With analytics seen as the game changer both within the firm operations and support support functions as well as for sales and marketing, it is important that the PM get a good grasp of analytics as a subject and the associated tools. The Wikipedia entry for Analytics calls it “the discovery and communication of meaningful patterns in data.” Additionally, wikipedia entry for Google Analytics calls it “a service offered by Google that generates detailed statistics about the visits to a website.”
It should be clear now that analytics comprises of statistics, reports and patterns in data generated from different sources. Additionally, in the software world, analytics is also used to identify features of interest, and in some cases, especially for web products and apps, analytics actually becomes an important module (to track feature and product usage) to be developed in the product.
So what does a product manager need to know about analytics? Well, first of all he should understand the type of analysis that is useful to clients, senior management and engineering teams (web analytics, customer segmentation, product performance reports, product usage reports or something else). So here’s a glimpse of the analysis that is useful:
- Product performance/usage reports (how many people use the product, what segmentation is possible etc)
- Product sales reports (customer profiles, segmentation, geographies etc)
- Website usage reports (if its e-commerce, then products looked at, purchased etc, for other sites the browsing patterns, exit pages etc)
- Website experiments (A/B tests, multivariate testing and so on)
- Customer surveys and forum post analysis (to identify features of interest, trouble areas etc
Something that is clear here is that all this analysis must be performed periodically, at least once a quarter and must be an important input in future product planning. And this also means that product managers must a) gain expertise in statistical analysis and b) build a good rapport with the analysts or analytics team.
So how does one learn about these different analytics domains?
Here’s one suggested route.
- Start with a book or online tutorial on statistics and learn the fundamentals from that
- Download some free statistical software such as pspp (or use MS Excel) and go through the some hands on exercises available on the net there
- Add Google Analytics tracking code to your personal website or blog and start viewing the reports available. Then read the entire GA help documentation.
- Read some good books on website design to increase your knowledge of experiments on website usage
- Learn SQL, a vital tool for querying databases and getting aggregate results
- Finally, read articles, books and tutorials on Business Intelligence and Dashboards
The “resources” section of the blog has links to various useful software, tutorials and books. You can use them as a reference.
If you spend “An Hour A Day” on these every weekday, it will take you a good 3-4 months to get to the final part. After completing these, you should be in a position to talk meaningfully to the analytics teams, create useful performance reports from raw data, and support your feature specifications using live data.
All this will definitely turn you into a Data Driven Product Manager, which is an immensely vital skill for product management today.
Note: There is a strong relation between Big Data and Analytics, however, for most part it is not relevant unless you are either designing the big data infrastructure or you already have user access to the infrastructure.
In a future post, I will summarize the connection between product managers and analytics, given the 4 types of product managers in India.