Every week you can find a new survey report that tells you things that aren’t very interesting: How many recruiters use LinkedIn, that digital advertising continues to overtake print, that an increasing number of companies are using CRM systems.

These reports aren’t interesting because a) they tell us stuff we already know or can guess, and b) they don’t tell us anything we can do anything about.

If you want to write a survey report that will grab readers’ attention instead of putting them to sleep, these three strategies can help:

1. Say something counterintuitive. For example, it’s been a given in business for over 20 years that good risk management starts with tone from the top. I’m not giving you this one for free (unless you ask me), but I’ll bet 10 cents to the dollar it isn’t true. I will celebrate the day someone publishes a survey that disproves this ancient canard.
   
2. Describe what distinguishes the leaders from the laggards. A survey finding that 39% of companies use cloud CRM is of no interest to anyone except CRM vendors. That 80% of the best companies – as rated by their salespeople and, perhaps, their customers – do, versus 20% of the worst, might be of great interest. Unless you take the trouble to find and reveal meaning in the numbers, they’ll be, well, meaningless.
   
3. Explain the why. Surveys reliably reveal the what. For example, how many companies use demographic data to target their direct mail (still a powerful marketing channel, by the way). But they rarely tell you why. (Perhaps it’s because today’s databases allow for more precise targeting.) Your data will be much more relevant to your audience if you add the why.

All very well, you say, but how can I do these things? Here are three ways:

1. Start with a hypothesis. Don’t write your survey questions until you know the conventional wisdom you want to dispute. Even if you don’t quite overturn the conventional wisdom (tone at the top, for example, isn’t irrelevant; it’s just not the most important factor), the exercise of thinking about it hard enough will give you enough good questions to get to the heart of the issue and tell a more interesting story.
   
2. Ask differentiating questions. The most revealing queries are those that separate respondents that are successful from those that are not. For instance, if you are examining how well companies manage risk, ask them how much their company has lost by misadventure in the last three years and what the company’s revenue is so you can normalize for size. Then ask them how well they think they manage risk. That’s a qualitative self-assessment, of course, with all the hazards that entails. But it can be a useful double-check. If the qualitative varies significantly from the quantitative, it might give you a story about self-delusion among risk managers.
   
3. Include interviews about what works and what hasn’t. If you anticipate that more and more companies are moving applications to virtual private clouds, for example, make sure the interview questions ask those who are moving to the cloud why they are doing it. You can get a superficial rationale by asking multiple-choice and open-ended questions in the survey. But an interview will give you infinitely more depth, and quite possibly a supporting story or case study you can feature.

Too many companies think a survey report all by itself will be something people will want to read. They’re wrong, especially when their surveys ask questions for which they and most readers can predict the answers. Asking those kinds of questions certainly mitigates the risk of coming up empty-handed. However, it also almost guarantees that the survey will confirm what everyone already knows, and what good is that? However, with a well-designed plan and a sincere desire to discover something new, you are almost certain to return from survey-land with insights no one else has. That will grab reader attention, as well as inform future business decisions.