ASSURE: Automated Regression Testing for IFS Cloud.
Author: David Kidd, Managing Director, Cooper Software
Why your IFS Cloud update is only as safe as your last regression test
Picture this. It's a week before payroll runs. Your payroll lead has just been told there's a new IFS Cloud update to test before it goes live. She's also got, well, payroll to run. So, she does what most people in her position do: she tests the happy path, ticks the box, and hopes for the best.
That's not a criticism of her. It's what happens when testing is left to the people who are already busy doing their actual jobs.
Testing isn't an IT problem. That's exactly the issue.
Here's the thing that catches a lot of IFS customers out: proper regression testing can't be done by IT alone. IT doesn't know what "correct" looks like for a finance close, a picking flow, or a payroll run. Only the people who do that job do.
So the burden falls on your finance team, your supply chain lead, your HR and payroll people, the ones who already understand what the system is supposed to do. Multiply that across six or seven departments, each needing tens of hours to properly test their patch of the system, and you're looking at a process that eats weeks of calendar time and hundreds of hours of business time, every single update cycle.
Faced with that cost, a lot of businesses either put off updates or quietly do less than they should. Not because they don't care, but because there simply isn't the time. Testing gets compressed into whatever a stretched team can manage in the days they've got, and the business lives with that risk until, eventually, something breaks.
The part that breaks isn't the bit IFS tested
Here's a distinction worth sitting with. When IFS ships a new version, IFS has tested it. Thoroughly. What IFS hasn't tested, and can't test, is your business.
Every customisation, every modification, every bit of bespoke logic you've built on top of IFS Cloud to fit how you work, that's only ever been tested by you. In the IFS ecosystem these are often called CRIMs (configurations, reports, integrations and modifications), and they're consistently the part of an environment most likely to break during an update.
The mechanism is simple once you see it. Your customisation attaches itself to something in IFS: a field, an object, an endpoint. That's a legitimate, sensible way to build. But IFS changes that underlying structure every release, because IFS is adding genuine value and new functionality as it goes. Your customisation was built on the old foundation. When the foundation shifts, the customisation doesn't always know.
Sometimes that shows up as an error, and honestly, that's the best-case outcome, because you find out immediately. A field a customer order depends on gets renamed, your validation logic can't find it, the order won't save, and you know straight away that something needs fixing before you go live.
The dangerous version is quieter. No error. No red flag. A calculation that used to return one value now quietly returns a different one. A field that used to populate stops populating, and nothing tells you. You promote to production, everything looks fine, and it isn't until months later, at year end, or at a VAT return, that someone notices the numbers don't add up. By then you're not looking at a quick fix. You're looking at unpicking however many transactions have been running on faulty logic ever since.
What ASSURE changes
ASSURE is Cooper Software's automated regression testing product for IFS Cloud. What it changes isn't the need for testing, that doesn't go away. What it changes is who must spend their time doing it, and how thoroughly it gets done.
At the centre of it is an AI-powered process recorder. A business user records themselves going through a process the way they normally would, an order to cash flow, say: raise the order, take the payment, match it to the invoice, settle it. Once, start to finish, the way they'd do it anyway. ASSURE then works out what in that recording needs to stay fixed and what needs to become repeatable, so the same test can run repeatedly without breaking on things like a serial number that already exists from the last run.
It's worth being precise about what this is and isn't. It's not a tool that hands testing to just anyone with no context. It's test acceleration for people who already understand the process. A management accountant who knows their month-end routine can record it once and get a repeatable, automated test out the other end, without writing a line of code.
The result is a genuinely strange thing to watch the first time. A process that takes a person two or three minutes to click through by hand runs, once recorded, in about a second. Do that across a business with a hundred or more tests covering every department, and testing stops being a two-week, all-hands exercise and becomes something one person can kick off and watch complete in a couple of hours, with full evidence attached: screenshots, recordings, a proper attestation of what passed and what didn't.
That changes the maths on updates too. When testing no longer means pulling people away from their day jobs for weeks, there’s far less reason to put off the next release. Organisations can take updates more often, stay closer to the latest version and get the benefit of new functionality sooner, without waiting it out until they’re forced to move.
"My tests broke" is a good sentence, not a bad one
There's a useful way to think about what a good test suite gives you across an update cycle. If you've been running on a stable IFS Cloud version for a while and every test has been passing, and then you update to the next release and a batch of tests suddenly fail, that's not evidence something went wrong with ASSURE. That's exactly the signal you wanted: proof that IFS changed something underneath you, caught before it reached production, not after.
From there it's a straightforward conversation about what needs to change. Sometimes it's the test that needs updating to look somewhere new. Sometimes it's a genuine process change your business needs to make. Either way, you're making that decision deliberately, ahead of go-live, instead of discovering it by accident three months down the line.
Managed or self-hosted, and safe either way
ASSURE is available as a fully managed service, where Cooper Software owns and maintains your test library on your behalf, or self-hosted through your own internal portal for teams who want to build and run their own tests day to day. Neither route compromises on safety: ASSURE is designed to be production safe, with environments explicitly flagged as production or non-production, and separate test-only scripts so testing never interferes with live operations.
Where this fits if you're weighing up your next update
If you're already juggling an IFS Cloud upgrade decision, regression testing is worth building into that plan from the start rather than bolting it on afterwards. The same is true if you've had to navigate compliance-driven changes, like the payroll module challenges we helped a customer through in Solving the compliance issue for unsupported IFS payroll modules. Testing and compliance sit closer together than most testing conversations acknowledge.
If you want to see what setting a hundred tests running in a couple of minutes looks like, get in touch with us and we'll show you against a live IFS Cloud environment.
About the Author
David Kidd is Managing Director of Cooper Software and serves on the company’s board. He has held senior leadership positions across multiple sectors, including Technical Director for a national retailer and manufacturer and Operations Director for hospitality businesses in central Scotland. He has guided organisations through growth and transformation, securing private equity investment and working closely with investors to shape long-term strategy.
David combines deep expertise in ERP, supply chain and manufacturing systems with a proven track record of delivering large-scale technical implementations. Before becoming Managing Director, he was Delivery Director at Cooper Software, leading the Project Management Office, Service Desk and Consultancy division. He now sets the strategic direction for the business, with a continued focus on delivery excellence and helping customers achieve measurable business value from their ERP investments.
Talk to us.
Start a chat, call on 01383 840 700 or complete the form, we will respond to all enquiries.