Web analytics is hard!
The blogger Avinash Kaushik recently won an award for his post that largely echoed the content of my first book, Web Analytics Demystified. While I have never heard of the award, I am happy for Mr. Kaushik and excited because his repeated mention of my book’s title is driving incremental sales. Thanks Avinash!While I’m happy for Avinash, I have to wonder about his repeated insistence that “web analytics is easy.” I wrote as much in a comment I posted to his blog. The comment has not yet appeared (not sure why, maybe Avinash is on vacation) but I’m interested in your reaction to my position so I figured “hey, I have a blog …” and have published the comment below.
Basically I think that web analytics is hard — not easy or necessarily complex — but I think that this is very good news! “Hard” is something that all of us can overcome with determination and effort, just like the New York Giants did this past Sunday … all we have to do is recognize the level of effort and determination that is required and set expectations accordingly.
The verbatim content of the (missing) comment follows:
Avinash,
Congratulations on your award. I’ve never heard of the award but I don’t work nearly has hard on SEM/SEO as you do. Regardless I’m sure winning this award is quite an important accomplishment for you.
I also wanted to thank you. As the author of “Web Analytics Demystified” — the classic work you say your post pays homage to — it was interesting to hear your take on the subject. More interesting was how your post positively impacted sales of the book — direct sales were up nearly 40% from the month previous!
Perhaps your post reminded people that some of us have been around “demystifying” web analytics for a long time.
Either way, despite some people’s complaints that you were inconsiderate in your attempt to game my brand and copyright, I can assure you that I do appreciate your willingness to continue to bring awareness to my work, both as an author and as a consultant. And I sincerely hope you didn’t mind my tongue-in-cheek title for my AMA presentation …
Still, I wanted to ask you one question: Do you think the Giant’s thought that winning the Super Bowl was complex? Or do you think they thought it was hard?
I ask because you seem hung-up on my assertion that “web analytics is hard”, going out of your way to try and convince people that it is “easy but complex, not hard” and that somehow complex is preferred.
And while I’m not sure why you’re so violently opposed to my assertion, I do disagree with you — web analytics is hard, and that is fantastically good news! Web analytics is hard just like an underdog with a young quarterback winning the Super Bowl is hard. Nothing good is easy, but anyone, anywhere can do something difficult simply by being dedicated to the end product.
- Sacking Tom Brady is hard (hell, very hard) but the Giants defense did it again and again and again
- Converting on third down is hard, but Eli Manning and his receivers did it when it counted the most
There is nothing particularly complex about football, perhaps save contract negotiations, and there is nothing overly-complex about web analytics. In my experience the complexity that does come up in web analytics usually comes from “over-promise, under-deliver” and end-user needs for basic tools like segmentation and data integration that are lacking in some of the most popular applications, not from the practice of web analytics itself.
- Web analytics is hard, but anyone with enough motivation can overcome this obstacle and, in your own words, create amazing, magnificent, and wonderful success!
- Web analytics is hard, but simply by recognizing this and moving beyond the “web analytics is easy” hyperbole companies are able to create the right expectation about the effort required to be successful.
- Web analytics is hard, and that is excellent, excellent news. Because if web analytics were impossible there would be no need for Google Analytics or any of the other great applications available today, but if web analytics was easy, most of us would be massively under-performing.
- Web analytics is hard, but just like the Giants and every other Cinderella team in the history of sports, hard can be overcome through sheer determination, effort, and a well-formed plan for success: often the result is greatness.
I hope you’ll print this comment, and I know my thoughts will do nothing to dissuade you from your “web analytics is easy” mantra, but I had some time (ironically waiting to fly to New York where I hope to catch a Giant ticker-tape parade this week) and wanted to comment.
Again, congrats on the prestigious award! I’m sure that Stephan Spencer and Tamar Weinberg were quite bummed to have been nominated against you — I know I would be!
Sincerely,
…
Some have accused me of “over-messaging” on this point and pointing out the obvious. Perhaps, but as long as I keep seeing the relieved look on my client’s and audience’s faces when I tell this simple truth, and as long as people continue to come up to me and thank me for clarifying expectations about their use of measurement technology, I’m going to stay on message. “Web analytics is hard” seems to match well with people’s experience, and more importantly, nobody appears too perturbed about this statement (except for perhaps Mr. Kaushik and his business partners.)
What do you think? Is web analytics easy? Is it complex? Is it hard? Does it matter? Are you perturbed by my assertion? If so, why? What am I missing?! I’d love to receive your feedback …
The Blackbeak Blog…. Arr! Web Analytics » Blog Archive » Web Analytics, A challenge per month added the following ...
[...] I’ve seen posts which have described web analytics as hard, or complex but not [...]
Steve Jackson added the following ...
This led to a big reply which I ended up posting on my own blog;
http://blackbeak.conversionchronicles.com/2008/02/06/web-analytics-a-challenge-per-month/
Bryan Eisenberg added the following ...
Eric,
I actually agree with both you and Avinash. The critical factor is where you are in the optimization life cycle. If you have never done any optimization and are just getting started with web analytics it is easy to take a dive in, kind of like getting your feet wet in a pool. At that point it is true any one can do it.
However, as your insights and issues grow it gets harder, kind of like doing the breast stroke in the ocean with a strong current. The breast stroke is challenging, but not once you learn it. However, once you try to swim in the ocean, the current, the weather, your physical conditioning all add the the complexity and difficulty (how hard it is) of the task.
The complexity of web analytics comes from all the factors beyond our control, primarily that we are measuring complex human behavior. Humans by their nature are chaotic like the waves of an ocean. A non-random, complex system larger than we can easily comprehend.
This is all about point of view and the audiences you are both speaking too. Funny enough I think you both agree with each other more than you think.
Keep the demystification going!
Mark Dykeman added the following ...
Eric,
To as certain extent, we early-adopters, net-veterans, or digerati, or whatever you want to call us, are being hoisted by our own petard.
For years we’ve been cajoling those less eager to get online: “Build a web site, it’s easy!” Information architecture, it’s easy! Setting up your own server, it’s easy! Running your own business online, it’s easy. Measuring your website, it’s easy!” I believe at one point in my career I told someone that I could be replaced by a typing pigeon.
Now I find myself explaining to clients, that “uh-yes, it’s all very easy to do, and can be done very quickly… but it’s actually quite time consuming, knowledge-intensive and work-intensive to do *well*.”
It all boils down to the adage - you get what you pay - and work - for. If it’s easy, you’re either Avinash, or you’re probably not working hard enough.
eric added the following ...
Jacques: Good point. Perhaps I am just responding to hundreds of emails over the past few years that have a common theme of general disappointment with web analytics. The expectation that all this would be easy when your comments are far closer to the truth seems to blindside people. I guess I repeat the “web analytics is hard” message to let people know they’ll need to dig and sweat …
Steve: Great post! The funny thing is that the experience we had in Helsinki has been repeated all around the world — with my clients, at Emetrics, on webcasts, everywhere I go. I appreciate your feedback and your sharing your own experience.
Bryan: Nice to hear from you and good perspective.
Mark: Brilliant perspective! It’s time to tell the truth and deal with that truth with whatever tools and experience we have at our disposal. And I have little doubt that web analytics is easy for Avinash — anyone who can explain this stuff as brilliantly as he does must think it’s easy! But there are too few Avinash’s in the world (sad, but true.)
Thank you all for your comments!
Web Analytics is Complex and Hard - Mymotech added the following ...
[...] Two of my favorite web analytics stars are at it again. It wasn’t difficult to see before that the divergence in their thinking centered around Eric Peterson stating that Web Analytics is hard, and Avinash Kaushik stating that web analytics is complex, and now it is out in the open. [...]
Jeffrey Toms added the following ...
Hi Eric,
You’re congratulations to Avinash Kaushik rings rather hollow…
Congratulations on your award. I’ve never heard of the award but I don’t work nearly has hard on SEM/SEO as you do. Regardless I’m sure winning this award is quite an important accomplishment for you.
Do you really think he will post your comment to his site? Your opening comment is a slight and condescending. I wonder if you will post and respond to my comment? Time will tell.
eric added the following ...
Jeff: Avinash did post my comment as I have yours. Perhaps I could have phrased that comment differently, but I can assure you that I am happy for Avinash in his successes. Ours is such a young industry, and I’m sure we can agree that we all need to work together!
Thank you for your comment.
Michael Feiner added the following ...
Hi Eric,
Web analytics is easy? Hard? Complex? Challenging? A web analyst would reply “well, it depends”.
I put myself in the (third) group - web analytics is easy and hard. It is all a matter of perspective, audience and time.
Echoing Bryan’s comment, WA can be both.
Perspective - initially WA can seem quite daunting (too many reports, which metrics work together etc.). Overcoming this problem is easy. Avinash does a great job helping people overcome the first hurdles in WA.
The issues you contend with are often more advance (e.g. how to measure engagement etc.). These are much harder to overcome.
Audience - Avinash’s blog attracts many online marketers and aspiring web analysts. They come to his blog to get ideas of how to use WA more effectively. They are not concerned with the future of WA or the intricacies of engagement models (though they would love for someone to just tell them how to do it).
Your audience is here for the debate and keeping up with latest developments in the industry.
A quick look at the typical comment on each blog illustrates the differences in audience. Comments on Avinash’s blog are shorter and often a thank you message. Yours are longer and more of the question/observation type.
Time - we all agree WA is nascent. As with anything else, things get easier with time as more and more people are using it (whatever “it” is).
Avinash is helping expedite this process. You are helping demystify the front line.
Building the Wright Brothers’ plane in 2008 - easy
Building a Boeing 787 in 2008 (even for Boeing) - hard
Painting a Picasso, Van Gugh in 2008 - easy
Becoming the next Picasso, Van Gugh - hard
Avinash’s blog is so popular with established web analysts as well as new comers because he constantly reminds us of all the great stuff we should be doing with WA and are not.
So it is all a matter of segmentation - easy and hard can co-exist.
Thanks,
Michael
Web Analytics Bloglinks der Woche (KW 6 / 2008) | Webanalyse & SEO - News added the following ...
[...] Web analytics is hard! (Eric T. Peterson) [...]
eric added the following ...
Michael: Thanks for the comment and I see your (and Bryan’s point.) The only problem I have with this is that I tend to view web analytics as a more holistic endeavor than some of my peers — I believe that web analytics is not just about “making” the measurement but is about actually “using” the measurement in practice throughout an organization.
No doubt there are aspects of our practice that are easy! Google Analytics makes it a no-brainer to determine the bounce rates for our pages, etc. which is nice. But what do you do with that information? Do you just go back to someone in your organization and say “our bounce rate sucks, fix it!” and see results?
It’s not that easy, and we all know it.
I very much agree with your assertion that things will get easier over time — inevitably companies will come around to the central principles of RAMP and other measurement best practices — but we are not there yet. Because of this, I see Kaushik and Hamel’s messaging as dangerous if not necessarily incorrect. It sets the wrong expectation for NOW, not the future, and in my personal and direct experience leads to confusion and ultimately disappointment.
But anyway, I get your point and thanks a ton for your comment. Some people haven’t been able to take a step back and think objectively about this issue (perhaps myself included!) and your point does that.
Michael Feiner added the following ...
Eric,
Sorry for taking so long to reply - I just noticed your reply now.
From my acquaintance with Avinash and having reviewed Stephane’s recent podcast, I think it is safe to say that they too take a holistic approach to our world.
The more I think about it the more convinced I am that both statements are true.
You are spot on about setting expectations too high. Managers, unaware of difficulties of the day-to-day WA process, get nice graphs and think - “That’s great, now I want to know how this impacts XYX”. When they’re made to realised that it isn’t that easy they can easily get frustrated with WA.
How many times have we all been in this situation before?
But equally, if online marketers and aspiring web analysts are confronted with the difficulties of WA upfront than the risk is they will abandon ship before it even sails.
There main two concerns would probably be:
1. Lack of confidence to read the data and make some basic conclusions
2. Concerns over presenting the data to management and having to field subsequent questions
Therefore, I think both of you are playing an extremely important balancing role in our world. Avinash is making more people use, practice and trust WA whilst you are making sure that they don’t get disillusioned too quickly and are made aware that there are no silver bullet solutions (only this week I’ve been asked by a salesperson to get me XYZ data for a site I’ve never analysed before and am unfamiliar with the Omniture implementation on it – he just thought it is a simple exercise to get the data out of the box).
Hope this makes sense.
Keep on the debate - it is very healthy. It makes us all think harder.
Thanks,
Michael
Defining the Five Aspects of Web Analytics « Web Optimization Blog added the following ...
[...] of site traffic has become quite complicated. Even Eric Peterson has come right out and said “web analytics is hard“. On that we [...]
Patrick added the following ...
Hehe,
this debate reminds me sooo much of the ‘SEO isn’t rocket science’ debate :-).
I think there are two types of careers (and of course many shades in between):
1) it’s very very hard to learn the knowledge you need, but once you’ve done that you’ll probably be successful (e.g. earn a lot of money) getting the knowledge of somebody with a PhD in physics or mathematics.
Very, very difficult to acquire that type of skill. This is ‘hard’ and ‘complex’.
2) it’s not overly hard to learn the knowledge as long as you have enough time…for example marketing: it doesn’t take a genius to ‘learn’ marketing knowledge.
But what is hard is applying that skill very successfully.
Another example for this could be running track: It’s not hard to run 100m (it’s neither ‘hard’ nor is it ‘complex’ for the average human being to acquire the skill of ‘running’) - but it sure is hard to become really good at it (achieve a sub-10 time).
I think we should talk about whether the knowledge you need for a skill (like web analytics) is ‘hard to acquire’. Becoming really good at it sure is hard (even with something as simple as running).
Whenever I hear somebody bash a field because it is ‘easy’ - for example ‘SEO is easy’, ‘SEO isn’t rocket science’ (I know Avinash is NOT trying to bash the field of web analytics :-)), I find that funny, because often it’s true that learning the skill is ‘easy’ (or at least not ‘rocket science’), but becoming very good (or even great) at it is almost never easy for most people (after all becoming great at it usually means becoming better than the majority of people).
eric added the following ...
Patrick: Thanks for your comment, but apparently you’re wrong. Or at least Matt Belkin of Omniture thinks so:
Seriously, I don’t know why people like Avi, Matt, and Stehpane Hamel struggle with this truism. It’s not like web analytics being hard will stop companies from investing in their products and services any more than SEO being hard will stop folks from trying to game Google.
Thanks for your thoughtful comment.
eric added the following ...
Michael: Oops, same problem. I saw your reply when I replied to Patrick, sorry!
You make a good point, but see my comment above. I sincerely doubt that telling people the truth about web analytics and clearly setting expectations will make people “abandon ship”.
“Web analytics is hard” sets expectations, not requirements. “Web analytics is hard” clarifies for your sales person that you might need more than a few minutes to satisfy his request. “Web analytics is hard” reminds us that this all pretty much new and we have to stay on our toes.
I know people want both Avi and I to be right and that’s okay. I respect his right to have a different opinion, even if I think he’s wrong. Hopefully he extends me the same consideration.
Thanks for your comment!
Web Analytics Demystified » Blog Archive » Matt Belkin of Omniture: Web Analytics is Easy! added the following ...
[...] baseline. That’s it! That’s why I think it’s easy. I know other bloggers have argued that analytics is hard, but I’ve done this for a living and I can tell you that it’s [...]
Patrick added the following ...
Hi Eric,
I had a bit of trouble understanding your point (because Matt Belkin actually disagreed with your notion of web analytics being hard), but then I read the comment again and (hopefully?;)) understood you were being a bit sarcastic before continuing with ‘Seriously,…’ hehe.
Patrick added the following ...
ah, I’d like to say something else:
I found this post wondering if web analytics is hard or not (I typed in web analytics + hard or something) or well mostly wondering what other people would say about it.
But I always had a thought in mind:
If web analytics is so easy then why does it seem to be so hard to find good web analysts? I have Avinash’s book (and am a regular reader of his blog) and I know he is of the opinion that it’s not easy to find good web analysts (he blogged about it and has a chapter on this in his book), so I’m actually surprised to read that Avinash is spreading the message of web analytics being so easy (I’m not very immersed in this industry, yet).
I know he’s trying to break it down and explain web analytics in a way that anyone can understand it (I think he blogged about this being one of the reasons why he thinks his blog has become successful), so I would have thought he’s doing this in order to spread the use of web analytics (”if you make something easy to spread it spreads further” is a well known concept in SEO and word-of-mouth marketing (if you say web analytics is incredibly hard chances are fewer people will give it a try and it won’t spread far).
But I do see your point about this setting too high expectations which then would be bad for the industry, of course.
If you ever talk about this with Avinash in person, please ask him why he thinks it’s so hard to find talented analysts if web analytics is so easy :-)
Giadascript » Blog Archive » CPA: l’incubo del marketing manager added the following ...
[...] quindi dell’opinione che la Web Analytics non sia semplice. Esistono dei software che offrono dati più o meno facili e veloci da aggregare e da leggere, ma [...]
Analytics Talk » Blog Archive » Do we really “get” Google Analytics? added the following ...
[...] is web analytics easy? That’s debatable, but the simple truth is that there are millions of people out there getting into web analytics via [...]
Building An Optimization Culture | FutureNow's GrokDotCom / Marketing Optimization Blog added the following ...
[...] community has been split on this issue. Eric T. Peterson, Web analytics consultant, argues Web analytics is hard, while Google’s analytics evangelist Avinash Kaushik argues Web analytics isn’t hard. [...]
analyticsfan added the following ...
A few observations and questions.
1. Exactly what aspect of it is hard or easy? Math? IT? Selling to the business?
2. Certainly compared to some quantitative fields, there is nothing quantitatively difficult about most basic web analytics that companies need - that does NOT take away from its value - difficulty does not imply greater value.
3. Why is it hard to find a “good” analyst? I suspect it comes down to money and lack of business understanding by buyers, resulting in would-be candidates going elsewhere:
- most job listings are searching for someone who cranks out simple reports: the business does not understand the potential
- what analytically talented individual would go for that kind of job vs. other more interesting, lucrative options?
4. Certain vocabulary in this are is overblown: “analytics”, “optimization”, and “segmentation” come to mind. These are better understood terms in more mature areas such as database marketing but the web analytics popular usage dilutes them into something a little oversimplified. In these usages, all of those areas are “easy” in my mind. In the broader area of marketing analytics, that is not necessarily the case.
The realy hard part imho is that business doesn’t get it and it is difficult to explain something somewhat technical to non-technical, non-quantitative people who do not know what they do not know.
Jacques Warren added the following ...
Well, I am not so passionate about this; I’m happy my name is not big enough to care.
I am not saying this because I am an independant Web Analytics consultant; yes, Web Analytics is hard. I have analyzed many web sites, and assisted with many more implementations. This ain’t easy stuff. Not impossible, not insurmountable, just hard. Why is that a problem?
You don’t just pick up gold nuggets. You dig, you sweat, you earn them.