Radio myth-buster #4

Co-Founder and Insights Director


Thanks to recent RT poster 'Tracking schmaking' for inspiring this article.

There are many myths surrounding tracking, a mystique sometimes promoted by those in the reporting loop. That in the absence of hard evidence to the contrary, takes on a life of its own. What T-s called "that dark art of intangible numbers".

Read on for 5 Truths About Tracking

1. Tracking is exactly that. Not a standalone census frozen in time, but a continuously-moving measurement of station share and other KPIs (Key Performance Indicators), as a management reporting tool. No different to revenue, inventory-yield, profit margin and so on.

Not exclusive to radio, but a widely used technique by major brands to keep tabs on all sorts of data important to their business such as marketing awareness, image perceptions, and product satisfaction. Google "brand-tracking" for more background.

2. Why the Secret Squirrel? Just as it would be poor corporate governance to allow financial performance figures to be let loose in the corridors and beyond, it's just as irresponsible to have your product/promotion/marketing KPIs up for gossip on the forever active radio grapevine.

Executed properly, year-round, tracking is a major investment. Not the internal cottage industry when it started 30+ years ago on sales phones after-hours, to collect music research.

At the same time, it's also weak and negative people-management, especially of creative teams, if dark hints are dropped about tracking, as some kind of mind-game to supposedly drive better performance. This tactic usually backfires, as sphincters tighten and creative risk-taking gives way to safety first.

Any internal heat about tracking should stop at the Content Director and go no further. Even then, the CD shouldn't spend all day obsessing about the scoreboard (see here).

3. Tracking can't predict survey results. "Tracking has been known to be horribly wrong". A view shared by many over the years, but depends on what you mean by "wrong".

Firstly, all forms of radio research carry more caveats than a mining heir's inheritance.


Radio's great strengths of universal access, mobility, and flexibility, are also the most frustrating for researchers. How do you pin down, even for a moment, such an elusive type of consumer behaviour that's like catching smoke?

I often mention people-meters, for the simple reason that it's the closest method yet of capturing the reality of radio listening, only glimpsed through the veil in focus groups and traditional surveys.

Even then there's controversy. Sample size a major point of contention in the US, as sample costs big money, and when you're dealing with all those tiny units of data, you need it to be robust.

So expecting one elusive methodology to precisely predict the outcome of another is unrealistic.

The best use of tracking is trending, leading to the next home-truth …

4. Look at the trends not the numbers. Especially not decimal points :). The longer the trend, and the larger the accumulated sample, the better. A number in isolation means nothing. If you feel like plucking, take up the banjo.


"We've got an 8.2 this week" … means what, compared to when… last week, last month, last year? What's the long-term trend, up or down?

Which format are you in? CHR is driven by cume, rock by TSL, talk radio by retirees with time to fill out diaries. Tracking, whether by phone or online, has to be viewed through the format lens. The method of data collection also influences the result.

It is possible during a survey period to gain a reliable feel for the market, when your KPIs such as share, cume, dayparts and image perceptions are set up properly on long-term graphs. Where movement is meaningful.

Over time, your "gut" becomes more and more educated about what to ignore and what to take seriously. What is just "noise", and what will impact the survey.

Context is everything. What's going on in the market during that period, not just on radio? Because life drives radio listening. e.g. extreme weather, season change, election campaigns, school holidays, major sport. As well as inside-radio influences such as the hit music cycle, major promotions&marketing campaigns, and personality changes.

Parts of Surveys 6  and 7 in Australia, will be influenced by several of those life elements during September.

5. Tracking is not the end game. If you make the investment in some kind of KPI tracking, you have a platform for added benefits, and not wasting a contact.

e.g. deeper, more flexible analysis of the dataset to learn more about listening behaviour, additional studies, and recruitment for music research and focus groups.

Research-permission is the lifeblood of quality feedback these days, when pure "random" research is cost-prohibitive, giving way to pre-recruited "permission-panels".

The Takeaway

It's understandable that tracking in the past has acquired its own mythology.

In commercial radio, a lot of revenue hinges on ratings. Which are also a scoreboard for creative achievement, self-esteem and career prospects. So anything involved with affecting and predicting those ratings brings inevitable emotional baggage.

What will happen to traditional tracking, when evolving ratings methodologies capture more "real-time" listening and report results more frequently? And there's more integration with measurement of other media, for a total picture of the media consumer experience. Does it become redundant?


In my view radio research in the future will be more and more based on social methods, permission-panels and brand communities, allowing more in-depth and creative questioning and analysis. Tracking target segment KPIs, not the total market trying to predict ratings, and providing more actionable support for content, promotions and marketing.


Eriks Celmins is Managing Director of Third Wave Media, Adelaide-based consultant for media research, strategy and content.


Comment Form

Your email address will not be published.

Recent comments (0)
Post new comment


See all