Google Lost $500bn in a Day. What Does AI Search Mean for B2B Lead Generation in 2026?

On 29 October Alphabet lost $500bn of market value in a single session, and Google gave part of the reason itself: its chatbot-style AI Mode makes less money than traditional search. If you buy B2B leads through Google, Meta or LinkedIn, that's your channel being rebuilt while you're paying for it. And the prospects you most want are the ones least likely to see your ad. A phone call with someone in your ICP who picks up still works. That's what Sentrama sells.

This piece is about the infrastructure problem: getting your ICP in front of you at all. What you say once they're listening is the pitch, and I'll keep the two apart.

What happened to Google on 29 October?

Alphabet shed $500bn, 12% of its value, the day after its third-quarter results, The Economist reported. CNBC called it the company's worst day since it listed in 2004. Ad revenue from search and other businesses grew 11%, the slowest in five quarters. What rattled investors was Google's admission that AI Mode brings "lower monetisation" than search.

Meta's ad revenue grew 26% in the same quarter, and on a like-for-like basis it's now about the size of Google's. The Economist puts much of that down to Meta letting AI run more of its ad business. Alphabet says it's now scaling ads inside its AI answers. So the attention platforms will adapt, and they'll do it by putting more of your budget through AI systems you can't see into, in auctions you don't control.

Where have the buyers gone?

Into the answer. A randomised field experiment by researchers at the Indian School of Business and Carnegie Mellon University, posted in April 2026 and not yet peer-reviewed, found organic clicks fell 38% on Google searches with an AI Overview. Searches ending with no click at all rose from 54% to 72%.

The B2B shortlist has moved too. G2 surveyed 1,076 B2B decision-makers in March 2026. Just over half now start software research in an AI chatbot more often than in Google, and chatbots shape the shortlist more than anything else, at 54%. Vendor salespeople come in at 19%. A third bought from a vendor they'd never heard of. G2 runs a review site and surveyed software buyers, so treat the numbers as directional. The direction is hard to miss.

What did an investor tell me last week?

I spoke last week to an investor with a portfolio of B2B companies, and he put it plainly. His companies have mastered online conversion. If an interested prospect sees one ad, they convert. His problem is everyone else: the decision-makers who aren't online all day and never see the ad in the first place. That market was the reason he took the call.

Before we spoke, he'd asked an AI to explain Sentrama and got five different answers. That's how much of a buyer's first impression a chatbot now writes.

Google, Meta and LinkedIn are fighting over your prospect's eyes. We sell their ears.

What should B2B marketing and sales leaders do about it in 2026?

Stop putting your whole pipeline into channels that sell attention.

Ads charge for the impression or the click, then hand you the job of turning it into a conversation. That only works while the prospect sees the ad, and fewer of the ones who matter do. Retainers and outsourced SDR desks have the opposite problem. They reach out, but they bill you for the days whether anyone picks up or not.

For sales leaders it's the same maths from the other end. If a chatbot wrote the shortlist, your reps are either on it or they aren't, and vendor salespeople shape fewer than one shortlist in five. A conversation with someone in your ICP who picks up is one of the few ways onto a list no AI wrote.

Sentrama sells the conversation instead. A Sentrama Lead is a prospect in your ICP who picks up the phone to hear your pitch from our human SDRs, at a fixed price per lead agreed before the first dial. There's no auction, so nobody's AI can reprice it. There's no click, so it doesn't matter whether they ever saw your ad. Our platform does every task before the call: it finds the prospect, removes the wrong records, finds the direct line and works out when they'll answer. On our data, a Sentrama Prospect is 3.5 times more likely to book a meeting than a prospect called from a standard database record.

Google built pay-per-click on predicting the click. We charge per lead because we predict the pickup, and that's how we're creating pay-per-conversation.

What happens once they pick up is the pitch, and that's yours. Every call is recorded and transcribed, and each one comes back as an SQL or an MQL with the reason the prospect gave. So you find out why they said no, which no AI summary will ever tell you.

What's the Monday move?

Pull last quarter's pipeline and mark every deal that started with an ad click or a form fill. Then take your ICP list and ask how many of those people you've ever reached that way. The gap between the two is the market that never sees your ad.

A fixed cost per lead, no auction, and a human on the end of every one: see how Sentrama works for marketing leaders.

Onwards and upwards.