CommerceBolt — a Perigeon company
Talk to an expert

Service

Unified Commerce

Unified commerce means one stock position, one customer record and one order history, whichever channel the customer used. Most retailers have several of each, which is why click-and-collect is harder than it looks.

Diagnosis first

The four records that need to become one

Omnichannel sounds like a front-end problem. It is almost entirely a question of which systems are allowed to disagree.

  1. Stock differs between web and store

    Two systems both counting. Until one owns the position and the other reads it, every promise either oversells or hides sellable stock.

  2. The same customer is three customers

    One record in the storefront, one in the till, one in the email platform. Loyalty, service history and personalisation all fail quietly on this.

  3. Store staff cannot see a web order

    So returns to store become a negotiation. The order record has to be visible wherever the customer turns up, or the policy is unenforceable.

  4. Click-and-collect promises what it cannot fulfil

    Because availability shown online reflects a nightly sync rather than what is on the shelf now, and nobody reserved the item when the order was placed.

Decide what is allowed to disagree

Every multi-channel business has systems that hold overlapping data. The question is never whether they can be kept identical — they cannot, there is always lag — but which one is authoritative and how stale the others are permitted to be.

Answer that for stock, for customer and for order, and most of the architecture follows. Leave it unanswered and you get the familiar symptom: a nightly job that overwrites a correction someone made that afternoon, and two teams each certain their number is right.

Pilot one store

The temptation with an estate of locations is to design for all of them and launch together. It is the wrong shape for this work.

Stock accuracy, pick reliability and collection timing vary by location and are learned from real orders. Running one store for a month tells you what your actual accuracy rate is, which is the number every other decision depends on — and getting it wrong in one location is recoverable in a way that getting it wrong in forty is not.

Where this sits

This is integration work with a retail operating model attached, built on the storefront covered by eCommerce development. If you sell to businesses as well, the account and pricing mechanics are covered under B2B commerce.

Platforms

Where we build.

Stock accuracy is an operational problem, not a software one

No integration can make a system report stock that the shop floor has not counted correctly. Unified commerce projects succeed or fail on whether stock accuracy in store is good enough to promise against — and improving that is a process, training and cycle-counting question. We will tell you if the integration is not your limiting factor.

Talk through your channels

Scope

What a build includes.

  • A single inventory position

    One system owning stock by location, with reservation at order time so the same unit cannot be sold twice in two channels.

  • One customer record

    Identity resolved across web, till and marketing, so service history, loyalty and preferences follow the person rather than the channel.

  • Orders visible everywhere

    Web orders readable in store and store orders readable online, including returns and exchanges across the boundary.

  • Fulfilment routing

    Rules deciding which location ships or picks an order — proximity, stock depth, capacity — rather than defaulting to one warehouse.

  • Click-and-collect that holds

    Reservation, pick confirmation and a collection window, with availability that reflects reality closely enough to promise against.

  • Returns across channels

    Buy online, return in store and the reverse, reconciled so finance sees one transaction rather than an unexplained credit.

How it runs

From first call to live.

  1. Channel and systems map 2–3 weeks

    Which systems hold stock, customers and orders today, which is authoritative for each, and where they currently disagree.

  2. Ownership decisions 2 weeks

    Naming the single owner of each record, and deciding the tolerated lag. This is a business decision with technical consequences, not the reverse.

  3. Build 10–16 weeks

    Integrations, reservation logic and routing rules, built against a single pilot store rather than the whole estate.

  4. Pilot in one location 4–6 weeks

    Real orders, real staff, real returns. Stock accuracy and collection reliability get tuned here, where a mistake affects one shop.

  5. Roll out Ongoing

    Store by store, with staff training and the reporting that tells you when a location's stock accuracy is drifting.

Frequently asked questions

What is the difference between omnichannel and unified commerce?

Omnichannel describes the customer-facing goal: a consistent experience across channels. Unified commerce describes how you achieve it: one set of underlying records — stock, customer, order — rather than several that are periodically reconciled. You can present an omnichannel experience over fragmented systems for a while, but it breaks at the edges, which is where customers notice.

Do we need to replace our POS?

Not necessarily. What matters is whether your POS can expose stock, customers and orders through an API in something close to real time. Many can. Where it cannot, the choice is between a middleware layer that compensates and replacing it — and the middleware route is often cheaper and less disruptive than it is given credit for.

How accurate does store stock need to be?

Accurate enough that a promise made online is kept. In practice that means knowing your current accuracy rate before you design against it, and holding a buffer where it is weak. Selling click-and-collect on stock data you do not trust produces more disappointed customers than not offering it.

Can we do this in stages?

Yes, and you should. The usual order is unify stock visibility first, then customer identity, then orders and returns across the boundary. Each stage is useful on its own, which also means the programme can be paused without leaving something half-built.

Is this relevant if we only sell online?

Less so. If you have one storefront and no physical locations, most of this is already unified by default. It becomes relevant once you add marketplaces, a second region with separate stock, or any physical presence.

Want to talk through Unified Commerce for your store?

Tell us where it hurts. We will tell you honestly whether we are the right people for it.

Talk to an expert Book 30 minutes