Sound Online Advice New Media Analytics and Metrics for the Radio Industry
Archive Receive Newsletter

AG News: Tuesday - 4/7/2009

If Radio Likes Numbers, Why Not Embrace Analytics?

You live by the book. You die by the book. So the radio industry saying goes. For as long as I can remember, nearly every station in medium to large markets has waited for Arbitron to publish its ratings so they can determine advertising rates and what needs adjustment in programming. Other numbers that seem to be bantered about as if they still mean a great deal are those you saw when Arbitron released its "RADAR 100" network ratings last month; there the claim is that now 186 million people 12 and older heard one or more network radio commercials in each week of this "survey."

With such a strong attachment to numbers, you'd think it would be a natural occurrence for the radio industry to embrace what it is doing online and to start wrapping the numbers of analytics and metrics around its web sites and the advertising they carry. Only (and I may be missing this), how radio uses the numbers supplied by server logs, analytics programs, and online ad buys is a topic I've rarely seen discussed. Radio trade publications seem stuck on presenting news of format flips, downsizing, and the ever-present danger of having to pay a royalty "tax" to record labels instead of giving you something to work with on the numbers front.

Knowing that numbers mean so much to those working in radio, let's go over a few which show - in detail - that what radio's been following is not as important as you think; and why the numbers the radio industry follows are going to shrink even further, relative to the price advertisers are willing to pay for airtime. The good news is we can use real numbers for this demonstration, those obtained from ad buys which have been analyzed to determine who gets future advertising dollars.

There is a caveat though. These numbers represent advertising in various radio industry trade publication web sites. But, you can know that this is exactly the style of assessment being used to decide if buying a radio schedule today is still worth as much as the industry thinks.

First, the setup. What follows are hard numbers from various ad buys that center around two similar campaigns. No names will be used, but elements of each will be discussed because that is a determining factor in problems plaguing traditional media today. The price paid - as measured by cost per thousand (CPM) - is being measured against new media.

I'll leave search engine keyword ad buys out of this and concentrate on what's called content placement, where the site content attracts the desired type of people. (Keywords are used to determine placement by matching them with a web site's content.) Over the past three months (Q1, 2009), $5,913.83 of advertising was purchased for this "content" slice of the campaigns. When viewed as a group, the CPM price paid across these "content sites" was $0.44. (That is a correct number, 44 cents per 1,000 persons reached.) This was not a shotgun approach. The buy was on web sites that delivered the desired type of person who is interested in the advertisement's offering.

Remember that $0.44 CPM, and watch it change as we talk about the different sites where these buys were placed.

Online Network #1 delivered 1,226,173 impressions, resulting in 173 "clicks" - which is better described as 173 persons ending up at the landing page I designate. It's my job to convert these 173 individuals into a sales by giving them: 1) a reason to click on the ad, and 2) once they've reached the landing page, a reason to purchase or register (whatever the action deemed "success" may be). In this case the click through rate was 0.01% - not good enough to continue the campaign, so I dumped it after spending a total of $38.45. Chump change, but enough to show me where this campaign was headed on this particular network.

Online Network #2 delivered 4,403,647 impressions at a cost of $1,490.47. It delivered 4,262 persons to the landing page at a CPM of $0.34. (Again, that's not a typo - it cost 34 cents to expose 1,000 persons to the ad.) The click-through rate, though a weak 0.10%, was still enough to continue spending on this campaign because we could determine the number of persons who followed the landing page through to our success metric. Ultimately, the cost-per-click for this campaign was a whopping $0.35.

There was one search engine where a content keyword ad buy was made in Q1. This "content" keyword ad buy differs from the over $18,000 spent on buying keywords for searches in Q1. Content is a different animal than simple keyword ad buying.

I spent $962.94 buying content placement. The result was 6,852,645 impressions delivered, 4,469 responses (clicks) received, at a CPM of $0.14. (Yes, that's a 14 cent CPM.) The CTR (click-through rate) for this campaign was a meager 0.07%, but strong enough to show successful metrics and make the buy worthwhile.

Now let's shift this just a little and bring in the "traditional media" side of the equation. To help, I'll define the traditional media types as those web sites operating under the premise they apply a cost for serving an ad based on their evaluation of what they think an ad on their web site is worth. (Note: Each of the examples listed below served ads that were delivered to the same types of audiences described above, and each was measured over the same period of time.)

Radio Industry Trade #1: Because this campaign just started in the last week of the quarter, only $142.84 was spent. It brought a CTR of 0.32%, at a CPM of $24.28. Cost per click on this campaign was $7.52. Respectively, the costs per click for each of the new media campaigns mentioned above were $0.22, $0.35, and $0.21 (no typo - 21 cents per click).

Radio Industry Trade #2 was slightly more expensive. After spending $1,500, here's what I ended up with: CTR was 0.35%, CPM was $59.21, cost per click rested at $17.05, and it ended up with 25,334 impressions (of which 88 traveled to our landing page).

I know there are a lot of numbers being flung about here, but let me give you one more set that's sure to draw your attention. Then, I'll close.

Radio Industry Trade #3 is a perfect example of why there are more media buyers moving to an accountable ad serving environment. Here's how this publication's web site performed with $1,671.88 of my money. The result was 17,733 impressions served with 25 persons going to our landing page. Total cost - hold onto your hat - CPM was $94.29, CTR was 0.14%, and it cost me $66.88 for each one of those persons who clicked.

Ad Placement Impressions CPM Cost CTR CPC
Online Network #1 1,226,173 0.03 $38.45 0.01% 0.22
Online Network #2 4,403,647 0.34 $1,490.47 0.10% 0.35
One Search Engine 6,852,425 0.14 $962.94 0.07% 0.21
Radio Industry Trade #1 5,883 24.28 $142.84 0.32% 7.52
Radio Industry Trade #2 25,334 59.21 $1,500.00 0.35% 17.05
Radio Industry Trade #3 17,733 94.29 $1,671.88 0.14% 66.88

Here is but one of many spreadsheets I use to determine the value of a campaign. It's unreadable for a reason, but you can see that it is very organized in how its numbers are presented.

Know that there are another half-dozen spreadsheets used in completely tracking an advertising campaign. But, when the annual spend is in the hundreds of thousands of dollars (or even the thousands of dollars), it is well worth the minimal effort required to keep these updated. With a little knowledge in creating charts, the data is presented in a graphical interface that gives the client a one-look view of how their campaign is performing.

These are cutting-edge advertising metrics and analytics that are directly applicable to the radio industry, either through use on web sites or - as I've demonstrated here before - in over-the-air campaigns.

You live by the book. You die by the book. If so, why is it taking the radio industry so long to grasp that "the book" no longer just refers to the numbers in its audience? These days you have to glance at what's presented above in advertising dollars. If you embrace the old book you have to give a hug to the new numbers in the room, the analytics and metrics of advertising campaigns.

If the radio industry fails to keep pace with these, radio advertising revenue numbers will just continue going down... to a point below that $0.44 CPM used to buy a like-minded online audience.

About Contact Indie Artists Radio Stations Audience Data Privacy

President, Audio Graphics
Ken Dardis
Online Since January 1997

Radio Industry News
All Access
Holland Cooke Media
Radio Ink
Radio Business Report

Search Audio Graphics

Search Web