To be most helpful on ridiculous questions, just say no

When someone asks you “what is the average click through rate for Google,” or, “what is a good conversion rate,” find a kind way to let them know that the question is unanswerable, or that the answer is counterproductive. For years, we’ve been answering unanswerable questions in an effort to appear smart and helpful, or out of fear that the organization will lash out at you for not being prepared, but answering these questions damages you and it damages your business.

Our role is not to serve the needs of people with questions, but to help people make those questions and the organizations asking them smarter over time. One of the best things we can do is try to understand the motivation for a question like this, rather than just answering the question. If someone asks you what percentage of people click on Google paid listings vs. natural search, try to get more information about their needs, rather than simply answering the question. Of course, there is an answer; there is a percentage of clicks on each one, but does that percentage translate equally to The King’s Speech Showtimes, Atlanta weather, how to tie a bow tie, and brown leather shoes? Of course not.

It’s likely that what they really want to know is whether an investment in SEO or paid search is worth it. What would we expect in terms of traffic, how does that translate to sales, new customers, ROI, etc. versus the costs involved? What should we target first?

These are interesting questions, and ones you can be very helpful with.

But when we try to be helpful and answer questions about the “right” this, the “best” that, or the “average” such-and-such, we slow down our organization. Why? Because information is about action, and this type of information gets us to action more slowly, in some cases. More often than not, you’ll find that these questions are about comparisons: how are we doing vs the right, the best, or the average? Yes, identifying these gaps will cause action, but the answers to these types of questions are “numb”: we can see a gap between us and the norm or the best, but we can’t really identify its specifics or see a clear course to improvement.

As an analyst, you will be asked all sorts of whacky questions. When you get them, find a good way of helping mold these questions into more useful, insightful, and actionable ones. When I say, “just say no,” I don’t mean to not be helpful. I mean be incredibly helpful, but about the root of their concern, not the surface.

The first few times you do this, especially if you haven’t before, it may get a little ugly. But be confident that you’ll save face: you have the information they need, and what you have is so much better than they expected. Engage them and tell them that you think behind their question is some really interesting stuff. If you need a day or two to dig in, ask for it, and tell them they won’t be disappointed. They won’t be.

What happens next is that people will no longer come to you with a list of surface questions. They will come to you with their real problems. And that’s a much better version of our jobs.

What if it blows up in your face? All I can tell you is that it’s a really good job market out there, and more and more organizations are looking for thoughtful analysts, rather than question answerers. It’s getting easier to make this the job you want it to be. Don’t regret not trying out of fear.

No Trackbacks


  1. Absolutely true. With the right delivery, this will work in your favour. You can eventually build trust by explaining that their question is stupid and you would be stupid to answer it (paraphrase please). Time is valuable – if they want a metric, they don’t need you, they can find it in an out-of-the-box report. If they want to understand what is happening on a site, then you can step in and save the day. In the long run, they’ll thank you for it.
    And the black eye will heal eventually.

    Posted February 28, 2011 at 11:34 am | Permalink
  2. John

    Generic versions of approaches that are useful in getting to where Evan wants you to be:

    “That’s an interesting question, would you mind telling me how you came up with it?”

    “Wow, other people have asked that too. Why does everyone want to know that?”

    “I’m just a simple analyst, but I’d like to learn more about the business — can you tell me what decisions you’re trying to make with the data you asked for?”

    “I don’t think we can do that, but if you tell me what you’re trying to do with that data, I might be able to come up with an alternative.”

    Posted March 2, 2011 at 12:37 pm | Permalink