The funnel most marketers grew up on – see the ad, click, browse, add to cart, buy – did not gradually erode. It came apart almost overnight, as shoppers got comfortable discovering, deciding, and buying inside social feeds and LLMs, often in a single moment.

Recorded live at K:LDN 2026 in London, this is the close-out conversation for the series, a step back to ask what actually replaces the funnel and what it demands of the brands trying to keep pace.

Alicia Esposito of Future Commerce sits down with Kelsey Capps, who leads Klaviyo's product work across analytics, personalization, and AI, and so has a front-row seat to the change. A shopper can now find a product in a TikTok comment, buy it, and post about it before the marketing team has time to react, which means the advance-planning window marketers relied on has all but vanished.

Kelsey's argument is that brands have to rebalance, away from a roughly 80/20 split of campaigns over automated flows and toward signal-triggered automation that can react in the moment, like a flow that fires when someone arrives from a ChatGPT referral. Underneath the tactics sits a bigger shift: optimizing your site and product data so the machines doing the discovering can actually surface you, and building an operating model around understanding customers rather than merely targeting them.

Her closing note is optimistic. The data advantage that once belonged only to enterprises is now within reach of the smallest brand, and because customer intelligence compounds, the brands that start building the model first are the ones that pull away.

What you'll learn

What replaces the linear funnel once discovery, comparison, and conversion happen at once
Why the advance-planning window has collapsed, and what to do about it
How to rebalance from campaigns toward signal-triggered automated flows
Why optimizing your product data and brand voice for machines is now table stakes
How a single wrong recommendation breaks trust, and why understanding beats targeting
Why customer intelligence compounds, and why first movers pull away

Pull quotes

"Customers don't want to be just targeted, they want to be understood." — Kelsey Capps [13:50]

"You can see when someone came to your site from ChatGPT. That's just a UTM source on an event, and a brand could automate a workflow around it." — Kelsey Capps [16:15]

"This conversation used to be confined to the four walls of industry events. Now your family and friends are having the same conversations about how their data is being used." — Alicia Esposito [23:01]

"Even your smallest mom-and-pop shop can take advantage of their first-party data in a way they just never could before." — Kelsey Capps [24:55]

Chapters

0:00 Cold open and introductions
5:23 Perpetual commerce and the collapse of the funnel
6:11 Discovering and buying inside a TikTok comment
11:43 Understanding customers, not just targeting them
14:46 Optimizing your data for the machines doing the discovering
15:40 The vanishing planning window, campaigns versus flows
17:08 Composer and orchestration pulling it together
19:03 Vision in the age of autonomy
22:00 The turkey-sub story, how a bad recommendation breaks trust
23:43 What Kelsey is optimistic about, and the urgency for laggards

In-Show Mentions:

Learn more about Klaviyo’s Composer

Associated Links:

Check out Future Commerce on YouTube
Check out Future Commerce Plus for exclusive content and save on merch and print
Subscribe to Insiders and The Senses to read more about what we are witnessing in the commerce world
Listen to our other episodes of Future Commerce

Have any questions or comments about the show? Let us know on futurecommerce.com, or reach out to us on Twitter, Facebook, Instagram, or LinkedIn. We love hearing from our listeners!

Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.