your site: If you are going to run complicated database queries, do it on a backup or dedicated statistics server so you don’t slow down your site. Servers are easy to replace, disappointed customers are not.
4. Use your database for deeper testing: Readers of thinks know that testing is important for optimizing online performance, but tests can be about much more than “Did they buy today?” The most valuable customers are the ones that come back again and again, and you can use your database to analyze who comes back and why.
5. Some analysis is better than none: If you’re stuck on which database analytics are right for your business, don’t sweat it. Start with something simple, like segmenting your sales data by a category or two and go from there. Even if you are not yet a database pro, a little bit of insight is better than none at all.

Are you getting enough value out of your small business website? Want to make sure your business makes the most of the local, mobile, social web? thinks helps you understand how to grow your business via the web, every day. Get more than just news. Get understanding. Add thinks to your feed reader today.

Or subscribe via email.

And while you’re at it, don’t forget to follow Tim on Twitter.

Image credit: quinn.anya via Flickr using Creative Commons Attribution-Share Alike 2.0 Generic.

Technorati Tags:
analytics, bounce rate, douglas hubbard, how to measure anything, measurement, metrics, performance, skills, solutions, the numerati