Actionable Web Analytics, by Jason Burby and Shane Atchison
Book review by Andreas Ramos
Actionable Web Analytics, by Jason Burby and Shane Atchison (Wiley, 2007. 256 pg., illustrations, tables, index. $29.95. ISBN 978-0-470-12474-1). Available at Amazon, Borders, Barnes & Noble, etc.
Jason Burby is the Chief Analytics Officer at ZAAZ and Shane Atchison is the CEO of ZAAZ. They use analytics to build better websites for corporate customers.
A practical book on how to use analytics to get KPIs, measure what you need to know, and how to implement the information as business decisions. Well-written book by two guys who actually do this for a living.
The authors start by saying the book is for marketing. It helps marketers to improve sales, profits, and ROI. It is not a how-to book, with step-by-step explanations. Instead, the book shows you why you should use analytics. You learn how to look at websites as customer experiences and use analytics to measure that experience.
Chapter five states the four steps of analytics:
1) Establish your KPIs. When you build an ecommerce project, you have specific goals in mind: leads, subscriptions, sales, and so on. These goals must be converted into business goals, namely, you need to monetize the goals. How will something make money? By converting a goal into dollars, you can measure it. These business numbers are your Key Performance Indicators (KPIs).
2) Track your KPIs. Create the reports that let you measure the KPIs. If a table of numbers isn't a KPI, then it is not a report. It is just clutter. Figure out your KPIs and build reports to measure them.
3) Analyze your KPIs. Understand what is happening in your trends. Why is a trend going up? Why did it dip in July? What are you doing that is causing the trend to move? Instead of reports ("we sold $320,000 on the site in September"), give an explanation ("Because we introduced the new product, sales increased 24% from Q2 to Q3.")
4) Take action on your KPIs. Based on your KPI data, you can make actionable business decisions. "Our new organic apple juice produced $400,000 in April profits in NYC. Let's release it in Chicago." "We discovered that 32% of our visitors are from California. Let's roll out a new product line for the California market." "Although we spent $200,000 in Google Adwords in October, we only got $40,000 in profits. The ROI was 20%. We spent $50,000 in Yahoo and got $35,000 in profits. The ROI was 70%. Let's shift budget from Google to Yahoo."
This is the heart of analytics: it's not about reports. It's not about numbers. It's not pretty charts. Analytics is roll up your shirt sleeves, figure out what is going on, measure it, understand it, and improve your business. Analytics works.
I've been working with the web since it started. In the first few years, you could build just about anything and it made money because there were only a few websites. By 1999, you had to optimize for search engines, so SEO started. In 2002, PPC was added. From 2002 to 2006, our clients asked for SEO and PPC, but they didn't know about analytics. Analytics was an add-on to SEO and PPC. There wasn't even a trade show for the analytics industry. By early 2007, this flipped around and analytics has become the main issue. SEO and PPC, just like banner ads, email blasts, and other tools are ways of driving traffic into the analytics tool. Analytics unifies the entire process and turns it from a black art into profits. This is why analytics has become the core of the web industry.
For example, we establish that user conversion is a trackable KPI. Let's say we're getting a 32% conversion rate. What is the benefit of increasing the rate to 33%? 34%? On page 101, the authors show you how to create talbles to show the expected profits and ROI for each percentage point increase in a KPI. If you make this change and it improves the KPI by 4%, you can calculate the increase in sales and use that to justify the action. And even better, they show you how to calculate the cost of a delay in implementation of a change. By putting off the change for three months, you can also calculate the loss in profits.
There is also the reality of implementing an analytics project at a company. It's all nice to talk about theory, but if you've tried to set up an analytics project at a large company, there is a great deal of politics. The larger the company, the worse the politics. How do you convince six VPs to take action on an item? In chapter ten, Burby and Atchison show you how to prioritize projects according to the profitability of each project. Do the item first that will bring in the most money. This is an excellent chapter and will be very useful for you.
Throughout the book, Burby and Atchison apply analytics to the four types of websites (ecommerce, lead gen, content sites, and customer support sites) and how it can be used for each type.
The Bottom Line
Actionable Web Analytics has a wealth of practical information and ideas. It's a good investment. Get the book.