Ask a famous blogger, a published author, a random twitterer or your mom how to succeed in web analytics, or how not to be a Reporting Squirrel. The answer will invariably be:
Before you provide the data, ask the requestor what is the business question they are trying to answer. Then fulfill that need.
It is a good answer.
Most of the time they, Marketers /bosses /HiPPO's, ask and we puke data out.
The result is also almost always the same.
After a while of doing this you, poor Squirrel, want to jump off a building. As if that was not painful enough, during the course of your employment the company made no actual decisions based on web analytics data. Ouch.
I call it a lose-lose.
If the outcome is so obvious. . . why do we still have this problem? Why is it that we don't ask for the magical thing you were told to ask since the day you were born into this world? Business questions.
Part of it is that we might not be in the position to ask for that question (I don't buy this 80% of the time, sorry).
Part of it is the case that we don't often understand the difference between a business question and a report request. Even for experienced Analysts / Consultants.
Let' us solve this problem today.
I really want you to become an amazing Analysis Ninja and help ensure that this issue never comes back to bite you in the butt.
Come with me.
Why Ask For Business Questions?
One word: Context.
I am sure this will not surprise you, but it turns out you are a very unique person. You are distinct from all your other siblings, even your twin.
And it turns out every business is unique, every website is unique. There is one and only one of its kind in the world.
Even if you completely copy and paste someone else's website and start executing, your business is probably unique in terms of the individuals you have collected around you and how they work together.
Or perhaps while you sell via the retail channel like everyone else, your strategy is different in its focus on driving purchases through the web, or maybe you are obsessed only about offline store sales. Or perhaps while you have 100% copy pasted GroupOn's business and are executing it in Russia. It is 100% likely that you are solving the problem completely differently than the US GroupOn site you are copying.
Or maybe you are HP and are obsessed and 100% focused on solving for Customer Satisfaction on your eCommerce site and your closest competitor Dell is obsessed 100% on getting more and more conversions.
This is a long preamble to tell you that the strategy we normally execute in measurement is wrong. We read books / blogs on Metrics & KPIs and we think we know what to do for an ecommerce site or a blog or support site etc and we start reporting what the blog / book recommended.
Then we are shocked no one cares one bit about the data.
If you are unique, why should you crack open a standard analytics tool with its standard reports and metrics and get going?
Or why simply respond to a "report request" and start data puking? The person at the other end is probably uninformed about Analytics and Segmentation and what is possible (even as they are supremely qualified to do their job in Marketing / Sales / HR).
You need business questions because
1. Rather than being told what metrics or dimensions to deliver you want business context: What's driving the request for that data? What is the answer the requestor looking for? Then you apply smarts because you have context.
2. Best practices are highly overrated. If this is your first day on the job, sure go ahead and puke out what "industry experts" recommend. But know that it won't impress anyone because you don't actually know what the business is doing / cares about / is prioritizing.
Convinced you need to only accept business questions?
I am glad.
Three Attributes of Business Questions.
We have done a pinky swear that you are going to start your daily web analytics journey by asking the business what questions they want answered.
I have to warn my budding Analysis Ninja that what you will typically hear is:
+ I want to know how much traffic is coming to our website
+ I want a conversion rate
+ I want a path analysis for our visitors (oy vey!)
+ I want to the list of top exit pages on our website
+ I want to know how many leads we got on website this month
+ Or give me a report that shows click-through rates of our home page promotions
All of these are requests I am imploring you to rebel against. They are not business questions.
Take a deep breath. Simple. With your eyes, not your mouth, say: "I really want to help you."
At this point what we really want to do is refocus the discussion and increase the likelihood that you can be something more than a reporting squirrel. With a twinkle in your eyes politely say:
We are executing a true Web Analytics 2.0 and Multiplicity in our company. In as much as we have access to many sources of data that can help answer What, Why and What Else questions quickly and efficiently.
Based on my expertise I can help you pick the right tool and metric if you could share the question you are trying to answer. What problem are you are trying to solve?
That will probably take all of 20 seconds to say (don't forget to twinkle your eyes).
They'll be struck by your sincerity, and shocked that you want to help this much.
The trick now is to make sure that you are able to recognize if what comes out of their mouth (or in words over email) is actually a business question and not a rephrased report request.
Business questions have these three simple characteristics:
#1. They are usually open-ended and at a much higher level, leaving you room to think and add value.
#2. They likely require you to go outside your current systems and sources to look for data and guidance in order to measure success.
#3. They rarely include columns and rows into which you can plunk data you already have.
If what you hear fails any of the above tests then you have to go back and try again.
Try not to be impatient or show off how smart you are or pick a fight. Keep the twinkle in your eyes, highlight what was different about what they said compared to the first time around, and then gently ask them a specific follow up question.
If you read the three characteristics carefully you'll notice that they are encouraging the best from your requestor (context, business priorities, problem framing), while at the same time encouraging the best from you (your knowledge of data, systems, analytical strategies).
That is the basis for the magic that converts you and me from data puking Reporting Squirrels to Analysis Ninjas who leverage custom reporting, advanced segmentation, statistics, surveys, competitive intelligence tools and to deliver reports with specific insights.
Guess what's the result?
People who just wanted data are now running around taking action based on your insights. Why? Because you did not provide data. You answered questions that were important to the person, tied to business priorities.
In rare cases your requestors might not even know how to reply when you ask them for business questions. Let me send you off with a little gift for the times when that will happen. . . .
Sample Business Questions for Ninjas & HiPPO's.
Look the person in the eye, still with the twinkle, don't forget the twinkle, and say:
My dear friend Avinash Kaushik asked me to share these sample questions with you. He said it would help us identify what's most important for our business.
If you hear the questions below you've hit the jackpot, because these are questions, have the three characteristics outlined above. . . .
- How can I improve revenue by 15 percent in the next three months from our website?
- What are the most productive inbound traffic streams and which sources are we missing?
- Have we become better at allowing our customers to solve their problems via self- help on the website rather than our customers feeling like they have to call us?
- What is the impact of our website on our phone channel?
- How can I increase the number of customer evangelists by leveraging our website?
- What are the most influential buckets of content on our website?
- If we could only do one thing to increase revenue on our website what would it be?
- What is the incremental impact of our display ad campaigns?
- Are we building brand value via activity on our website?
- Do fully featured trials or Flash demos work better on the website?
- What are the top five problems our customers face on our website?
- What is the cost for us to earn $1.00 on our website?
- What is the effect of our website on our offline sales?
Open ended. Force you to go to new sources. Don't contain columns and rows.
That's when you know you are on the right track.
It is quite likely that you will run into other business questions that might be more pertinent in your case.
But the theme that you are looking for is tough, highest-level business problems that you can help solve by analyzing the data you have (or data you don't have but will figure out how to get).
These are the questions that give you goose bumps. These are questions that give you a drugless high. These are questions that validate your decision to come into this field. These are questions that get you out of bed in the morning and feel excited to be alive. (Ok so maybe that is just me. But you'll see what I mean!)
Questions that rock your world, and coincidentally make for a truly data-driven org.
Parting Words of Wisdom from a Practitioner.
For the longest time in the web analytics world we have been content to do one of two things:
1. Provide the data we have in our applications in the hope that in the deluge of visitors, page views, referring URLs, time on site, and exit pages, there is something that marketers and business stakeholders will find of interest and take action on.
2. (Eager to please.) Take requests for reports, create them, and figure out how to email them or publish them on the intranet.
Reality is rather messier as a result of this.
Business leaders feels frustrated that they are not getting insights that they can act on. On the other hand this can't be fun for you. It can't be easy for you to hold the title of Senior Web Analyst and you are reduced to running reports.
Hence the most important foundational element of any effective web analytics program is to ask real business questions, understand those business questions, and have the freedom to do what it takes to find answers to those questions by using Web Analytics 2.0 strategies.
So. . . .
If you are the Business Honcho, bare your soul and share the questions (samples above) that keep you up at night or the priorities that you think are required to go out and win against your competitors (again these are not reports you want).
If you are the underling, seek to get a peek into the said soul and understand the strategic questions that the business wants answered. When you learn what the questions are, go get answers, one at a time. You will now be on your way to truly adding value to your company.
If you are a powerless underling, provide the reports, puke the data being asked, play the useless non value added game that you are being asked to play. But all the while know that real glory lies some place else. Be on a question quest at every given opportunity, with a twinkle in your eye.
Identifying business questions is a journey.
As you solve one set, the next will come up. Or you may be in the middle of solving one set, and suddenly that set will become irrelevant and there will be a new set.
This evolution and change is a sign that you are actually answering business questions and not just doing reporting, because business is always evolving and changing and you have to simply learn to change with it.
From the bottom of my heart I wish you all the very best.
Ok now its your turn.
How do you spend most of your time at your job or with your client? Providing data rich columns and rows and reports with pretty font, or answering business questions? If you have tried this strategy already what are the biggest barriers to you being asked questions rather than data? If a strategy has worked very well for you in this context what is it? Twinkle in your eyes? [I knew it! :)]
Please share your experience / feedback / pain / joy.