From aisles to agents: what John Lewis selling via ChatGPT and TikTok signals for modern discovery
John Lewis recently announced a significant investment in AI-powered shopping, with plans for its products to be surfaced in AI apps such as Google Gemini and ChatGPT, and a launch on TikTok Shop.
What this signals for wider brands is really interesting. This is a simple, high-profile example of how discovery and decision-making are shifting, and how brands are adapting to meet customers where they already are.
If you have ever found yourself thinking “search feels harder now”, this is a big part of why.
What is actually changing
For years, many teams have operated with a fairly contained model:
People search (mostly on Google)
They click through to a website
They compare, decide, and convert
We measure performance through clicks, sessions, and attribution
That model still exists, but it is no longer the whole story.
Discovery is now multi-channel and multi-modal. People jump between platforms, formats, and sources of reassurance, often before they ever land on a website. In many journeys, the moment of decision is shaped earlier than most reporting suggests.
John Lewis’ move is a strong signal of this shift. Their stated aim is to make products highly visible on AI apps when customers seek ideas and inspiration, and to make items instantly shoppable through third-party apps.
So the transition is more towards: presence where decisions are forming, not just presence where clicks are happening.
Why this matters beyond retail
Even if you are not selling products directly through social commerce or AI apps, the underlying behaviours show up across sectors:
Travel: inspiration and validation through social, video, reviews, and aggregators, often before a brand site visit
Home and property: comparison, local validation, and trust signals across directories, maps, reviews, and communities
Professional services and B2B: shortlists shaped by reputation, proof, and third-party references long before someone fills in a form
The mechanics differ, but the pattern is consistent. Customers are assembling confidence from a wider ecosystem of signals.
The ecosystem that now influences the shortlist
If you want a practical way to think about this, consider the places people now use to form an opinion:
Reviews and ratings (Trustpilot, Google reviews, sector platforms)
Online mentions and third-party validation (press, partners, communities, “best of” lists)
Social presence (not just content, but consistency and credibility)
Video and creators (YouTube, TikTok, Instagram, shorts and explainers)
Directories and marketplaces (where comparisons and filters happen)
Forums and communities (the unfiltered reality check)
AI and LLM tools (summaries, shortlists, comparisons, recommendations)
Your website still matters. It is often the most controllable asset you have.
But it is increasingly not the only place your brand is evaluated, and it may not be the first place an opinion is formed.
The measurement challenge: why “search feels harder” now
As journeys fragment, measurement gets more nuanced. If more discovery happens inside platforms, and more reassurance comes from third-party sources, then clicks and last-touch attribution become weaker proxies for influence.
That does not mean measurement is impossible. It means leaders need to evolve what they consider “search performance” and “digital performance” to include:
visibility across the ecosystem, not only rankings
share of demand and share of attention, not only sessions
proof signals and trust indicators, not only conversion rate
conversion quality, not only volume
decision-led reporting that helps teams choose what to do next
In short, the task is not to build bigger dashboards. It is to build clearer decision-making.
What brands should focus on next
John Lewis’ announcement will trigger a lot of tactical discussion about TikTok Shop, AI integrations, and new formats.
Those are relevant. But the bigger opportunity for most organisations is to strengthen the system around discovery and decision-making.
Here are the areas we see making the biggest difference.
1) Map where demand forms, not just where it is captured
Most teams can tell you where leads come from. Fewer can describe where interest begins.
Map the real journey. Include social, video, reviews, marketplaces, directories, communities, partners, and AI-led touchpoints.
This is often where the quickest clarity comes from.
2) Make your proposition portable
When customers discover you across platforms, you need a proposition that survives being seen out of context. Ask:
Is it obvious who we are for?
Is it clear why we are credible?
Are our claims consistent across channels?
3) Build proof that travels
In an ecosystem world, proof is not one case study buried on a website. It is a set of signals that appears repeatedly:
reviews, ratings, and volume
third-party mentions
creator and customer content
clear examples, outcomes, and evidence
This is also the type of information that tends to show up in comparison moments and AI-led summaries.
4) Treat video as a trust format, not just a brand format
Video is increasingly part of validation. It answers questions quickly, shows reality, and reduces perceived risk.
The brands doing well here are not necessarily producing more. They are producing useful: explainers, comparisons, “what to expect”, walkthroughs, FAQs, and proof-led stories.
5) Get disciplined with feeds, data, and consistency
As discovery becomes more structured and platform-led, product and service data matters.
For retail, that is feeds and availability. For services, it is structured information: locations, service pages, pricing cues, proof points, and consistent facts across directories and listings.
6) Improve conversion quality, not just acquisition
When attention is harder to win, and clicks can be fewer, conversion quality becomes a bigger lever.
Many performance gains come from:
clearer messaging
less friction
stronger reassurance
better landing page relevance
tighter alignment between the promise (ad, snippet, post) and the experience
7) Fix the alignment problem
In our experience, most underperformance is not technical. It is organisational.
Organic, paid, social, web, PR, CRM, partners and agencies often operate separately, with separate reporting and separate incentives.
The teams that win simplify this:
one joined-up roadmap
clear ownership for outcomes
a cadence for decisions
reporting designed to prompt action, not just describe activity
A simple way to start
If you want a practical starting point, begin with three questions:
Where do customers discover and validate us today, outside our website?
What do they see in those places, and does it build confidence?
Do we have one plan and one view of priorities across the ecosystem?
If you can answer those clearly, you are already ahead of most teams.
How Saurce supports this kind of work
We typically support leaders through:
Search ecosystem and discoverability reviews (beyond Google, across the journey)
Trust and validation audits (reviews, mentions, directories, proof points)
Journey and conversion quality audits (from discovery to decision)
Measurement and reporting frameworks designed for leadership decisions
Partner and agency oversight to keep delivery aligned to outcomes
If you are exploring how your brand shows up across this wider ecosystem, or you want a clear view of what to prioritise next, we are always happy to talk.
Sources
1) John Lewis Partnership (09 March 2026) — “John Lewis invests in future of retail with AI-powered shopping and TikTok Shop launch”https://www.johnlewispartnership.co.uk/media-centre/latest-news/2026/23871
2) Google Ads & Commerce Blog (11 February 2026) — “What’s ahead for commercial experiences in 2026”https://blog.google/products/ads-commerce/digital-advertising-commerce-2026/
3) Google Ads & Commerce Blog (11 January 2026) — “New tech and tools for retailers to succeed in an agentic shopping era”https://blog.google/products/ads-commerce/agentic-commerce-ai-tools-protocol-retailers-platforms/
4) Yahoo Finance (March 2026) — “John Lewis expands AI shopping under £800m transformation plan”https://finance.yahoo.com/news/john-lewis-expands-ai-shopping-112502160.html
5) Search Engine Land (11 February 2026) — “Google outlines AI-powered, agent-driven future for digital advertising and commerce in 2026” https://searchengineland.com/google-shares-whats-next-in-digital-advertising-and-commerce-in-2026-468995
