EPD Engineering Solutions Limited

02/09/2026

A cleaned up drawing still had two columns too many, and the ones that were left were not sitting where the beam intersections said they should be. None of that showed up until I overlaid it against the real model and checked it against the structure, not the file.

Software gives you a layout every time. It cannot tell you whether that layout survives contact with a real building. That check, by hand, against something real, is still the job. It is slower than trusting the first output, and it is the difference between a drawing that looks finished and one that actually is.

What is the last thing you caught only because you checked it yourself?

31/08/2026

What actually happens when we finish a design job:

Every drawing gets its own number and revision, so nobody's left guessing which version is current.

The bill of materials is broken out by supplier: steel, fixings, anchors, laser cut plate, each with the exact standard specified.

Anything that needs a DXF or NC file to manufacture is issued as one.

One named contact on the document, not a support inbox.

Same standard whether it's one drawing or seventeen.

Full breakdown in the carousel below. If your last handover was "here's a folder, shout if you need anything," it's worth comparing notes.

26/08/2026

We built a staircase that cannot be drawn wrong. It is a parametric model in Inventor where Building Regs compliance is not checked after the drawing is done, it is built directly into the geometry. Change one number, the floor to floor height, and every riser, tread and stringer notch updates itself and stays within the regs automatically. That is not a time saving trick, it means every future stair, built by anyone, on any job, starts from a model that already knows the rules. We are still finishing the handrail, getting the upright spacing right without clashing the tread fixings has been the hardest part of the build. If you build parametric models for your own work, where do you draw the line between what the model should check for you and what still needs a human eye on it?

24/08/2026

I will not touch a drawing until I can answer three questions.

Most rework does not start with a bad calculation. It starts because nobody agreed what finished looked like before the work began.

Before any drawing opens, I check three things: what the client expects at the end, what information and format they need, and how quickly they want it against the standard the work has to meet.

I have redone contractors' work and brought jobs in-house to fix them, because nobody asked these three questions first.

Swipe through for how each one works, and why the scope of works conversation at the start saves more than it costs.

If your team is opening drawings without this conversation, that is the gap we fill. Get in touch and we will show you how we structure ours.

20/08/2026

Work has been piling up lately at EPD which is a nice problem to have, so I did the one thing most business owners skip when that happens.

A fortnight off gave me the distance to look at it properly. Instead of jumping straight to adding resource, I went through the tasks I repeat on every job and asked why they still took as long as they did.

The biggest one was parts lists on drawing packs. I'd been manually overriding them for years. Fine for a one off job. Not fine once jobs are stacking up and every pack is costing an extra couple of hours.

There's a sharper risk underneath that. Override a value and it stops updating with the model. Change something later and the drawing says one dimension while the parts list says another. That's the kind of mismatch a client spots before you do.

The fix took an afternoon. No new software, no extra headcount. Just time spent properly on a process I'd been working around for years.

Full breakdown in the latest article on LinkedIn, including the same fix applied to mezzanine drawing grids.

If you're not sure whether your own overload needs more people or a better process, that's usually the first conversation worth having. DM me.

Work was piling up, so I did the one thing most business owners skip when that happens. A fortnight off gave me the distance to look at it properly. Instead of jumping straight to adding resource, I went through the tasks I repeat on every job and asked why they still took as long as they did. The b...

05/06/2026

04/06/2026

Busy. Always delivering. Never quite as profitable as you should be.

That is the symptom. The triage finds the cause.

60 minutes. Free all of June. We look at your projects or your business and find where the margin is going.

No pitch. No obligation. Just answers.

DM me to book.

FabricationBusiness EPDEngineering SME FixedPriceProjects

03/06/2026

EPD is part of UK Small Business Week 2026.

Free 60 min triage call. All of June. Worth £500.

One problem project or the whole business. Your call. No pitch. No obligation.

You just leave knowing what is actually going wrong commercially and why.

DM me to book.

AutomationBusiness ProjectMargin EPDEngineering SME ShopSmall