People Inc. put numbers on the referral collapse this week that would justify panic at most publishers: Google search traffic down 40% year over year, unique visitors down 22%. And yet CEO Neil Vogel — one of the industry’s most vocal critics of AI companies taking content without paying — declined, again, to pull the crawler trigger. His reasoning was unusually plain: “If we were to turn off AI, we would turn off search… we’re not there yet.”
The phrase doing the work is “not there yet.” It implies a destination, and the rest of People Inc.’s disclosure shows the company building toward it. Non-session revenue reached $125 million, up 16% year over year from $108 million, and now makes up 43% of digital revenue, up from 39%. That line is a bundle of things that do not depend on a browser session: Apple News licensing, AI content deals, events, and the company’s D/Cipher targeting tool. Meanwhile session revenue — the ad money that does depend on visits — held flat while sessions fell, because rates rose.
Read together, the numbers describe a strategy, not a stalemate. Blocking crawlers is a negotiating chip People Inc. cannot yet afford to spend. So instead of spending it, the company is building the position from which the question stops mattering.
The numbers in this piece
01The 43% number is the strategy
Most of the crawler debate is framed as a binary — block or don’t — and most publishers experience it as a trap: allow the crawlers and watch your work power answers that replace the click, or block them and vanish from search. People Inc.’s disclosure reframes the problem. The way out of the trap is not choosing the right side of the binary; it is reducing the share of your business that the binary governs.
Two mechanisms are doing that work at once, and they are separable. The first is the non-session portfolio — licensing, events, data products — which grows revenue that no crawler decision can touch. The second is quieter and more available to everyone: session revenue held flat on falling sessions because rates rose. That is yield work — floor discipline, curation, direct and PMP mix — offsetting a traffic decline in real time. You do not need an Apple News deal or a proprietary targeting product to do it. You need to run your sell-side operation as though every session must earn more this quarter than it did last quarter, because for the foreseeable future it must.
There is also a negotiating logic underneath the restraint. A blocking threat is only worth something if the other side believes you could survive executing it. At 43% non-session revenue and climbing, People Inc. is visibly approaching the point where the threat becomes credible — which is precisely when it becomes most valuable unspent. Publishers who block from weakness get silence; publishers who can block from strength get meetings.
02Why this matters for publishers
| This is the clearest public model yet for surviving the referral collapse | Not "win the crawler fight" — stop needing to. A 40% search-traffic decline coexisting with a growing business is proof the trap has an exit, and the exit is revenue mix, not crawler policy. |
|---|---|
| Most publishers cannot state their own 43% number | Session-dependent and non-session revenue are rarely split in publisher reporting. You cannot manage a hedge you do not measure, and you cannot time a blocking decision without knowing what share of your business it would put at risk. |
| The rate story is the underrated half | Flat session revenue on falling sessions means revenue per session rose enough to cover the decline. That is an operating discipline available to every publisher immediately, independent of any licensing negotiation. |
| The chip only counts if you could play it | Vogel's framing treats blocking as leverage held in reserve. Leverage requires capability: a publisher that cannot technically block per-crawler, selectively and reversibly, is not holding a chip — it is holding an opinion. |
The industry keeps asking publishers where they stand on blocking Google, as though the answer were a matter of conviction.
03What publishers should do
04The bottom line
The industry keeps asking publishers where they stand on blocking Google, as though the answer were a matter of conviction. People Inc.’s numbers say it is a matter of arithmetic. At 39% non-session revenue the block was unaffordable; at 43% it is merely premature; somewhere north of that, it becomes a real option — and, paradoxically, at that point it may never need to be used, because a credible threat negotiates better than an executed one. Every publisher is somewhere on that same curve whether they have calculated their position or not. Vogel has calculated his. The useful response to this story is not to agree or disagree with his restraint — it is to go find out your own number.