Dan, I'm not sure how you deduced what my thinking about influencers is  based on the question I posed in my last comment. 

Dave & Ed,

Having repped influencers for over a decade, reading this exchange is frustrating because it conflates two completely different disciplines: ad-tech remnant inventory and creator partnerships.

Ed’s comments show a basic misunderstanding of what influencer platforms actually do. They aren't programmatic networks dropping hidden banner ads below the fold or rotating cheap TV spots. They are discovery, workflow, and attribution tools for creator campaigns.

When a brand approaches an influencer or talent platform, it’s a strategic conversation—defining campaign goals, brand guardrails, key messaging, and success metrics. Equating a creator who actively produces custom content for an engaged community to a Made-For-Advertising (MFA) farm selling 0.1-second scroll impressions is flat-out wrong.

Just as all digital ad networks aren't created equal, not all influencer platforms are the same. Broadly writing off an entire $20B+ industry as "junk impressions" without providing a single piece of data, case study, or platform example isn't media analysis—it's just uninformed generalization.  Any of the "top" CMO's Dave engaged surely would have explained how they approach the space and why they like the option.  But this commentary was one long accusation with zero date or case study to back it up.

It will be very interesting when AI Agents decide to remove the CEO's and other staff.

Ed, good points. I see the uasge of impression machines crossing both of those areas, plus one more. First, I see the influencer platforms putting out a lot of low quailty, self-measured/managed inventory akin to the digial walled gardens along the lines you mention. Second, while they might be sold like TV companies sold RON, I don't seem them as comparable quality or characteristics like TV RON is to the more desirable dayparts or progamming. But third, I see them very similar to the Made-For-Advertisng (MFA) sites that have plagued us for years, that fill up many of the programmatic buys seeking cheap CPMs.

Guys, just for clarification, are we talking about massive, rotating buys in broadcast TV networks, cable channels or streaming services or are we talking mainly about "Walled Garden's" buys on META, You Tube, etc. backed by funny data? It sounds like mostly the latter, to me. 

Dan is correct about big, heavily discounted upfront TV buys being a selling strategy for at least 50 years and this is mainly because such advertisers are assured that they get a fair rotation of ad placements--at 25-30% lower total CPMs-- than by buying more selectively. Wht's more, many CMOs agree with this----providing they are assured of honest reporting on impressions, where the ads run, etc. 

But times are changing and the differential between scatter and upfront CPMs has declined recently. This may cause a rethink in some quarters.

Thanks, Dan, for your comment on "The Blame Game". As you may have noticed, at the end of that chapter I agree with you--that many, often interlocking factors were involved and it's not realstic to blame the whole cord cutting fiasco on bad programmikng decisions---which is what many in Hollywood have been thinking. However, when surveys report that 15-25% of cord cutters say that one reason they dropped "pay TV" is its lack of fare they want to watch, I take that as evidence that this is one of the reasons for many cable dropouts--along with cost savings, no ads, etc.

Joe, excellent analysis. When audience ratings become increasingly dependent on proprietary publisher data and synthetic modeling, transparency, independence, and comparability become critical. Without a consistent measurement standard, are we really comparing apples to apples?

Dan, thanks for the comment. Yes. I did talk to a number of marketing leaders on this topic, which is why I decided to write about it. I also purposefully referenced some long ago endorsers speficially to harken back in a nostalgic way to individual endorsers who had enormous impacts on brands to create contrast to today's infuencers.
You are 100% right to compare these platforms and their impressions to many of the "run-of-network" RON media programs we see, used to dollar average down the CPM for more premium buys that are part of the package. I have exactly the same issue and criticism for many of them. So many of them are hollow, junk impressions that do nothing for the brand and are only there to make the buyer believe that they have some "media math" nirvana. Meta's impressions counted for be viewing in a scroll for "more than 0," many for less than a second, are the kind I include there, as well as the enormous number of impressions that are "below the fold" in web or mobile and counted in campaigns. And yes, many of the culprits for selling this are quality publishers, adding these "impression factory" units as part of "audience extensions."

Dave, labeling your view of efficiency platforms "blasphemous" shows a clear lack of real-world context regarding how legacy media has packaged inventory for decades.

Dismissing influencer networks as mere "impression factories" ignores the fact that they operate on the exact same structural logic as Run-of-Network (RON) in digital/TV or Run-of-Book (ROB) in print. Traditional media has monetized un-targeted, broad-reach inventory for decades to drive foundational cost efficiencies. If broad-rotation programmatic or print buys are accepted as legitimate foundational tactics, why apply a double standard to social media aggregation?

Did you actually consult any working CMOs to discuss why they allocate capital to these networks, or did you just decide to insult your peers' intelligence?

There is risk and reward in all advertising. Citing Michael Jordan and Cindy Crawford requires reaching back nearly 35 years for pristine, high-budget outliers. Meanwhile, modern high-profile single-endorser bets—like Dylan Mulvaney for Bud Light—demonstrate that concentrated single-asset endorsements carry massive headline risk and no guarantee of ROI.

Finally, performance ad networks survive by dynamically optimizing toward publisher nodes and content that yield higher user engagement. Modern influencer networks operate on the exact same mechanics: you establish broad top-of-funnel reach, gain data visibility into which micro-creators generate real traction, and immediately shift budget toward those high-performing individuals and content types. It is the ultimate real-time A/B testing environment.

It is what we commonly refer to as "media math."

You're not the first to speak this truth, but you're one of the few.
I suspect nothing will change real soon, though.

I have been tracking click-through sweepstakes entries for over 20 years. I will say 2026 is different and has been lower.  We have seen a increase in ad blockers but there is more.  Advertisers are holding back in higher quality prizes. Could it be an election year? Maybe. There is a degree of caution in the market place.

Question?

While 'Your Best TV Ad Is Already On Social Media' ... what is the comparison of the TV Ad response c.f. the response in that time differention?

Jack, that says it all!

If a tree falls in the woods and I don't hear it but my bot does .....

Josh, I totallly agree. That is where we are headed. The issue there, probably, will be to bifricate the measurment ... for example, in your toilet paper example, the human portion may be on exposure and impact and the agent portion may be performance measured/compensated only.

I wouldn't be surprised at some point CNN & CBS News merge into one in my opinion. I see many networks shutting down on the cable side like Cooking Channel, Nick Music, TeenNick, MTVU etc.

I'm not even convinced the human ever has to see the ad. That's the thing. 

Suppose I instruct my agent, "Every 3 weeks, buy me a 6-pack of toilet paper rolls, optimizing against these qualities, at cheapest price and same-day delivery."

I've already delegated brand consideration and purchase choice. Advertising to me, you're wasting your time. It's my agent you need to convince. 

I'm not even convinced the human ever has to see the ad. That's the thing. 

Suppose I instruct my agent, "Every 3 weeks, buy me a 6-pack of toilet paper rolls, optimizing against these qualities, at cheapest price and same-day delivery."

I've already delegated brand consideration and purchase choice. Advertising to me, you're wasting your time. It's my agent you need to convince. 

Dave, I think we’re all in agreement here. Verifying that an agent represents a real human still doesn’t verify that the human actually experienced the advertising. That certainly may become the critical distinction. We may need separate standards for machine activity and human advertising exposure and response. An agent can act on my behalf, but after years of developing all my unique human quirks and complexities, I’m not sure any bot is ready to see, feel, or experience an ad on my behalf. Just ask Joe.....

Frank & Josh,
I believe that you've zeroed in on the key issue. It's all about identity and how we treat "agents" working for humans. Bots are non-human traffic and are not countable under any of our current ad standards. We will see some bots that will consume ads and commerical communication, but even of verifiable as acting in baffled of specific known humans, can we also assume that they are going to deliver those ads to their humans with the same experience and impact that the advertisers are paying for?

It wasn't simply editing a video in either case, whether it be fictional or not.

Perhaps a more apt comparison would be an episode of 'The Simpsons' - Season 6, Episode 9

'Homer Badman'

https://www.youtube.com/watch?v=SqDP8SnPVA0

The Broadcast News reference misses the distinction here. The reporter staged a shot of himself crying and edited it into the story as though it happened during the interview. The problem was presenting a fabricated reaction as news, not simply editing video.

Agree with story, but too often there is a outdated assumption "traditional" media (prefer legacy media term) is distinct from social/digital or platforms like TikTok, IG, YT, FB or somehow diffeent. Legacy media are almost always inherently multi-platofrm and some have very large/robust followings which far exceed their traditional platforms. Earned media, unlike ads, also can and syndicate across outlets and platforms. And, media report on other media stories regularly too. Therefore, it's not about trying to garner "traditional" placements only, rather multi-platform story content which can travel. Increasingly there are people within legacy outlets who's focus is on social platforms.  Also the old school tactic of a news release has become more important in shaping AI, whereas few years ago not so much when SEO was more relevant.