What should I spend my money on first?

Typical hardware development budgets, the four early-stage blockers and where your first £15,000 should go. From an ISO 9001 design consultancy.

By Nathan Ruttley, CEO and Co-Founder·13 September 2026·8 min read

Small budgets don't always have to hold back development, they just need to be used wisely.

Determining your project budget is a daunting task, especially if it is your first venture into hardware development. What's the right amount to spend in the first instance? What will it cost in total to bring your product to market?

Let's put some figures on it: £15,000 is a lot of money to an individual, but might be a tiny amount of money to a large business. £25,000 might cover your tooling for a simple product, but maybe you'll need £300,000. The figures vary wildly, but one thing stands true for any number: it is only worth spending if it buys the right thing.

The number you actually need

Finding data on total product development budgets is extremely challenging. This data is often very commercially sensitive and usually varies wildly as you don't know the challenges that were faced in developing that particular product, or what was included in the scope of work.

As a design and manufacturing agency we have a unique perspective on the range of budgets our clients allocate to get their product to market, and to put it plainly, almost all of our projects land in the £50,000 to £350,000+ range when considering their entire lifecycle. These are the figures that we advise most clients to have in mind when considering the total cost to get from Discovery through to a completed Design for Manufacture (DFM) stage.

These figures are backed up by two of our competitors, who quote £30,000-£500,000+ and £60,000-£400,000 for their typical projects.

There is only one unbiased resource available to back up these figures, and that is MIT's entrepreneurship centre which quotes a design agency as charging $200,000 to $300,000+ (£150,000 to £220,000) to productise a startup's wearable.

To tie this back into the £15,000 mentioned in the introduction, this will likely account for 10-35% of a typical project's initial development budget (Embedism's stages 1-5, covering Discovery, Feasibility, Concept, Engineering Development, and Design for Manufacture). So enough to cover the initial work, but not enough to finish a project.

But how should you size your budget at all, how do you know if £15,000 is what you need to kick off, or if you need £50,000?

Match the budget to the size of the prize

If you are approaching a design agency, either as an established business or as a startup, it's because you believe that you have an idea that you think makes financial sense - and will maybe even set you up for life.

An idea that works and reaches commercial success must be worth many more times than what it costs to develop. This is what investors look for. It follows, then, that the budget should meet the size of the prize.

The budget referred to above is the total budget. Reality can be cruel and not every person or business has enough funds available immediately in order to take their product to market.

£2,000 is not going to get your consumer product to market. Not with Embedism, and not with anybody else.

This gives most founders and businesses two routes to get the cash to fund development: raise enough in one go to fund it outright; or stage the work and raise incrementally, with each stage giving confidence in the next.

One thing is clear whichever route you choose - spending £2,000 on a partial build of a product doesn't work for anyone.

Where to spend it?

At the early stage of a project, there are four blockers that we typically see. Usually only one of these is blocking you, so identify which it is and then focus your initial budget on resolving that.

(a) You don't know if anyone wants it

Risk is commercial, not technical.

• Spend on appearance models. CAD, renders, one or two physical parts that look and feel right and do nothing. These can be presented to potential customers as part of focus groups to get initial feedback and market validation.
• Approximate costs: £5,000 to £16,000.
• This doesn't get you any proof the product can be built, nor any indication of unit costings, but if there are other products in the market that have similar features then you can address these shortcomings through risk management rather than evidence.

(b) You don't know if it can be built

A specific technical question you can't answer: battery life, range, sensor accuracy, whether it can fit in the enclosure, whether the mechanism works.

• Spend on a feasibility study, and a basic technical concept if needed. This could be made from development kits, bench test equipment, representative loads and/or test conditions. The purpose is to understand an unknown and de-risk development.
• Approximate cost: £5,000 to £20,000.
• This doesn't get you anything to show a customer, but might be of interest to investors.
• There is always a risk of failure with a feasibility study, but it is better to find out after spending £5,000 than after £50,000.

(c) You don't know what the product should really be

You just have a rough idea, and no spec, meaning every design decision is a guess or based on "gut feel".

• Spend on a works-like prototype real users can hold and use, built from off the shelf modules and 3D printed parts. Then write the user journey from watching them use it, and infer the real product's specification from this.
• Approximate cost: £15,000 to £35,000.
• Doesn't give you: a real product, and certainly not anything manufacturable - you are buying information, insight, and knowledge about your customers and your future product.

(d) The market is proven and you need to raise

• Spend on one integrated unit that looks and works well enough to survive a live demo or be used by you in controlled conditions.
• Approximate cost: £20,000 to £55,000.
• Doesn't give you: repeatability. One unit that works is not a design you can build ten of.

How to pick

• No evidence of demand: do (a). Cheapest, and it unlocks the funding for everything else.
• Demand plus a big technical unknown: do (b). Never design around an assumption you haven't tested.
• Demand and no technical fear: do (c). Spec quality is the biggest single predictor of whether a development lands on budget.
• Only do (d) once the first three are genuinely answered.
• Doing two of these badly is worse than doing one properly.

Contingency

You should always plan for the worst too, and allow for a 20% (for a well-scoped project) to 50% (for novel technology with lots of R&D) contingency budget across your project, especially in the early phases. Overruns in the early phases are usually a sign that you are solving a tough problem, or spending more time (and money) in working out exactly what your product needs to be - this will pay for itself many times over in the later stages where the cost of change drastically increases.

Where the money gets wasted

There are a few areas that we commonly see clients trying to rush towards. We always ask "why", and usually the reason is the illusion that doing these things will push the product closer to launching. Sometimes these pay off, but more often than not, the risk far outweighs the reward.

• Going straight to a custom PCB. Lots of our clients come to us with dev kits in a box, and a custom PCB can be a great solution for them. However if there is no hardware to speak of, often the dev kit route is a smarter first step.
• Tooling before the design is fully validated across all disciplines. Tools can easily cost £10,000, £20,000, £50,000 or even £100,000; they are a huge investment. You wouldn't think that a firmware issue could have a knock on impact on the plastics, but if an extra reset button is needed to make the product boot-safe then the enclosure has to change!
• Certification too early. Full CE and FCC for a connected product can run from £10,000 to £30,000 and only applies to a frozen design. Pre-compliance screening during development is the cheap version, and we find that it saves multiple weeks per project.
• Chasing unit cost savings early on. At 100 units development will far outweigh unit cost, and if you are low-volume manufacturing the labour will often significantly outweigh savings of a few pence across the BOM. It is probably best to invest in Design for Assembly/Reliability at this stage.
• Spreading it too thin. £15,000 across a bit of electronics, a bit of mechanical and a bit of firmware gets you three unfinished things that won't move your project forward.

There are plenty more traps in product development, but these are the ones that are very easy to fall into - that's why it is always a good idea to speak with your peers (or design agencies!) to get an idea of how to best approach the development of your product and not just the product design itself.

Checking a quote that looks cheap

When money is tight it can be tempting to go with the lowest bidder. This might pay off, but here are some tips on how to spot if a quote is too good to be true:

• Is firmware development included? In the age of AI coding it can be tempting to think you can do this yourself (and maybe you can), or that someone without specialist firmware knowledge can AI-code their way to a working product - this is very unlikely to work out well. Check that the quote you have received has firmware development included and that the person or team delivering it is up to the job.
• Have any design revisions been budgeted for? PCBs often need 2-4 respins, 3D prints need iterating on, and CMF specs need tweaking to get the product feel you want. Check that these are included in the quote. We quote for a number of revisions that is appropriate for your project with extras billed on top. And remember that respins cost time, and not just money.
• Is there a handover or ongoing support included? Not essential, but some projects need it more than others.

What next?

£2,000 won't get your product to market, but it might buy you a written technical spec, an honest opinion from an experienced engineer on your concept, and a costed route to manufacture.

Tell us what you are trying to build and what you are trying to find out at your current stage. Then we can tell you what the smallest piece of work is to answer that question.