Signals and Noise

If you work in media relations at a company covered by the national press, it can feel like you are constantly putting out brush fires. As fast as press releases go out, media inquiries come in. Earnings and stock prices. Lawsuits and layoffs. ESG and COVID. They all make news.

In terms of media outputs, companies can efficiently track how much news gets published. The nagging question for companies covered in the national press, though, is which of those news stories do people actually see or hear? Which stories impact the firm's reputation? Which stories impact sales?

We've done this analysis dozens of times now. Here's the short answer. On average, for a national, publicly held company, about six news cycles a year generate enough coverage to move the needle. That's the Noise in the system.

And about half of those stories actually do move the needle. Those are the Signals.

A critical first step in managing Headline Risk, then, involves our ability to separate Signals from the Noise.

It's not difficult to do. Every semester in an undergraduate case studies class at N.C. State, we analyze news coverage published about 20 publicly held companies over the past year. We start by simply multiplying daily news volume by average daily tonality, creating a News Stream bar chart that looks something like this:

The triangles denote seven Big Stories last year for the company, in this case an innovative global automotive manufacturer.

We could do more math, but the pattern in the data hits you right between the eyes. The big spikes represent the Big Stories.

Then we overlay that Big Story bar chart with a line chart that depicts consumer perceptions about news people actually saw or heard. Buzz.

Buzz is tracked by YouGov, which on a daily basis poses a simple and insightful question to its online panel: Have you heard anything Positive or Negative about this company? By combining our News Stream chart depicting news volume and tonality with the Buzz chart, we end up with a Signals and Noise analysis that looks like this:

So what are we looking at here? Of the seven Big Stories we identified, four correspond to no discernible shift in Buzz. Those stories are Noise. In this case, the Noise consisted of two product announcements and two quarterly earnings reports. We'll have more on that in the upcoming Earned Earnings blog.

Three of the Big Stories, though, did correspond to discernible shifts in Buzz. Those stories are the Signals. Again, we can wrap some math around this, but for now we are sticking with that inter-ocular trauma test. What kind of news generates Signals? The positive Signals involved a quarterly earnings report and a significant spike in stock price. The negative Signal involved a product failure that generated national news.

So what does this tell us? For one, when it comes to identifying news that moves the needle for business, traditional media monitoring generates a significant volume of false positive readings. In our Signals and Noise analysis this semester, 20 national companies generated 116 Big Stories. About 60% corresponded to no discernible Buzz with consumers. Said another was, six in ten stories generated false positives.

Of the Signals, the four in ten stories that did break through:

  • About half were triggered by press releases -- that's headline risk that can be managed.

  • About a third corresponded to negative shifts in Buzz -- that's headline risk that can be mitigated.

Why is it important to separate the Signals from the Noise? Let's use a severe weather analogy. We could track storms based on wind speeds, barometric pressures and precipitation rates, and identify the most intense storms. Or we could track storms based on economic impacts from flooding, property damage and deaths, and track the most severe storms.

In business, we need capabilities to track, manage, mitigate and ultimately forecast the most severe media storms. Once we have the ability to separate Signals from the Noise, we can begin to gauge the impacts of news cycles on our business. That's especially critical if the Signal is strong enough, and becomes the catalyst for a crisis.

That's next. The Crisis Life Cycle.

Back to the Whiteboard.

If you work in media relations or risk management at a large national company

and would like to work with one of my students to separate Signals and Noise,

let's connect. I can be reached at

14 views0 comments

Recent Posts

See All