This I Believe [A Manifesto for Web Marketers & Analysts]
National Public Radio had a long running show called This I Believe.
In it a person shares their private philosophy. Each story is extremely inspiring and I am a huge fan of the show.
A couple of my favorites: Muhammed Ali, and from 1953 Margaret Chase Smith.
I wish I had something personal that was insightful enough to share. I don't.
But I had the program This I Believe on my mind when last year I was asked to write something for the Korean version of my book, Web Analytics: An Hour A Day.
Here's my little professional manifesto of a few things I believe….
I have to admit that I love the web. Love, love, love.
It's quite simple really, in our humble history there are few things that have transformed human beings in the way that web has. And to think that it is still just a baby.
I am amazed at how it has democratized access to information, it has allowed the little guys compete with the big guys, it has created opportunities for people and businesses where none existed.
The web also provides a unique and wonderful opportunity to understand our customers (visitors to our websites), to learn from their actual behavior and transform how we do business, to achieve the near impossible: improve satisfaction and revenues!
Here are five things I believe in when it comes to achieving success with web data:
1. I believe in the power of people (your employees), not tools.
I have a 10/90 rule . If your budget is $100 then spend $10 on tools and professional services to implement them, and spend $90 on hiring people to analyze data you collect on your website.
The web is quite complex, you are going to access multiple sources of data, you are going to have to do a lot of leg work. Blood, sweat and tears. You don't just need tools for that (remember 85% of the data you get from any tool, free or paid is essentially the same). You need people!
Hire the best people you can find, tools will never be a limitation for them.
2. I believe that reporting does not equal analysis.
Most "Analysts" and Marketers think that their job is to simply churn reports out and drop them over the fence and that's enough to create data driven cultures.
False.
Reporting is the act of providing data. Analysis is the act of providing actionable information.
No company will succeed by having a army of report writers, they will succeed by having a small group of "Analysis Ninja's " who transform data into information.
If you get just data, reject it and reject the person who gave it to you.
3. I believe in not just understanding the What, but also understanding the Why.
I am a Mechanical Engineer who has a MBA. Put another way I am a quantitative person.
It was tough for me to realize that all the quantitative data that we have access to can only help us understand the "What" on our websites. What happened, what pages were seen, what products did people buy, what pages did they leave from, what campaigns they came on etc etc etc.
But no matter how hard we tried all that data is not very good at understanding the "Why". Why did people click in the order that they did, whey did they abandon the cart, why did they buy from our website when we don't give discounts, why did they all leave from this page and on and on and on.
To understand Why use Surveys, Usability Tests, talk to your customers and really listen.
The above lead me to creating the Web Analytics 2.0 strategy, at the center of which is the desire to be smarter than we have ever been before.
It demands a lot of you because you deserve a lot more than you have today.
4. I believe that God created the Internet so we could fail faster.
In the offline world it is very expensive to experiment and test, the cost of failure is very high. As a result we don't take risks. We keep doing what we think "works", until the day we go bankrupt.
The web changes that.
You can take dramatic risks, at very low costs and learn big. Your website is nothing but a machine built to make you smart by taking lots of risks.
Why should you tolerate ideas getting killed on conference room tables or by your HiPPO's?
Why accept opinions when you can convert them into hypothesis and get them validated for cheap and quickly?
Why not let your customers actively be a part of helping you create customer experiences that deliver value to them AND to you?
The cost of taking risk on the web is low. You can try a idea. As soon as it is live data starts following it. If the idea is a total loser then kill it fast, does not have to cost you a ton of money. What is more likely is that you will find winners that you had never imagined.
Give it a try. Fail faster.
5. I believe that Web Analytics is complex, but it is also very sexy and hugely rewarding.
Nothing in life is easy. Everything takes some time to learn, some effort from your end, some investment in time. The same is true about Web Analytics.
But you'll be astonished how much progress you can make in just a few hours if you put your mind to it. Don't be scared of data, don't buy into people who try to scare you by telling you how hard it all is.
Don't believe me? Read the primer on web Analytics, it can jump start the process of changing your life – in three and a half hours!!
It is such a high when you dive into unraveling the mystery of data, discover yummy insights and to fix customer problems – helping people you have never met, helping make their lives just a little bit easier.
Start fast. Start today. Don't stop.
Five simple things I believe in, things that I hope will help ignite your own love affair with web analytics. It will be an exciting relationship, just like the one you have with your spouse or partner or boy/girl friend – every evolving with never a boring moment.
Good luck and god speed.
So what do you believe in ? Please share one thing, or five.
Subscribe 
July 27th, 2009 at 05:32
Short and sweet, baby, short and sweet. Change is upon us, the economy is turning around. Good post, as usual.
G
July 27th, 2009 at 06:02
Great!
I believe in the use of visual representations in web analytics.
They help to explore, make sense of and communicate data.
Once the meanings have been discovered and understood, they need to be communicated to others.
Visual representations are essential to perform data analysis.
July 27th, 2009 at 07:56
I think this is very close to a personal post Avinash :)
BUT… "5. I believe that Web Analytics is complex, but it is also very sexy[...]"
That last point is going to be the hardest sell you'll ever have ;)
July 27th, 2009 at 08:14
Is this officially the shortest Occam's Razor post ever? Analyze THAT! Great list, Avinash! One minor addition/different angle that I have — I absolutely see reporting and analysis as being two different things. And, my definition even aligns with yours. But, I look at reports as typically being "performance measurement" (of anything — a company, a department, a person, a web site) — they're very much the "What" in that they are grounded in metrics with pre-established targets based on what you *think* you're going to accomplish. They then tell you — objectively — *what* you accomplished. Did you accomplished what you thought you would accomplish?
Analysis is the *why* and the *how*. Reporting helps you determine where to point your analysis ninjas — if 75% of your metrics show that you achieved pretty much what you expected, then the analysis should focus on the 25% where you were wayyyy off and missed your targets the most dramatically. Analysis also comes into play when you are looking at a new project or initiative and are trying to figure out the best approach.
"Reporting" can and often is a recurring phenomenon that should be as automated as possible (with checks to make sure data's not just being puked out for the sake of puking out data). "Analysis" should be a series of one-time exercises.
July 27th, 2009 at 08:21
I LOVE #5, "I believe that Web Analytics is complex, but it is also very sexy and hugely rewarding," because it is so true, and you could replace "Web Analytics" with "every woman" and it would still be true. ;)
I believe that web analytics is the difference between visitors and customers.
July 27th, 2009 at 08:38
I believe that simplicity is the best policy in web analytics.
As Avinash has expressed many times, focusing on a *critical few* metrics – or even better, a *master metric* – that informs the design and continued improvement of a web site is the best way to realize the benefits of web analytics efforts.
July 27th, 2009 at 08:58
Avinash,
Good post and great opportunity to reinforce a foundational philosophy. Without a doubt, yours would be my top picks. I do have a few that I think lay a little lower on the importance hiearchy.
I believe that data alone can be misleading and sometimes even (gasp!) lie.
This is further support for your 3rd belief (I believe in not just understanding the What, but also understanding the Why).
It sometimes takes the voice of the customer to set us on the right path to understand what the numbers alone might not be fully telling us.
I believe in perpetual forward motion.
Keep moving forward. Looking for new opportunities to learn. New opportunities to track. New opportunities to optimize. Once you learn something and apply your learnings, don't stand still and pat yourself on the back. What's next? Move.
I believe in the power of the less than obvious.
I think that there is real power in taking the roads less travelled to gain unique insights and perspectives. I believe in custom reporting…marrying dimensions that on the surface, don't seem like the most obvious dimensions to merge. Sometimes, you discover something that never would have come to your otherwise. Segmentation in unique ways can lead to great discoveries. Don't just do the obvious…by source or campaign.
I think that this goes without saying, but I believe in the power of segmentation, too.
I agree with Tim's comments above on Reporting versus Analysis.
Good post, Avinash. I look forward to reading other comments.
July 27th, 2009 at 09:05
I also believe in giving before receiving. One point, as simple as it may be, yet so powerful to me is what Google Analytics has done. It's FREE!!! Google gives first and then asks afterwards and even that is optional, e.g Adwords. That alone speaks volumes.
July 27th, 2009 at 09:26
Excellent post! I believe that the web allows humans to continually do what humans have done for thousands of years…that is to connect and communicate to other people. That entails listening and understanding…the two things that the web has highlighted. Listening includes the web analytics, the surveys, the talking to people, the testing…now, we have to use the information from listening and understand what is happening. This is what you call actionable insights. The future is really about listening and understanding.
July 27th, 2009 at 09:32
Love this post AK.
July 27th, 2009 at 09:58
I believe that web analytics is a job. There is life beyond the cubicle walls.
WA may be the best job on Earth, but if it becomes too fun we'll do it for free. Haha.
July 27th, 2009 at 10:06
I believe in the people behind the analysis – Smart people can fit data into the mold of their assumptions and hate to be wrong… Really smart people continually challenge their own assumptions and are happy to be proven wrong because it means they've learned something. It's part intellect, part integrity, part confidence (to admit you are/were wrong). Not only can these people make a project/company/Web site successful, they make everyone around them a little bit smarter.
July 27th, 2009 at 11:44
I believe that analytics is really about making your visitors happier.
Anything that gets in the way of making your visitors happy should be fixed or discarded.
Analytics is a tool for helping people find you and gives them the opportunity to experience your brand – without being encumbered by sloppy landing pages, flash introductions (yuck), or bad navigation.
July 27th, 2009 at 11:46
Avinash, once again great post. Very concise and insightful.
Was it Peter Drucker the one who use to say that great ideas normally come together with a lot work? In any case web analytics and online business optimization are certainly great ideas.
Cheers,
Jose
July 27th, 2009 at 13:07
Benoit: I believe in that very strongly as well, death to tables! Ok maybe that was too strong. Let's try this again: gentle passing away to tables. :)
Tim: You are not saying my posts are too long are you! :^)
I like your framing. Reporting tells you What Happened, then you have to think. Analysis tells you What To Do.
[PS: That is not my shortest post! Actually maybe my shortest on-topic post. This probably was my shortest post, even though I updated it after a week:
Life, Liberty And The Pursuit Of Happiness]
Tara: I absolutely LOVE this: I believe that web analytics is the difference between visitors and customers.
Alice CS: I can totally sign up for: "data can be misleading and sometimes, maybe often (!), lie". : )
I like the idea of perpetual motion.
I gave my wife this refrigerator magnet a while back:
"Marriage is not a destination, it is a journey."
Replace marriage with web analysis.
Jenn: I agree with you 100%. The web allows us to do things humans have done for a millennia, but now we can do them at scale. That I think is the real advancement.
To your thought about the future I would add one minor thing (in a business context): The future is really about listening and understanding and two way conversations.
Alex: You and me, two peas in a pod! See my "rule" #9 here: Nine Rules To Work / Live By
ODA: Lovely thought. Real smart people are: Part Intellect. Part Integrity. Part Confidence (to admit you are/were wrong).
Matt: I like that sentiment.
We talk about customer satisfaction and what not. But I like the thought behind we should make our customer happy. So much better than "satisfied".
-Avinash.
July 27th, 2009 at 13:10
Hi Avinash,
Very good post :). Great manifesto.
I only disagree that Web Analytics is complex. I think it is very very easy in comparison with traditional analytics.
I have a background in business intelligence and I think web analytics is much easier, simpler, faster, more rewarding, more measurable, more fun etc…
Let’s see some differences from my perspective.
1) Traditional analysts would kill to have the automated way of data collection. I used to work in a company that there were 25 people just to collect move and distribute data from one client to our databases. At least 25!!!
2) The interface for web analytics is so simple that everybody can use. My previous interface was most of the time command prompt and scripts in order to handle databases. It was not unusual to do 2-3 day project just to summarize some data. Can you compare this with web analytics?
3) Quality Assurance was a separate process for every project and it was not unusual to take 30-40% of the time in the project. All this just to have more confidence in the data.
I really could have a very long list. I really think that web analytics is not complex. However doing proper analysis and finding actionable insights is challenging but a lot easier than traditional Analytics. I also love web and web analytics because it is such a fun playing with it.
Christos
July 27th, 2009 at 13:21
I believe the web was invented for all of us with a Liberal Arts education. We have diverse interests in many different disciplines. The web finds a home for all of us – making a living at something we like to do!
July 27th, 2009 at 14:44
Hi Avinash,
Thank You for the offer to share our beliefs. I am a big lover of internet. I love it, live it, celebrate it and keep busy thinking how I can help to unleash its power further and contribute my 2 cents to change lives for better through it.
Here is what I believe in.
1. I believe internet "magnifies" good and bad luck.
One can get lucky or get stuck using its power. Lives are made and destroyed on web depending on what one indulges in.
For ex, I found my first USA job and every job thereafter in this blessed country on the internet.
Again for India and Indians like me in general, internet brought loads of luck. In an extremely orthodox and closed society; at-least on the surface, the liberal, entrepreneur voice and lot of pent up energy got a chance to express itself.
2. I, like you, believe internet is experiment pad for wannabe enterpreneurs.
I have started, failed and tasted "mild" successes already in few online ventures and am more than ever determined to do even better execution for my next ones.
3. I believe human experience is mostly passe when it comes to simple matters of customer service
If I could sign up and "cancel" most of my services by myself online, I do not want to subject myself to the mostly untrained or half trained, not empowered and sometimes rude and bully manpower of most of the companies.
4. I believe, internet is going to usher in huge home based economy.
GDPs will go down for a good reason and an individual would be able to compete with big business and its the human experience, where big businesses, inspite of their fancy technology, will fail to compete with the neighbourhood guy.
Some of it is already happening.
5. I believe web analytics still does not and can not give us the "why".
The "why" can only be interpreted by human talent and experiences of the analyst. The "why" lies in this analyst's higher awareness. Higher awareness that "why" is the longing of every human to experience this life in an ever improving way.
Its human nature to get excited when we see a person, place, product or service that enhances our life experience.
This "why" applies as much to friendships, relationships, marriages and countries and towns we choose to live in as it applies to websites and products and services from businesses and governments.
This "why" is the absolute bottom line for any company.
It provides explainations into organization's thinking. For ex. why Motorola thought Q is the next big thing after Razr and why Apple thought iphone is. Why yahoo's new mail system didn't work correctly for few months for many users including me and they thought a link to classic mail would be enough to make people continue using it. I for one shifted to Google completely after years on yahoo.
Few of these "why" have already been resolved for companies willing to pay attention. That we can not as a website throw useless forms in the face of our user. That we can not lag behind our competitors when it comes to user experience. That we must think of customer experience and work backwards to accommodate our internal agendas and process problems rather than the other way out.
This "why" explains the disclaimer that past successes are not an indicator of future. It explains why "Success can be never ending and failure is never final"
Thanks
Rohitaash
http://rohitaashsays.com
July 27th, 2009 at 17:13
Hello Avinash,
Nice professional manifesto, great post, and amazing conversation stimulator.
While I don't have a separate set of beliefs for my professional life, here is what I use for all things interpreted for Analytics.
1. Have integrity. Tell the client what the data says, not just what they want to hear.
2. Give up "being right". If you analyze with preconceived ideas, you will be right; if you keep an open mind, you will be surprised.
3. Be powerful. Be straight/clear/focused in your communication. Show what you see with insights that make sense for the business.
4. Be courageous. Remember: it's what the data says and you are only an interpreter. Don't pull punches regardless of the audience.
5. Be peaceful. Some things cannot be fixed, so just ride them out.
July 27th, 2009 at 22:22
Nice post Avinash.
I believe in Occam's Razor :-)
I agree that Reporting is not Analysis. However, I also believe that both are needed. Reporting is never about future — it is about past and current performance (monitoring year-over-year, vs Plan, vs Forecast etc.). Analysis on the other hand can and should play a significant role in shaping the future of your business.
Enjoyed the post and a good read as always.
July 28th, 2009 at 06:07
My favorite part was the website providing cheap, endless experimentation potential. And I answered you with what I believe on my blog.
And I mean every word. Especially the first three.
http://aprilwilson.net/blog2/2009/07/28/this-i-believe/
July 28th, 2009 at 10:29
Hmmm…. my wife had trouble sleeping last night. By 2am she grew desparate and asked me to talk to her about how web analytics works. By the time I reached sessionization and visitor recognition she was deeply asleep for the rest of the night. 8-)
Sadly, then I couldn't sleep anymore…
Akin
Unica
July 28th, 2009 at 10:59
I believe web analytics makes most creative meetings redundant. Yeah yeah, we all like pretty designs but we want the site to make money!
My girlfriend often asks me what I dream about. When I tell her "reporting" she gets grumpy but I try to explain to her, "I report at night so I can make insight in the day".
She doesn't like that much. :)
July 28th, 2009 at 21:02
I believe in the focus on the funnel. We don't necessarily need to widen the top to get more out faster. We need to focus on spillage and widening the bottom.
July 29th, 2009 at 06:47
I believe that I love reading your blog. I used to listen to This I Believe back in the day. I can't remember which episode, but I remember distinctly being moved by the passion of the people on the show. This passion is something you bring to the Internet and web analytics.
I listened to your interview with Kelvin Newman of AI Digital last night. Great stuff there too. In it you mentioned how free tools like Google Analytics and Yahoo Analytics have both opened up the game to a world of new players as well as raised the bar for the paid tools.
I believe firmly that one person's ability to positively impact online goals has shot through the roof in recent years. No longer do you need to contract with the most expensive consultants and agencies to get some insights. You need a free tool, a passionate team of "Analysis Ninjas" and a culture that's willing to accept that insights will get you further than HiPPOs.
Every two weeks you fuel the fire, Avanish. Thank you so much for sharing your insights, passion and energy with the world.
And as luck would have it, I strongly believe that there is a growing demand for this type of passion and insight. Perhaps that's one reason we all keep reading.
July 29th, 2009 at 15:32
Christos: I grew up in the traditional BI world, I totally understand where you are coming from. I agree that so many problems we might have there don't exist in WA (we have new ones! :)).
Kristen: Brilliant!!
I wrote in one of my facebook status replies that I am a "web made person". I have a mechanical eng degree and a MBA. But nothing there prepared me for what I do. I had no idea at the end of all that what I was good at or what I really wanted to do. I was on a healthy job track just doing a job.
Now I do what I am good at and what I enjoy doing. I wound never ever have thought this is what I wanted to do.
Massimo: I love your #2. Especially in our world of data that is so profound because if I torture data enough it will tell me anything I want. :)
Ned: I agree that reporting is necessary. Since the balance today is 99.9 to 0.1 in favor of reporting I am going to be "mean" to reporting. :) I promise you though when the balance is closer to 75 to 25 (still in favor of reporting) I'll be nice!
April: You are such a sweetheart, thanks for sharing your own list!
Akin: I need to talk to your wife.
I am sure I can absolutely convince her of: 1. how magnificent her husband is and how valuable he is to the community 2. how amazing data driven decision making can be. :)
Rob: I don't know if it makes them redundant (no my UCD friends are not holding a gun to my head as I write this!).
I think people have opinions. It is ok to have opinions. It is ok to be creative. It is ok to be quantitative and have wild and crazy opinions.
I prefer that every opinion that people have is expressed as an hypothesis. Then there is a built in assumption that it will be tested. :)
Opinions: Bad. Hypothesis: Good!
Josh: When you said podcast I had no idea what you were talking about. :) I had to go to Google and find it and then I thought "aahhh… I did do that". :)
Thanks so much for the kind words.
-Avinash.
July 30th, 2009 at 01:14
I believe web analytics is an attitude.
I believe it's a perpetual challenge and thirst for knowledge.
And I believe it's also the pleasure of being able to read you! :)
July 30th, 2009 at 11:23
Avinash – this is a fantastic post and one we'll reference at our company in the foreseeable future. Thanks again!
July 30th, 2009 at 15:04
I believe in Web Analytics as in life, we have to do the path, but with someone like Avinash the learning curve becomes smaller, although the experience is only gained doing.Better than doing the reports is to know which reports do you need to accomplish your goals and have the perspective 360º of the customers.
Thanks for sharing
P.S. I am finally reading your book
July 31st, 2009 at 13:53
Hi Avinash,
It is an irony that the invention of the Web has not been received as much recognition as other great inventions. (Except that it has been included in the list of inventions of the century)
Web has given great power of "sharing". Web analytics has made it more powerful.
TGIF
Ketan
July 31st, 2009 at 15:57
I believe that web analytics is not just about numbers. It's about quality decision-making.
I believe that web analytics is more than just analysis. It is the very heart and essence of marketing.
I believe that analysis pays the best ROI when the numbers rise up off the page and become a story.
I believe that analysis is only worthwhile (and a whole lot more fun!) when it is actionable.
I believe that web analytics is great work if you can get it, but I would still do it if I won the lottery!
August 3rd, 2009 at 19:56
Avinash:
Great post and one that I will be having my students read this semester — it gets at some of the most important things to consider when using the web for marketing in all its forms.
I believe that it is our job to TEACH upcoming professionals just how important data-driven decision making is.
I believe that analysis of metrics is actually creative critical thinking and I hope to show my students that being creative isn't just about drawing a pretty picture.
I wish more people believed.
Thanks for a great post!
Elaine
P.S. see you on the 14th here in BTV!
August 4th, 2009 at 06:19
I believe it's an analyst's job to take a vast amount of data and boil it down to one easily digestible, graphic and/or sentence; so an action can be defined and implimented.
August 5th, 2009 at 09:20
I like that you believe in people, Avinash! I believe in them too :) People are the one who made the progress for over so many centuries. Machines (and tools) can help us but cannot do all the work for us. And besides where will the fun be if we let the tools everything for us? :)
August 6th, 2009 at 13:56
I followed some of your links and learned a few things. I never even paid attention to the "Bounce Rate." I'll be checking back again. Thanks for the knowlege!
August 10th, 2009 at 08:54
I believe web analytics can impact the world of nonprofit fundraising.
Through analytics we can understand the hearts and minds of our users / donors.
We can tell them a life changing story, challenge them to get involved, and respond to their needs through data analysis.
I believe in data analysis for good.
August 18th, 2009 at 00:46
In your blog entry correct?
You mentioned that "Bounce rate … measures the number of people who landed on your site and refused to give you even one single click!" ( http://googleblog.blogspot.com/2009/02/stop-bouncing-tips-for-website-success.html)
What happens if a user clicks a javascript button that then downloads software from another domain, like a subdomain, and then restarts the browser? In such as case, the user came to the page and left from the same page. The user did "click" the js button.
Those seem like clicks, but Google Analytics would still report them as bounces since they are not page clicks. Is that correct?
Please share thoughts.
It's a healthy debate going at my company :)
August 18th, 2009 at 09:12
Alec: It is not that complicated. :)
If you have said buttons you can simply encode them with a onclick event. Here's my feedburner link as an example:
a onclick="javascript:pageTracker._trackPageview('/goal/feed');" href="http://feeds2.feedburner.com/OccamsRazorByAvinash" title="Subscribe to my feed, Occam's Razor by Avinash Kaushik" rel="alternate" type="application/rss+xml"
Above is not a javascript button of course. But you can do something similar.
The javascript:pageTracker will send a "fake page view" into GA (or whatever too you use) and that second "hit" will be recorded causing that session not to be counted as a bounce (as it should not be).
Hope this helps.
Avinash.
August 18th, 2009 at 13:08
Thanks so much for the tip, Avinash! We'll give it a try.