Are You Being Duped By Digital?


(By Matt Nystrom) This is an open letter to my colleagues in the radio industry. If you’ve spent any time working in advertising, you’ve probably heard some variation of the century-old quote, “Half the money I spend on advertising is wasted; the trouble is, I don’t know which half.” Despite its dubious origin or veracity, advertisers and media sellers have long pursued attribution models that aim to eliminate advertising waste. To that end, it has become easy to gorge on a glut of up-to-the-minute marketing insights.

Digital marketing metrics, in particular, have created a Pavlovian response in advertisers conditioned to see immediate insight and results. Appetites for on-demand information have traditional media sellers feeling compelled to measure the efficacy of their products in ways that undermine their core strengths.

Case in point, for us working in the radio industry, it has been hard to miss recent chatter about digital attribution programs designed to track the minute-by-minute impact of radio advertising. Pundits and executives across the industry have been quick to herald the arrival of such programs as a way to prove radio’s worth and demonstrate real-time ROI.

In simple terms, these attribution programs simultaneously hook into a radio station’s automation system and a client’s web metrics. Every time a commercial runs, the station monitors that client’s website traffic for a predetermined period (usually eight minutes). Any activity observed during that window is attributed to the radio commercial.

The fundamental flaw in this attribution method is that it’s impossible to know precisely what portion of a client’s web traffic was indeed a direct result of a particular radio advertisement. Furthermore, many clients run their radio campaigns across multiple stations in a market. If two or more stations run the same commercial at roughly the same time, which station gets the credit?

Telling our clients that radio’s ROI can be measured in real-time misrepresents how radio works and how it’s consumed. Radio listeners are often on the go and tuning in on a diverse set of devices. When you consider that the average commute in the United States is nearly 27 minutes, do we really believe that listeners pull over on the way to or from work to visit a website in response to radio ads? If they don’t, should we assume those commercials aren’t working?

What about industries with long sales cycles? Is there no value in reaching listeners who aren’t intenders today, but will be weeks or months from now?

A radio market manager told me of a client who refuses to run commercials outside of business hours out of fear that the phone would ring and no one would be around to answer. Although it’s encouraging that the client is certain radio commercials work, it illustrates the way that ad buyers are being conditioned to expect an instantaneous response from all advertisements.

Gauging radio’s ROI using minute-by-minute measurements is a dangerous cocktail that will harm our industry. If this trend becomes standard practice, we’ll face the unintended consequence of turning every campaign into tight sales-activation messages that ultimately sabotage radio’s most effective application: brand building.

Across the pond in the United Kingdom, Les Binet and Peter Field have turned out incredible research for IPA that has demonstrated the cumulative effect of long-term branding campaigns. In a 2013 report titled “The Long and the Short of It: Balancing Short and Long-Term Marketing Strategies,” the pair establish what they call the 60:40 rule, which states that the optimal marketing-message mix is 60% brand building and 40% sales-activation.

Binet and Field found through extensive research that the sales effects of brand building are cumulative and increase slowly. Short-term sales-activation messages bring a sharp lift in sales, but the effects decay quickly.An update published in 2017 declared:

“Activation effects are relatively easy to measure, because they tend to be big, immediate and direct. In the short term (six months or less) they tend to produce the biggest sales responses. … Because the effects of brand building only become apparent over the long term, short-termism is dangerous. It can lead to excessive activation (which is inefficient), and under-investment in the brand (which can lead to long-term decline).”

Even if marketers are patient or diligent enough to adhere to a 60:40 marketing mix, it’s not a given that their employer will be in it for the long haul. When you consider that the median tenure for CMOs at the top consumer brands in the U.S. is only 31 months, it’s rare for marketing executives to fully implement and reap the rewards of long-term strategies.

Quick hits from sales-activation are important to harvest the fruits of your brand building, but it’s not a sustainable strategy. Mark Ritson sums up the perils of not following the 60:40 rule this way:

“Marketers are increasingly short-term in their focus. They spend money on immediate activation rather than longer-term brand building. They opt for bottom-of-the-funnel tactics because in a one-year time period that will always pay better. … Too much time spent picking the low-hanging fruit means less time watering the tree. Eventually, the tree stops growing.”

The radio industry cannot allow the increase of new competition to turn us into bad or lazy marketers. The uneducated demands of radio clients should not compel us to misconstrue the long-term benefits of radio’s core strengths. Radio works over the life of a campaign to build emotional connections that last far longer than eight minutes, eight days, or eight months.

Building a healthy brand is much like trying to get your body in shape. Meaningful improvements in your physical health are the culmination of a number of factors such as eating well, regular exercise, and getting enough sleep. The business impact of just one radio commercial is roughly the same as the health effects of having a salad for lunch. Yes, salad is a good choice, but the results of healthy living are observed only over time.

It’s the job of radio sales executives to educate advertisers about how to use our medium effectively and demonstrate how to measure a return on marketing investment. Well-crafted branding campaigns running at a high frequency establish a top-of-mind awareness that contributes to an increase in demand, market share, brand recognition, and customer loyalty, as well as a decrease in price sensitivity.

When a consumer visits a company’s website, the direct referral might have been a search result, display ad, or online promotion. Good marketers know that long-term traditional-media exposure can have a huge influence on the behavior resulting in that user’s visit.

The hard truth is that the radio industry has not done a good job telling our story and establishing our worth. We’ve allowed outsiders to brand us as old-fashioned, ineffective, and dying. Worse, we’ve stood by while digital marketing companies have taken an oversized share of credit for their part in sales-activation. It’s time to reclaim our reputation as the number one reach medium, the best way to connect with audiences on an emotional level, and a powerful tool for unaided recall.

Stop measuring your effectiveness by the wrong standard. Don’t feel compelled to fit radio into the mold of the latest marketing trend. You’re doing a disservice to your clients if you’re not showing them the value of building a brand. Quit allowing your clients to gorge themselves on the sugar high of short-term activation and show them the value of balancing their marketing diet with messages that will have a sustainable, long-term impact on the health of their business.

Matt Nystrom is the VP of Digital Media for Saga Communications and can be reached at [email protected]


  1. I had an advertiser receive a call from an 800# in a radio ad that had not aired anywhere but Rush Limbaugh 12 months prior to the call. Talk about HALO EFFECT !!

  2. I agree that attribution doesn’t make sense for radio. Your premise is solid However, what’s missing in this discussion is that most (I’m being generous) radio advertising sucks. AEs are not trained to be brand marketers. So they crank out radio spots that are formulaic, relying on good old “yell and sell” tactics. Radio has a huge opportunity to shine in the digital era as people gravitate away from commercial TV. Learn how to tell stories and engage listeners with emotion. Facts tell, but stories sell.

  3. Matt, well said, well researched, well done. You know the best way to help a client quantify the results of a campaign? 1) Have a well trained AE ask the right questions starting at CNA – “how will you determine the success of this campaign.” 2) Have a well trained AE create a long-term campaign that they believe in so much they would sell it to their grandma. 3) Have a well trained AE continue to communicate with the client after they close the deal and find out what is working and what isn’t. 4) Have those hard conversations with the client upfront…educate them about how our medium works best. Don’t be afraid to tell them they need to spend more to see better results. Don’t be afraid to tell them their copy won’t work. Don’t be afraid to turn their business down if you don’t believe they’ll be successful.

    Radio owners, train your people well, invest in them, trust them, and the money will follow. You don’t measure the success of a campaign through one narrow metric – and that’s exactly what attribution is…an 8-minute window of Google Analytics.

    Thanks again for a very well thought out article.

  4. The systems and tools exist for radio to measure radio-attributable advertiser sales results on a daily, weekly, monthly or annual basis – for any radio station’s advertisers. Using a number of them, it beats me how our radio industry continues to ignore our advertisers’ requests for measurement.

    The systems and tools out there generate more documented, radio-attributable sales for advertisers and more premium rate, 52 week sales and profits for radio per hour of prospecting, preparing, presenting, closing and servicing.

    The branding vs sales activation ratio of radio advertising dollars is easily validated: only 1-2-3-4% or so of most any advertiser’s prospects are “Now” buyers. The other 96% or so are “Future” buyers.

    Via replicable systems that produce customized, consumer-needs-focused advertiser strategies, tactics and tools, the radio branding campaigns help capture our advertisers’ prospects’ contact info. Then, these ultimate consumers’ interests can be further identified and fed as they’re continuing to be marketed along their consumer journey by radio, email, other digital, direct mail, etc., until the prospect says “Put a fork in me, I’m done, and ready to buy”.

    That our clients spend more on advertising than they invest in their retirement funds is telling. As they’d fire their financial planner if they didn’t get a statement each month showing how much they made, why expect any less from a larger investment in radio?

    How much of a competitive advantage do you have when you say to the client “How much money did make you last week / last month / last quarter / last year? With me, you’ll know. We’re good to start next Monday or the following Monday – how soon do you want to see the results?”.

  5. Thanks for starting the discussion on radio attribution, Matt. While the pot is clearly being stirred, I think you raise some interesting concerns that newbies often have as well. But for those broadcasters who have pioneered attribution programs and helped thousands of advertisers optimize their spend, the concerns you have do not hold practical relevance.

    For full transparency, I run LeadsRx, one of the two vendors doing original research in this area of broadcast attribution (Analytic Owl being the other). We’re extremely proud to be working with leading-edge broadcasters to quantify the impact of radio… not the FULL impact, but impact none the less. We’re all working together as a team, constantly improving the metrics we use, enhancing the methodology for obtaining these metrics, and presenting results to advertisers in actionable ways. Baby steps lead to bigger steps, and that’s a good thing for any industry.

    Your quick summary of how broadcast attribution works misrepresents the sophistication of the today’s methodologies. We are not limiting attribution windows to a preset eight minutes, and all web activity during a window is not attributed to radio. There are numerous filters and knobs to adjust that fine-tune probabilistic models to closer capture true consumer behavior. And, with spike analysis curves, we can clearly defend these approaches while also showing why 10 minutes is better than eight, how drive time impacts digital response, and when the “second breathe of radio” is present. Frankly, as a vendor who also measures digital attribution, I sometimes feel our broadcast attribution is more accurate than digital. Ad-click tracking parameters get mis-used more often than you think creating many false-positive results.

    Regarding your comment about industries with long sales cycles… I, too, believe this group is important. The auto industry, for example, experiences long cycles, and while measuring immediate digital response to radio spots is important, so is measuring how many cars are eventually sold because of those spots. Broadcast attribution actually does this.

    But the real evidence that broadcast attribution is not a “disservice” to advertisers (as you stated) can be found by speaking with the advertisers themselves. When you do, you’ll hear the stories that we hear: advertisers who finally know which ad copy delivers the best result, advertisers who expand to new markets because they can now measure the impact; advertisers who no longer struggle choosing dayparts or station format, advertisers who discover radio produces twice the lift as their digital spend, and advertisers who finally have an industry benchmark against which to measure themselves. Then, speak with the innovative broadcasters who use attribution tools to compete in an increasingly digital world… broadcasters who are seeing double-digit growth in ad spend after clients optimize based on attribution results. These are happy people.

  6. It’s a shame these so called attribution companies are taking advantage of an industry trying to prove their worth to advertisers.

    If the data was quantified like this from the start, it doesn’t matter how many responses to ads were tracked in present day.

    Matt, your article is really eye opening.

  7. Actually, Matt, broadcast analytics can help demonstrate the power of radio in both brand building and short term response. The 8 minute measurements roll up to show an advertiser how much more web traffic they get over time when they are on air versus off air. Very powerful stuff and really resonates with advertisers who might buy based on research, but then expect proof of performance.

    At industry leader AnalyticOwl, where I serve as General Manager (with 31 years of prior ad sales experience), we actually encourage sellers to talk about the intangible value in addition to the tangible value of radio…the “halo effect” of hearing a spot over time and its impact on brand awareness and purchase intent which eventually manifest themselves in a website visit.

    And we think about morning drive more broadly than just the driver in their car for 27 minutes…when someone’s alarm clock goes off in the morning playing their favorite station, what’s the first thing they do? Grab their phone. When they’re eating breakfast listing to their favorite station, what are they doing? Staring at their phone. When they get to the office and tune in their favorite station, they have a phone and computer in reach. Plus, aside from drivers, many people listen to radio on mass transit, while walking, waiting in line for coffee, etc.

    Matt, broadcast analytics might not excite you on the digital operations side, but it is making a world of difference to broadcast sellers and broadcast advertisers alike. Many of your other concerns like overlap are not major flaws and can actually be explained. I would love to give you a private demo to better educate you for a follow up article.

    Also, I say ignore what marketers want at your peril. Ask any seller how many calls get returned when they leave a message or send an email about some “exciting research”. Broadcast attribution is a game changer, and can be used not only for short term insights (A/B testing for creative, for example…you can’t deny that is a valuable service for an advertiser) but also to demonstrate to overall power of radio.

  8. When Matt sez, “It’s the job of radio sales executives to educate advertisers about how to use our medium effectively and demonstrate how to measure a return on marketing investment. Well-crafted branding campaigns running at a high frequency establish a top-of-mind awareness that contributes to an increase in demand, market share, brand recognition, and customer loyalty, as well as a decrease in price sensitivity”, I suspect he is also as aware that the radio industry, itself, needs to understand these strategies, as well. When I read, “…well crafted campaigns….”, I get tingles.

    I commend Matt on his well-informed entrance into the field of educating radio – a more important exercise, perhaps, than bringing advertisers up to speed.

  9. This completely misses the point that advertisers are leaving Radio in droves because it cannot provide accountability or proof of performance without these types of approaches. Frankly, radio NEEDS to ‘play ball’ and provide some tracking, even if it is limited in scope.

    Looking for repeatable patterns is a good thing. “Every time a radio spot airs, we see this type of spike”. Over a statistically-significant period of time and number of ad spots, true patterns emerge that take into consideration all of that other traffic and marketing.

    It is interesting to see a VP of Digital essentially telling broadcast Radio folks to ‘stay in your lane’ as a branding-only medium. Radio can be BOTH a Direct Response and Branding medium – even at the same time!

    Like any reporting methodology, these types of measurement shouldn’t be the only factor taken into consideration. They should be included with several other forms of accountability and reporting to make an informed choice. A little human oversight and wisdom goes a long way.

    • Providing a form of tracking of some kind is ideal, yes. The problem is, at the local level, often times these attribution metrics are used as the only factor to prove radio’s worth by companies that sign a client onto the attribution program. And products such as the one referenced in the article aren’t looking at “statistically-significant periods of time” – they’re using arbitrary numbers as a reference point and simply assigning attribution in that. There’s no analysis of historical performance. It’s simply “we ran an ad and in the X minutes following, here was your traffic.” Attribution like this would be much more palatable if presented like this: “after an analysis of the prior three months of web traffic to your site with no marketing efforts, we saw a lift of 20% in your total web traffic in the three months following when we executed your on-air campaign.”

      If your client sends an email to their subscriber list in the minutes leading up to a commercial, these attribution programs will take credit for any of that traffic generated during the X minute window after the ad runs. This particular brand of attribution is disingenuous, and we’re doing our clients a disservice steering them towards this method of tracking radio’s effectiveness.

      • Actually, Brian, AnalyticOwl does give you the opportunity to show the type of lift you describe. And we have analyzed 2.4 billion responses to radio spots, so there is statistical significance to what we do. I see you are also on the digital side at Saga…would love to get on a call with you, Matt and ideally some of your sales leadership to discuss how other companies are successfully using analytics.

  10. This is too smart for Radio Ink but, wow! This is like Moneyball and Limbaugh and Howard Stern mixed in a blender. One is all data while the other two are just dang good at what they do and probably don’t spend any time thinking about which topics will spike attention at which times. Well, well, written.


Please enter your comment!
Please enter your name here