Agentic Commerce
Universal Commerce Protocol setup
Get your catalogue into the protocol Google and Shopify built for agents, properly. Scoped as its own piece because that is how it is usually bought.
In plain terms, what is a universal Commerce Protocol setup?
Get your catalogue into the protocol Google and Shopify built for agents, properly. In practice that means an assessment of what your platform already exposes, which for Shopify merchants is usually more than they realise, and then the harder half: catalogue mapping to the real-time structure UCP expects, including the multi-item cart support added in March 2026.
Why this work exists
What makes this awkward is that it is obvious in hindsight and invisible in advance.
Agentic Commerce holds the whole discipline. This page is one defined piece of it.
The nearest piece of work to it is monitoring parity continuously, because that is where the trust breaks. Where that is already handled, this moves faster than the estimate; where it is not, the two get done together or neither of them holds.
What good looks like
It is healthy when the people who disagreed with it still use it.
The gap between that and where most accounts start is why this is scoped as universal Commerce Protocol setup rather than as a line inside something larger. Treating feed health as a revenue control rather than a maintenance chore is the usual second step once it is in place.
When this is the right work
You can tell this is the constraint when:
- You sell online and have never checked whether an assistant can read your catalogue
- Shopify turned Agentic Storefronts on by default and nobody looked at what it exposed
- Your competitors appear in assistant shopping answers and you do not
If none of those is true, something else is probably the constraint, and the Agentic Commerce overview maps the alternatives. The usual neighbours are getting the product feed accurate enough for an agent to act on and making stock and shipping truthful in real time rather than nightly.
What the work covers
What gets done:
- An assessment of what your platform already exposes, which for Shopify merchants is usually more than they realise, revisited rather than finished, because the inputs keep moving.
- Catalogue mapping to the real-time structure UCP expects, including the multi-item cart support added in March 2026. One pass is rarely enough; the second is much faster.
- Verification that an agent can actually complete the journey rather than failing silently at a step, done before anything downstream is committed to.
- Fallback behaviour, so a shopper who arrives from an agent lands somewhere sensible. Straightforward in principle; the detail is where it stops being quick.
How it runs
How this gets done without the usual two-month discovery phase.
Agreeing what failure looks like
The conditions under which this would be called off are written down at the start. It is uncomfortable and it prevents a year of drift.
Handing back the keys
Access, documentation and whatever was built end up with the client in a form they can use without us, which is the only honest definition of finished.
Before anything is changed
Nothing moves until the current position is written down and agreed. That takes days, not weeks, and it is the only defence against arguing about whether something improved.
What each part runs into
What gets in the way, part by part:
- An assessment of what your platform already exposes, which for Shopify merchants is usually more than they realise, which in Chennai runs into an enquiry path that runs through a messaging app rather than a form
- Catalogue mapping to the real-time structure UCP expects, including the multi-item cart support added in March 2026: the Chandigarh version of this has to deal with a catchment smaller than the ambition on the website
- Verification that an agent can actually complete the journey rather than failing silently at a step, where a Manchester engagement contends with an incumbent whose position comes from presence rather than effort
- Fallback behaviour, so a shopper who arrives from an agent lands somewhere sensible – working in Chicago this means accounting for a buying pattern that resolves faster than the reporting cycle
The failure mode
Building a custom integration when the platform already supports the protocol. It is work you will maintain forever for no advantage.
The damage is less in the wasted work than in the conclusion it leads to.
How it is measured
Eligible catalogue share, agent-sourced sessions, and completion rate from agent referral. That measure is agreed before the work starts, so there is no argument about the result afterwards.
Agreed before the work starts, so there is no argument about the result afterwards.
What you end up with
The output is a protocol readiness report, the catalogue changes needed, and a tested path from assistant to purchase.
Scoped so the first action can be taken the same week it arrives.
How it differs by sector
What this becomes in practice depends on who it is for:
- Finance and fintech brings its own limit: regulated claims, compliance review in the publishing path, and trust signals that have to be real.
- Travel and hospitality, constrained by the fact that aggregators hold the demand, so direct booking depends on being the specific named property.
- Marketplaces and platforms: Two sides to acquire, and a programmatic page set that can sink the whole domain. so this work has to account for that before anything else.
The industry pages carry the rest of that detail. Finance and fintech and Travel and hospitality are the two where this piece of work most often decides the outcome.
Common questions
What is UCP?
The Universal Commerce Protocol, launched by Google and Shopify in January 2026 with more than twenty backers, which lets assistants read catalogues and build carts in a standard way.
Is this the same as OpenAI’s protocol?
No. OpenAI’s Agentic Commerce Protocol came first and underpinned Instant Checkout. They solve a similar problem differently, and which one matters depends on where your buyers shop.
What do we actually get at the end?
A protocol readiness report, the catalogue changes needed, and a tested path from assistant to purchase. Documented, so it survives whoever was in the room, and specific enough to act on without a follow-up conversation. There is more on this in Automotive, and a different angle on it in B2B services.
What if this is not our problem?
Then we say so. The signals on this page are the honest filter, and work sold against the wrong diagnosis wastes a quarter. Where this gets detailed is Analytics and Measurement; where it gets argued about is Answer Engine Optimization.
What could go wrong with this work?
The usual risk is acting on a finding before it is solid. The diagnostic phase exists to avoid that, and where the data is too thin to be conclusive we will say so rather than proceed. GEO or SEO goes one level down from here, and Google Ads or Meta Ads two.
Ask about universal commerce protocol setup
Tell us what you have tried already. That is usually the fastest route to a useful reply.
Find out whether this is worth doing at all
We would rather tell you the work is not worth the money than take it and report on activity.