Dear Diary …
To a radio person this traditionally says Survey Diary. Where are they placed, where do we send the street team, who's filling them out, will I get one myself (it happens!), will they fall our way this time, and so on?
But there's another diary, totally in your control, that you should be just as concerned about.
Is There a Doctor in the House?
Every survey day Radio Today reflects the flurry of stats, snap-analysis, press-releases and emotions, filling the boardrooms and corridors of radio stations around Australia.
It's all part of the energy, theatre and immediacy of radio that we're addicted to, but hardly conducive to a cool-headed review of the implications for your network, station or program. Where the programming doctors need to take over from the spin-doctors.
And there's a great chance of survey-specific myths being created and perpetuated, that at best are just noise, but at worst, reputation and career-damaging.
Margin of Error
When your key number might be plus or minus 3% (or whatever) of the reported result, it's seen as an inconvenient truth, something only for research-geeks. That gets in the way of the great gladiatorial game of declaring winners&losers on the day.
I can't wait for (a) people-meters and (b) radio ratings presented as part of a much bigger 360° picture of people's total analogue+digital media consumption, to end the simplistic scoreboard.
BTW it's not any fault of the Nielsen survey implementation, which is top quality and audited. It's how the data is used, being technically called an "estimate", not an absolute.
The Market Diary
Is the one you should be keeping at least every week, in or out of survey, of major events that could affect your audience share, like a Facebook Timeline.
Not just radio-related, such as personality moves or promotions&marketing spikes, but anything in your listeners' environment that could influence their choice of station at the time. Whether it's election campaigns, school holidays, blockbuster news, extreme weather, sports finals etc.
Very easy in the heat of survey day to see the result through the lens of just the last few weeks. When in the rolling-wave system the real action can stretch back over 12 weeks including time inbetween waves.
Was a great example, reflected in our discussion threads over whether the election was a factor, and also considering the Ashes, Royal Baby, peptides and so on.
I mentioned in my Tracking Myth-Buster article in June here how Surveys 6 and 7 could be influenced by life elements. As it happens Survey 5 from May to August had a lot going on as well.
Radio is both habitual, and subject to continuous surfing.
Add to that, survey diaries are placed over 10 weeks, so your own "cume-sample" … how many diaries actually contained any listening to you … is much smaller than the total sample and spread over that period. Needles in haystacks!
So you just can't bet the farm on, and over-interpret, one result. Like tracking, it's the trend not the one-offs that matter.
Yes, media-buyers are mostly short-term, bulk numbers-driven, but programming strategy cannot be plotted the same way. It's all about long-term trends and multiple inputs to be carefully sifted through, away from survey day.
The Best Mix of Old&New
To make sense of it all, I suggest starting with your Market Diary or Timeline. What really happened over those 10-12 weeks that could have any bearing on the life of your audience?
Start broadly, don't jump to early conclusions, but consider the ripple effect of events through your target. It may not always be obvious, but in a fast-moving culture where everything's connected, you should try and picture a day-in-the-life of your target's family&friends, and how the events may affect them.
This is where regular conversations with your listener community via social media, and LABS using InsiderFocus (blatant plug!), can be useful for lifestyle insights. Not hard stats, but thoughts and feelings that will ultimately impact your share, because radio listening is driven by emotion.
If you're not tracking social media analytics … why not? They're not the complete answer, but now a vital part of KPIs to place alongside your Market Diary, looking for correlations.
Lay any tracking over the period of course, and don't be fazed by detailed differences with the ratings. There will inevitably be both similarities and differences – again, look at the trends.
Finally, and only then, pull the survey apart as relevant to your brand. Usually it's the other way round, what's the survey result, and what else do we know?
But try turning it around, so you're starting off in the hearts&minds of your listeners over the Timeline. Then review the numbers while in that mode, and you're in a much better position to take actions on what's real vs. what's merely a stats wobble.
Survey analysis and strategic decision-making is as much an art as a science. You have to use "soft" information and experienced gut, because numbers without context are guaranteed to lead you off course.
Eriks Celmins is Managing Director of Third Wave Media and InsiderFocus, consultant for research, strategy and content. Find out more here.