First of all, congratulations. Winning an Innovate UK or NIHR i4i award is a real achievement, the success rates are low and the assessors will have picked your project apart before funding it. What I see quite a bit of, though, is teams treating the award letter as the finish line when it’s actually the starting gun. The first 90 days shape the whole project, and in my experience the teams that do well are the ones that use that window to get organised rather than diving straight into the fun engineering work.
So here’s a practical view of what actually happens next, based on the projects we’ve supported at MDIN.
Get through project setup first (the clock is already running)
For Innovate UK awards, the funding offer comes with a setup process run through the Innovation Funding Service portal, and the deadlines are tighter than people expect. Within 5 days you need to confirm your project manager, finance contact and bank details. Within 30 days you upload your exploitation plan (and a collaboration agreement if you have partners). Within 90 days the whole setup needs to be complete, including responses to any financial checks, and Innovate UK are clear that the offer can be withdrawn if it isn’t.
You’ll also be assigned a monitoring officer through Innovate UK’s monitoring service provider, who reviews progress against your plan at regular intervals for the life of the project. NIHR i4i works a little differently (contracting runs through your host organisation and progress is tracked against the milestones in your application), but the principle is the same: someone independent will be checking what you said you’d do against what you’ve actually done.
One point worth knowing early: your project can’t start, and you can’t spend grant money, until the grant offer letter is signed and the project has formally gone live. Costs incurred before the start date generally aren’t claimable, so hold off on ordering that equipment until you’re through setup.
Keep on top of claims and record keeping
Grants are paid in arrears against claims, typically quarterly, so there’s a cash flow gap to plan for and your finance contact will need timesheets, invoices and evidence of spend to back each claim up. It’s dull administration compared with the engineering, but in my experience the projects that keep claim records tidy from week one save themselves quite a bit of pain later, particularly if a claim gets queried or the project is picked for an audit. Half a day setting up a simple system for timesheets and cost tracking at the start is time well spent.
Turn the application into a development plan you can actually run
Grant applications are written to win funding, and as such they tend to be optimistic documents. The work packages sound tidy, the Gantt chart is neat, and everything finishes exactly on time. The first real piece of work after the award is converting that into a development plan you can run day to day, with dated activities, named owners and milestones your monitoring officer can review without wincing.
My view is it’s worth being honest with yourself here about what the application glossed over. If a work package says “design and build prototype” in one line, break it down properly now (concept work, design iterations, procurement lead times, build, test), because those quarterly reviews come around fairly quickly and slippage is much easier to manage when you spot it early.
Decide how you’ll handle quality and design controls
You don’t need a full ISO 13485 quality management system on day one, and I’d generally advise early-stage teams not to try to build one immediately. What you do need from the start is the habit of capturing design decisions as you make them: user needs, design inputs, the reasoning behind choices, test results. This becomes the backbone of your design history and, eventually, your technical file, and retrofitting it two years in is genuinely miserable work compared with keeping it as you go. I’ve written separately about what actually goes in a technical file if you want a sense of where all this ends up.
The sensible middle ground for most funded startups is lightweight design control procedures early on, then a proper ISO 13485 QMS as you move from feasibility into formal design and development. When to make that switch depends on your device class and your route to market, which brings me to the next point.
Set the regulatory strategy now, not at the end
I’d say this is the single most common gap in funded projects: the regulatory route gets left as a task for year two, and then it quietly reshapes the whole development plan when someone finally looks at it. Classification, intended purpose and claims should be pinned down in the first 90 days, because they drive how much evidence you need and how long everything takes.
For UK-focused devices the picture in 2026 is reasonably workable. CE marked devices are currently accepted in Great Britain until 30 June 2028 for devices certified under the old directives and 30 June 2030 for devices certified under the EU MDR, and the MHRA ran a consultation earlier this year on recognising CE marked devices indefinitely (it closed in April 2026 and the outcome is still awaited at the time of writing). UKCA remains available as a domestic route. Which to pursue first depends on where your customers are, and it’s a decision worth making deliberately rather than by default. Also worth noting, the UK’s strengthened post-market surveillance requirements have been in force since June 2025, so PMS planning belongs in the technical documentation from the start rather than as an afterthought.
If your device delivers a medicine, you’re likely in Article 117 territory, which has its own quirks and is worth understanding early rather than discovering later.
Get to a physical build quickly
Monitoring reviews go much better when there’s something on the bench to point at, and more importantly, a physical build is the fastest way to find out whether the clever idea in the application actually works. For most projects I’d push for a proof-of-concept build in the first quarter, something quick and unpolished that answers the riskiest technical question, before committing to a full prototype. The distinction matters more than people think, and I’ve set out the difference between proof-of-concept and prototype builds in a separate guide.
That was broadly the approach on the HelloCath project, where early concept models and silicone prototypes let the team test the core idea and settle the regulatory route without burning the budget on tooling. Cheap builds that answer hard questions are, in my experience, the best money a funded project spends.
Where an external partner fits
Plenty of funded teams are two or three people with deep clinical or scientific knowledge and no in-house mechanical design, prototyping or regulatory capability, and the grant is often what makes bringing in outside help affordable. We work with teams in exactly that position at MDIN, covering the design, prototyping, bench testing and regulatory side while the founders stay focused on the clinical and commercial work.
Frequently asked questions
Generally no. Costs incurred before the formal start date in your grant offer letter usually aren’t claimable, so wait until the project has gone live before committing spend, however tempting the lead times make it.
Not necessarily on day one, but earlier than most teams think. Lightweight design controls from the start, capturing inputs, decisions and test results, then a full QMS as you move into formal design and development. Retrofitting records later is slow and expensive.
It depends on where your customers are. CE marked devices are currently accepted in Great Britain (until 30 June 2028 or 30 June 2030 depending on certification route, with indefinite recognition under consultation), so many UK startups pursue CE first, but it’s a strategic decision worth making early.
Progress against the plan in your application: milestones, spend against forecast, your exploitation plan and any risks to delivery. Reviews are typically quarterly, and honest reporting of slippage goes down far better than surprises.
In the first quarter if you can. A quick proof-of-concept build that tests your riskiest assumption is usually the most valuable early activity, and it gives your first monitoring review something concrete to point at.
If you’ve recently had an award confirmed and want to talk through your first 90 days, I’d be happy to have a call. You can book a free 30-minute slot here, no obligation, and even if we don’t work together you’ll hopefully come away with a clearer plan.
Rob Wakeford, MDIN Ltd, www.mdin.co.uk