Mistakes to Avoid in Large Hardware Projects

Volume work surfaces problems that smaller projects can hide. A specification that reads cleanly for ten handles can quietly fail at three hundred. A finish that looked right on a single sample can drift across production batches. A lead time that felt generous when the project was being planned can become impossible halfway through the build.

We've produced hardware for some of the largest commercial rollouts in the UK over the last few years. Coffee chains, retail estates, hospitality groups, tier-1 contractors. The same mistakes come up across very different projects, and they cost everyone time and money when they do.

Five of them, and how to avoid each one, are below.

1. Specifying hardware too late in the programme

Hardware tends to be one of the last items to get specified on a project, despite being among the longest in lead time. The brief often arrives eight weeks before install when twelve was needed.

Late specification narrows the options. The right finish gets replaced by whatever's available. Suppliers stop being properly compared because there's no time. Decisions get made under pressure, and pressure produces compromises that the rest of the project then has to live with.

If the project completes in October, the hardware conversation should be happening now. Not the order, the conversation. Finishes, sample approvals, programme alignment. Fifteen minutes at concept stage prevents a series of expensive calls later.

2. Treating samples as decoration, not validation

A finish sample is a reference point that the production batch has to match. If the delivered handles don't match the approved sample, there's ground to reject them. Without a properly approved sample, there's nothing to push back against.

There's also a meaningful difference between a marketing sample and a production sample. The marketing sample is hand-finished and kept in a display case for meetings. The production sample is whatever came off the line that week. Asking for both, and comparing them, is one of the quickest ways to understand how consistent a supplier really is.

3. Underestimating finish drift across large batches

Brushed brass is a family of finishes rather than a single one. The same name covers slightly different appearances depending on bar stock, brush head, operator pressure, and the day's humidity. The same applies to bronze, satin nickel, and most PVD coatings.

Variation that's invisible on a small project can become visible on a three-hundred-handle rollout produced across several runs, particularly when the handles sit next to lift surrounds, signage, or other ironwork from a different supplier. Worth asking how your supplier controls finish consistency across batches, whether they keep a master reference sample, and whether your order will be produced in a single run or spread across several.

4. Accepting vague lead time promises

Lead times are easy to commit to in March and harder to deliver in August, once material supply has shifted, machinery has needed maintenance, or another customer's order has overlapped with the slot you were planning to use.

A confident supplier will tell you what their lead time actually depends on. Material availability. Production slot booking. Finish process capacity. They'll tell you when they need confirmed sign-off to hit your install date, and they'll hold the slot for you once they have it. A vague supplier will quote a number and hope. Asking for the conditions behind the number reveals which kind you're dealing with.

5. Skipping the post-install conversation

Most jobs end the day the boxes ship. The hardware gets installed, the project gets handed over, and nobody talks to the manufacturer again until the next brief lands.

That's a missed opportunity on both sides. Most of what we've learned about how to write a clean specification has come from the projects that didn't go entirely smoothly. The brief that was ambiguous. The finish that drifted. The fixing detail that didn't anticipate site conditions. A short review after install, even by email, sharpens the next specification and tends to be the start of a longer working relationship.

What runs through all five

None of these are really hardware problems. They're consequences of how seriously the specification is taken, and how early. Projects that go well tend to have the hardware conversation at concept stage. Projects that go wrong tend to treat hardware as a procurement detail rather than a design decision.

If you're scoping a project for the second half of 2026, this is the moment to start the conversation.