Web Analytics: An Hour a Day

Avinash Kaushik focuses on the key issue: the visitors. Learn what your visitors want and then give it to them. Use analytics, competitive intelligence, and user polls to find out. Many sites focus only on customers (the ones who actually buy) and pretty much ignore the remaining visitors (who make up 60-80% of the traffic). By looking at visitors, you can improve the overall experience, and very likely the conversion rate as well.Avinash's book is far away the best book on analytics. He has solid experience in using analytics tools at large companies. He also has a degree in engineering and an MBA, so he understands both the technical and the business aspects. The book is a solid presentation of how to use analytics to establish and reach business goals. As you've figured out by now, analytics is not about tracking URLs or auto-generated reports. It's business.For Kaushik, there are three main strategies: discover the visitors' intentions, compare your website against benchmarks, and use analytics.The first item is to understand your visitors. Learn what they want. Find out what they are seeking at your website. It's a mistake to focus on conversions; at best only 20-30% of your visitors will convert. If you concentrate on them, you're ignoring 70% of your visitors. To learn their intentions, ask them. Add polls and surveys at your site. Avinash offers several questions to ask your visitors: What are you looking for at our website? Were you able to complete your task? If you were unable to complete your task, please explain why. How can we improve our website to make it more useful for you?The second item is comparison with your competitors: how are you doing in your industry. It's very nice to say that you have two million monthly visitors, but... compared to your top competitors, is that low? If they're getting one to two million visitors per month, then you're fine. But if they're getting 50 million monthly visitors, you're in trouble. So, find out. Avinash describes two services for competitive information and the features of each. You can find out your competitor's traffic share, level of activity, conversion rates, demographics, and so on. Other tools let you compare the amount of traffic for you and your top competitors. It's all in the book.The third step is analytics. There is so much useful content in this book that I can't give a short summary. Avinash has solid experience in setting up KPIs and dashboards for dozens of companies and you'll learn how to do this. He describes what is useful, how it matters, and how to use it. He tells you why you should avoid real-time reporting.A major issue in analytics is the soft numbers. Computers and the web gave us the promise of fully-trackable activity. Web analytics itself implies accurate measurement of data. This turned out to be an illusion. Nearly all of our clients are unaware that the numbers are off by as much as 30%. This is caused by a number of factors: the various analytics tools use different definitions for an event. Users block JavaScript, so tagging can't collect data. As much as 40% of users delete cookies every day. There are tracking problems with Ajax and Web 2.0 sites. And there are many more problems. The book has a clear explanation of log files, tagging, web bugs, and packet sniffing, along with the advantages and disadvantages for each one. You need to understand the technical issues to understand the strengths and limits of your numbers.Along with describing what to do, Avinash also tells you what to ignore. Analytics is not reporting, so don't deliver reports. Reports are not useful for business decisions. If they want reports, set up automated reporting. Nobody reads these anyway. You should also ignore page views, clicks, and exit pages. These are useless for business decisions. It sounds very nice to say you have four million monthly page views, but so what? What matters is the visitors' intentions and your KPIs.Avinash doesn't shy away from bold statements. He points out that analytics is interpretation and recommendation. But expensive analytics tools (which cost $50,000 to $100,000 per year or more) can simply be report-generating engines for your company, unless you have mastered their complexity. With tools like ClickTracks Analytics (which costs $90 per month) and Google Analytics (which is free) you can get all the reporting you want at a very low price. This allows you to invest in people or getting help from external consultants to move from reporting to doing analysis. You can even use both ClickTracks and Google Analytics. Google Analytics uses tagging and ClickTracks uses log files (or tagging); ClickTracks is good for SEO and Google Analytics is good for PPC. Each has features that the other is not capable of producing. By using easier tools, you can focus on the business goals instead of using the tool.He also brings up the problems with Web 2.0. We have a number of Web 2.0 clients. How do you use analytics in a Web 2.0 world? On Web 2.0 sites, the concept of page views is irrelevant, because Ajax and Flash tools don't require a new page or a page refresh. There are lines of code from Google that allow you to track JavaScript events in Google Analytics. His book is current enough to discuss these issues.A number of chapters are recommendations from Avinash: what is important? What is useful? What can you ignore? What is useless? He includes a list of best practices. The recommendations are extremely useful. That's why he is asked by Fortune 200 companies to help them with their analytics strategy. The same advice is in this book.Avinash Kaushik's book is easily one of the best books for analytics, SEO, PPC, SEM, or the web industry. If you read only one book on web commerce, this is the one.

I concur with the reviewer who said "could have been better." This book could have been written better, especially in regard to organization. Perhaps, if it was organized better, it would not have been so repetitive.The good point about the book is thatAvinash Kaushik provides some very useful practical tips from his years of experience as a web analyst. If you are a web analyst or would like to become a web analyst, then his book is a definite read for this reason. In addition to the book content, he also supplies a CD with some audios and PDFs.Other than that, the book is extremely irritating. It is very repetitive. Avinash keeps talking about how web analysts need to look at more than clickstreams, that web analysts should be solving business problems, not simply generating reports. Okay. I got it. It's not until Chapter 6 that he dives into web analytics concepts.His division of the book makes absolutely no sense. He allocates no time for the first five chapters, and then one month per chapter after that. Some weeks are less than seven pages. Does that mean it's supposed to take me more than an hour to read one page?He says that it is difficult to read "nine-page reports that are in Excel in six-point font." Yet he includes samples of reports in very tiny font. If he knows that it is difficult to read reports in small font, then why does he do this?And then Avinash has a habit of throwing in acronyms and jargon, without explaining what they mean. I had to do a lot of research on the Internet to figure out what he meant. For example:"Having your own environment means that you have immense flexibility in terms of bringing in lots more sources of data (for example, event logs from your Flash or rich Internet applications, Google search data, metadata from other parts of the company, and CRM or phone channel data)."What are these event logs he is mentioning here? Do Flash and rich Internet applications deposit event logs on the server? And what does Avinash mean by "metadata from other parts of the company"? What metadata? Avinash also never bothered to define CRM or explain what rich Internet applications and phone channel data are. Come to think of it, I don't even think he defined metadata.And lastly, Avinash has a tendency to discount certain data and make up or use new terminology or rules that often don't make sense. He seems to feel that browser statistics are irrelevant. These statistics are extremely important to web designers and developers. Certain designs look awful on certain sized browsers. Certain code works only in certain browsers and browser versions.All of his So-What examples passed for me. (See pages 122 to 124.) For the first one, I would survey visitors to find out why visitors are coming back to our site. In the second scenario, I would view all exit pages to see what kind of content they contain. In the third example, I would look closer at the data to figure out which of the top keywords to spend money on--the top 3, 5, 10, 20...His 10/90 rule of spending 10% of the budget on tools and 90% of the budget on people. Hmmm... We all know there are many high quality tools on the market that are for free and other ones that are extremely expensive. How can they both result in a ratio of 10/90.One more thing to note: In the book, Avinash says, "Visitors com to any website to accomplish a myriad of goals (regardless of why a website exits--for example, when was the last time you purchased on Amazon.com?)"Hmmm... Just recently. In fact, I bought this book from Amazon.

Lots of ideas, and the exposition is always clear. Kaushik writes quite well, his style is plain, straight and to the point, and easy to follow. He has the ring of authority, it helps that he is pleasantly sceptical about many things.However like almost all IT books it's quite repetitive...heaven knows why IT writers have to say things 3 times always. Make your presentations attractive, OK. But 3 times, at length, in different chapters? Segment, OK. But why tell us so very many times? And on it goes.It's genuinely a good piece of work, but could be so much better if he disciplined himself to keep the same content, but say it in 40% less space.

I'd like to give a minus star rating but I can't. BE WARNED! You will not get a book when you purchase this. You'll get an unreadable ring-bound copy of a printed PDF. Words and letters missing throughout the copy and fonts randomly replaced and words overlapping. I'd recommend avoiding this book AND the seller!

Web analytics isn't about numbers, page views, hits and sessions. It's about discovering what your users want, what they're doing on your website, and how you can help them get to where they want to be. Avinash does a superb job of explaining this in his book, and if you're in any way serious about doing web analytics the right way, you need to get this book.

This is an excellent start to analytics. I was already interested and got the opportunity to run the ompany analytics. This not only expains the how to but also explains the implications (much more useful). Definitely one that's going to get dog-eared!

This is the best hands-on guide to web analytics and the importance of analytics to any on-line marketing project I have yet found.The book balances both the high-level aspects of web analytics -- the philosophy, if you like -- with a huge amount of specific, practical, how-to information. It may seem like a big book, but I don't see how it could have been any shorter and still delivered so much.Fortunately, it's an extremely readable book. I like the language and the style of the book. Avinash's enthusiasm for the subject comes singing out of the pages. He makes the subject seem fun and he sustains that over hundreds of pages. That's a remarkable achievement when you consider the length.I have a shelf full of business or self-improvement books which try to make themselves readable and accessible by interjecting folksy anecdotes full of people with made up names every few pages. It's a very common approach. The result is often repetitious fluff. This book is very different. There is no padding of that kind: the examples given are all very clearly based on personal experience and are there for good reason. They are informative. They are not simply structural devices.The language itself is also distinctive and entertaining. I believe that Avinash grew up listening to the BBC World Service. Although it's an American book, there are notes in the language which echoe a different world in a way which I find refreshing.I cannot recommend this book highly enough. One of my colleagues has a copy where most of the pages are thick with highlighter ink. The quality of information is that high. It really is that good. If you can buy only one book on the subject, this is the one to get.