Every SME owner asks the same question eventually: what does an API integration actually cost? The honest answer is that API integration cost in the UK varies wildly, from a £30 monthly Zapier subscription to a £40,000 custom build, and most agencies won't tell you which one you actually need. We will.
This matters because the wrong choice in either direction is expensive. Buy custom when a £20 tool would do and you've wasted budget. Buy a no-code stopgap when your order volume needs a proper system and you'll spend the next two years firefighting failed syncs.
What drives API integration cost in the UK
Cost isn't really about the integration itself. It's about three things: how many systems are involved, how reliable the connection needs to be, and how much your data needs transforming on the way through.
Connecting Shopify to Xero for basic invoice syncing is a well-trodden path with mature tools either side. Connecting a warehouse management system to four different retailer portals, each with its own order format, stock feed cadence, and authentication method, is a different job entirely. That's the difference between a £50/month tool and a five-figure build.
The hidden cost driver: what happens when it breaks
Off-the-shelf tools are built for the average case. When an API call fails, a token expires, or a supplier changes their field names without warning, tools like Zapier will often just stop the run and email you. A custom integration can retry, log the failure meaningfully, alert the right person, and keep the rest of your operation running. That resilience is where most of the cost actually sits, not in the initial connection.
When Zapier or Make is genuinely enough
We're not precious about this: plenty of businesses turning over £500k to £2m a year genuinely don't need custom work, and we'll tell them so instead of quoting a five-figure build they don't need.
- Low volume, low complexity. Under a few hundred transactions a month, between two well-supported apps (Shopify, Xero, Mailchimp, Slack), with no complex logic in between.
- You can tolerate occasional manual fixes. If a failed sync means someone spends ten minutes re-running a task rather than a customer not getting their order, the risk is acceptable.
- The data doesn't need heavy transformation. Field-to-field mapping with maybe a filter or two. Once you're writing complex conditional logic across multiple steps, you're paying Zapier or Make prices for something that behaves like custom code anyway, without the reliability of custom code.
Cost here is genuinely low: typically £20 to £100 a month in subscription, plus a few hundred pounds if you pay someone to set it up properly rather than cobbling it together yourself.
When you need a custom integration
The tipping point usually isn't volume on its own, it's operational risk. Ask what happens if this integration is wrong or down for a day.
- Multiple systems with different data shapes. A fulfilment company pulling orders from six retailer channels, each with different SKU formats, needs mapping logic that no-code tools handle badly at scale.
- You need guaranteed delivery, not best effort. Order data, stock levels, and financial figures that must reconcile exactly, every time, with proper error handling and retry logic.
- Business logic in the middle. Allocating stock across warehouses, applying courier rules by weight and destination, or triggering different workflows depending on order value. This is application logic, not a Zap.
- Volume that breaks task-based pricing. More on this below, because it catches people out constantly.
This is the territory covered by our API and system integration work, and it's usually where a proper custom web application sits alongside the integration itself, because the data needs somewhere sensible to live and be managed, not just passed through.
Real cost ranges
Figures below are typical UK agency pricing for 2024/25, based on scope we see regularly. Treat them as a starting point for conversation, not a quote.
- Simple two-system sync (e.g. Shopify to Xero, one direction, standard fields): £2,000 to £5,000 build, £50 to £150/month maintenance if hosted and monitored properly.
- Multi-channel order integration (marketplaces or retailers into a WMS or ERP, bidirectional stock and order flow): £8,000 to £20,000, depending on how many channels and how bespoke the mapping.
- Full operational integration (custom WMS, courier APIs, Sage or Business Central, client-facing portals): £20,000 to £60,000+, usually delivered in phases so you see value before the whole thing lands.
- Ongoing maintenance: budget 10 to 20% of build cost annually for monitoring, API version changes, and the inevitable supplier field-format change that breaks a mapping.
We've built this end to end for fulfilment operators, including a custom WMS with multiple API integrations across several retailer channels; the Lama Fulfilment case study covers how that played out in practice, from initial scope to the point Zapier alone would no longer have coped.
The gotcha nobody mentions: task-based pricing at volume
Here's the practitioner detail that catches almost every growing SME. Zapier and similar tools charge per task, and a single order touching five apps (marketplace, WMS, courier, accounting, email) counts as five tasks, not one. At 300 orders a day, that's 1,500 tasks before lunch. Most businesses budget for the £50/month tier and end up on the £300+ tier within two months of scaling, having assumed the tool was basically free.
There's a second trap underneath that: rate limits. Platforms like Linnworks and most retailer APIs cap how many calls you can make per minute per endpoint. If a no-code workflow polls stock levels every five minutes across thousands of SKUs, you will hit that ceiling, and updates get silently throttled or dropped rather than erroring loudly. Nobody notices until a customer orders something that's actually out of stock. A custom integration handles this with batching and webhooks instead of polling, which is invisible in a quote but is exactly the kind of thing that prevents a bad Trustpilot review three months after launch.
A simple decision framework
- What's the cost of a silent failure? If it's a missed Slack notification, use no-code. If it's a wrong stock count or a missed dispatch, don't.
- Is your volume stable or growing fast? Growing fast means you'll outgrow task-based pricing within a year. Build for where you'll be, not where you are.
- Does the logic fit in a flowchart, or does it need real conditions and exceptions? If you're drawing more than five boxes to explain it, it's custom territory.
- Do you need an audit trail? Finance and compliance-sensitive integrations, VAT handling, multi-currency, Sage postings and the like, need proper logging that off-the-shelf tools rarely provide out of the box.
Most businesses we speak to actually need a hybrid: no-code for the low-stakes internal notifications, custom for anything touching orders, stock, or money. That's usually the cheapest path overall, because it stops you paying custom-build rates for a Slack alert.
If you're not sure which side of that line your business sits on, that's a conversation worth having before you commit budget either way. Get in touch and we'll tell you honestly, including if the answer is that you don't need us yet.