Ignoring co-viewing? Your campaign reach is bigger than you think

When advertisers plan and measure campaigns, they often rely on an overly simplistic equation: one screen = one person. But in reality, media consumption doesn’t work like that. Whether it’s a family gathered in front of the TV, a group of friends watching a game, or a couple enjoying their favorite show, ‘co-viewing’—that is, more than one person around the same device—is a very common behavior. And it plays a crucial role in the impact of advertising campaigns.

And yet, most measurement methodologies either completely ignore it or struggle to capture it. The result: campaigns are achieving a higher reach than reported, frequency is being miscalculated, and the true impact of advertising remains hidden. With so much at stake, it’s no surprise the debate is heating up.

At FLUZO, we often approach measurement differently from industry consensus—and co-viewing is no exception. Yes, we measure it. But probably not in the way you’d expect. Many new clients are surprised when we tell them not to worry: FLUZO data already incorporates co-viewing implicitly. The conversation usually goes something like this:

• Client: “Wait a second, are you telling me your data already includes co-viewing?”
• FLUZO: “Exactly.”
• Client: “Prove it!”

Let’s break it down. But first, let’s make clear why it matters so much.

Why measuring co-viewing matters more than ever

‘Coviewing’ is not a marginal phenomenon. It’s an essential part of how audiences consume media today. Over the last decade, the “big screen” has evolved into a social epicenter around which families, friends, and even strangers gather to share:

• Streaming series and movies
• Live sports events
• Gaming content
• YouTube sessions
• Podcast viewing
• Reality shows

And that’s just the tip of the iceberg… What about the content we share beyond the big screen? The match we watch in a bar. The reel our partner shows us on a tablet. The video we watch with colleagues on someone’s laptop. Or the podcast we listen to while traveling together in a car? Because yes, co-viewing can also be co-listening, but that’s a story for another day—for now, check out this post).

That screen, once purely analog, is now a digital extension. And its measurement has been tied to household peoplemeters (one TV = one household), ad servers, and unique user IDs designed for census-like counting (one TV = one viewer).

But as everyone instinctively knows, behind every video served there could be one person… or several. And that changes everything:

• A spot may be reaching more people than the ad server registers.
• At the same time, frequency is overestimated, because impressions are assigned to fewer individuals than were actually exposed.

In short, ignoring coviewing distorts the real impact of campaigns and ROI. Measuring it—or not—directly affects how much you think each euro (or dollar, or whatever currency you use) of ad spend delivers.

Traditional coviewing measurement doesn’t work

The industry has tried three main approaches to capture coviewing data.

1. TV Audience Panels (TAM)

These panels record who is watching TV in each household, including guests, and extrapolate to estimate coviewing rates.

Problems:

• They don’t always capture streaming ads.
• A large share of current consumption happens outside the home.
• Guest viewing is notoriously difficult to track.

And that’s without even mentioning market-specific issues: additional screens, second homes, out-of-home viewing, different universes, and varying definitions of “valid exposure.”

2. Digital census data (unique users and impressions)

This technical approach relies on logins, cookies, or unique devices exposed to a set of impressions.

Problems:

• A login or a device doesn’t equal a person (shared accounts, multiple people behind one session).
• The digital world is chaotic: impressions served but not viewed, videos played with the screen off, and many other issues.

3. Hybrid TAM + Digital Data Fusion

These combine panel and census data in unified measurement models.

Problems:

• As with other cross-media fusions, the challenge is correctly modeling overlaps.
• Lack of reliable observational anchor points.
• Unreliable projections of coviewing rates.

FLUZO’s solution: people-centric measurement

At FLUZO, we start from a fundamentally different premise: we measure people, not devices.

In our single-source panel, every media exposure is recorded at the individual level, naturally capturing coviewing behavior:

• A panelist watching TV alone at home ✓
• The same panelist watching a streaming series with friends the next day ✓
• Watching a video with colleagues at the office ✓
• Sharing content on a tablet at their partner’s place ✓
• [Insert any unexpected moment] ✓

A practical case

If media consumption is becoming more fluid and fragmented, doesn’t it make more sense to put the consumer—not the device—at the center of measurement? Let’s see how our approach delivers more accurate metrics across three households of different compositions:

Traditional ad server data

• 3 Unique Users detected
• 6 total impressions
• Calculated frequency: 2.0

FLUZO people-centric data

• 4+ actual individuals exposed
• 6+ real exposures captured
• Accurate reach uplift + frequency reduction factors.

To sum up:

Our methodology provides a disruptive yet sensible solution: capture coviewing from the ground up by measuring people. This approach makes coviewing statistically inevitable and enables precise calculation of its impact on both reach and frequency. In other words, by measuring individuals instead of households or device impressions, we reveal the true Reach & Frequency of the campaign—without projections or data fusions.

Ready to discover the true reach and frequency of your campaign? Contact FLUZO to learn how people-centric measurement can transform your metrics.

NOTES
(1) Converting cookies, unique browsers, unique users, or unique logins into real people has always been one of the toughest challenges in digital measurement—going back to the days of Webtrends Log Analyzer and Redsheriff. The shift toward privacy over the last 10 years hasn’t helped either, with GDPR and changes in browsers and operating systems. On the internet, nobody knows how many dogs you are.
(2) A similar issue arises with models that try to assess media overlap and incrementality using things like Sainsbury or similar approaches, as we already discussed here: The battle of the century: how much does Sainsbury reflect reality? – FLUZO