Hey guys! So, I wrapped up my ZX Spectrum crowdfunding project, and boy, it was quite a ride! The campaign went gangbusters with 146% funding, and I shipped everything out. Total revenue was £12,839.57, but after costs, I pocketed a neat £4,273.91—though, spoiler, that's before considering my labor! There was quite a learning curve understanding COGS, pricing strategies, and the hidden costs of hardware projects. It took about 13 months from idea to fulfillment, and I amassed 436 hours of work! I discussed vital strategies like working with distributors and how PCBs' cost plummets with bulk orders. Plus, the unknowns, like tariffs, which almost threw a wrench in the works. Stick around for the deets, and maybe you'll pick up a tip or two for your own crowdfunded adventures!
Some of you may be aware that I recently
successfully completed a crowdfunding
project. Everything’s been shipped out
to backers, so it’s a complete success.
Or was it? What was it? It was a
recreation of my childhood computer, the
ZX Spectrum. It’s pretty nice. It’s the
same size as the original. The keyboard
uses fullcolor silk screen printing and
it’s all capacitive touch. There’s a
built-in screen and speaker. It’s a
pretty cool bit of kit. So, I put the
project up on crowd supply last year and
we ended up smashing through the funding
goal and getting 146% funded. Not bad at
all. So, I know the big question you’re
going to have. Did I actually make any
money from this adventure? So, let’s get
into that first. I’ll cover it really
quickly. So, if you don’t want to watch
the whole video, you’ll have your answer
straight away. So, total revenue from
the project, £12,83957.
After subtracting all my costs, in
theory, my gross profit was £4,273.91.
But there’s quite a few caveats to that,
as you’ll see as we run through this
presentation. So, let’s dig into it all
and see how it works. Now, before I
started selling things online, I used to
wonder why do hardware products cost so
much when components and PCBs are so
cheap. So, let’s answer that question
and we’ll hopefully pull the curtain
back on what’s involved in a
crowdfunding project. Now, there’s a
rule of thumb that you should be
charging a minimum of two to 2.5 times
the cost of goods sold, often
abbreviated to COGS. Cogs is what the
goods cost you. There’s a great post
from Dave Jones of EE blog fame that
digs into this. I’ll do a link in the
video description, but you need some
margin to cover your costs and any
disasters that might happen. Now, how do
you actually calculate your retail
price? So, you can use bottom up
pricing. You base it on your COGS plus
the margin you need or you can do top
down pricing. You work out what people
would be willing to pay for your product
and you use that as your price. Now, in
general, you’ll end up using some
combination of these two approaches. One
caveat is if your top down price, what
you think people are willing to pay,
ends up being less than your bottom up
price, you really have to think
seriously about whether your product is
viable. Some things are just not worth
doing.
Now, how long did the project actually
take? It took just over a year. So, 13
months in total from initial idea to
shipping to backers. So, I had the
initial idea for this project back in
June the 14th, 2024. That’s when I got a
fullcolor silk screen printed keyboard
and it semi worked. We launched the
campaign November the 21st, 2024. We got
the funds from Crowd Supply January the
17th 2025.
Completed all the manufacturing and
packaging June the 4th, 2025.
And we shipped all the orders to backers
on July the 14th, 2025. So that’s it.
End to end, 13 months exactly. So let’s
dig into the numbers. So campaign
revenue, there’s quite an important
distinction on crowd supply that I’m
sure some people are not aware of. If
you order during the campaign phase, the
creator gets the full price, $99, minus
various fees. For pre-orders, post
campaign, Crowd Supply placed a house
order, and that’s done at the
distributor price, which in my case, we
worked out as $56.
Now, this house order from Crowd Supply,
they placed quite a big order. So, you
can see from the numbers here in the
campaign, I got 118 orders. And then
Crowd Supply placed another order for
another 118 units. It’s really
important. It pushes up the number of
units you get manufactured, which makes
a big difference to your unit cost. Now,
the fees during the campaign are also
quite important. We’ve got the breakdown
here. Fees per item worked out at
$18.76.
So, that gets knocked off the $99 price.
Now, here’s a couple of screenshots from
my dashboard on Crowd Supply. So you can
see the campaign orders 118 and you can
see the breakdown of the fees and you
can see crowd supplies house order at
the distributed price of $56. So that’s
all the numbers from Crowd Supply. Let’s
have a look at my cost. So here’s a
breakdown of my cost. Now it’s likely
I’ve missed quite a few things. I wasn’t
tracking everything religiously. Now I
had 300 units made. A few of them were
rejected at QA stage. I sent quite a few
samples out to people and I may have to
send some replacements out to friends.
Now, Crowd Supply ordered 236 units, but
you can’t just order exactly what they
asked for. There’s minimum order
quantities, and you got to allow for QA
failures, sample units, and
replacements. So, I ordered 300 units.
Now, this gives me a cost per unit of
$3060,
which is about $7 higher than my
original guesses for costs. So, gross
profit. This one is a bit tricky. We
have our sort of top level number of
4,273,
but I still have around 40 to 50 units
in stock. So, the final profit could
change depending on future sales, but
for now, I’ll ignore that and we’ll take
this as our gross profit. So, it’s not
too bad. Or is it? What about labor
costs? I put an awful lot of work into
this. So, here’s a breakdown with some
very rough figures of how long I think I
spent on the project. They’re very much
guesses. I wasn’t really tracking my
time, but we do end up with around 436
hours or 54 mandays. That might be a bit
high. I don’t know. I wasn’t really
tracking it. Now, if we look at UK
minimum wage, then our labor comes in at
quite a hefty £5,323.
So, our profit after the labor comes off
minus1,000.
But if I can sell all the stock in hand,
and I assume I can sell it for the
wholesale price, 30 actually get back
into profit. £1,000. Not bad at all. So,
how did I go about getting the ESP32
Rainbow made? Well, regular viewers will
know we got a long-standing relationship
with PCB Way. I put a link to them in
the description. They are pretty good.
They offer support for crowd supply
projects, and I was able to get a pretty
substantial discount, and it probably
knocked a dollar or so off my unit
price.
Now, how do you come up with a
recommended retail price? We talked
about bottom up pricing and top down
pricing, but you have to work out what
your bottom up pricing is. What is your
COGS before you’ve actually worked out
how many people will buy your product?
Now, you’ll see from the following
slides, there’s a lot of guesswork.
Crowd Supply use Mouser as a
distributor. Distributors want their
profit margin. Now, typically a
distributor wants at least 40 to 60%
markup on their side. This means that
your retail price needs to give you your
own margin and the margin for the
distributor. Now, here’s my guesses on
prices. Now, the really tricky bit is
you don’t know how many people are going
to order your product. So, you don’t
know what the actual cost is going to
be. More people ordering gives you
cheaper costs as components and PCBs can
be become cheaper the more you order.
Now, you can get some of these numbers
from online quotation tools, but with
components, you need to make a few
guesses. Here’s my rough guesses. At the
start of the project, I’d made a few
prototypes, so I kind of knew how much
component costs might be, and I could
get the PCB costs from PCB Bway online
quoting tools. Now, it’s pretty
remarkable how quickly the cost of PCBs
and SMD assembly go down the more units
you order. But you can see below a
certain volume, it just does not make
sense. So, below around 50 to 100,
there’s no point me even doing this
project. Now, as the volume goes up,
component cost will start to dominate
the total. So bomb optimization becomes
really important at higher volumes.
Crowd Supply give you this really handy
spreadsheet to see if your project is
actually viable. You can play around
with the figures in yellow boxes and see
what actually works. Gives you a good
idea of how many units and what price
you need to set to be viable.
Now you also need to work out what to
charge the distributor price. So Crowd
Supply will use this for their house
order and they’ll use it for any
subsequent orders. So you have to make
some guesses here. Again, these are my
finger in the air calculations. Now,
this table is based on RVRP of $99 and
it’s based on my estimated cogs. Given
my actual cogs, my margin is actually
quite a bit lower. My original target
price was $49.99.
It felt very sincere to go for this
price, but it would have meant me losing
a lot of money. Now, something we do
need to talk about is tariffs. This
could have completely scuppered the
project.
One thing no one knew at the start of
the campaign is what was going to happen
with tariffs. I started this project
before the US elections and Trump was
already talking about tariffs. Now
something that probably isn’t clear is
that tariffs are calculated on what the
importer is paying for the goods. It
doesn’t necessarily matter what the
underlying cost is. So that means it
should be calculated on what Mousa paid
me. £12,83957
doesn’t matter what I paid for my goods.
It’s what the importer is paying for the
goods. And since I’m outside the US,
it’s the total price that Mousa paid me.
So, who actually pays the tariff? Well,
we have two options. DDP, delivery duty
paid. That means the seller pays all the
import duties VAT and customs clearance.
Or you have DAP, delivery at place. That
means the buyers are responsible. Now,
Crime Supply stipulate DDP for shipments
to Mouser. So, that means I’m on the
hook for all the tariffs and duties.
As you can see from the table above,
this had the potential to cost me a lot
of money. At a certain tariff level, I
would have had to cancel the project and
issue refunds to everyone. Fortunately,
I was able to argue that the product was
sufficiently transformed in the UK to
justify a change of country of origin to
the UK. But that’s entirely at the whim
of the customs guys, and next time I may
not be so lucky.
Let’s talk about certification.
There’s a lot of confusion and incorrect
information around C certification. Many
people will quite confidently tell you
that you don’t need certification for a
hobby or dead board. You don’t need
certification if you’re using a
pre-qualified module. Now, neither of
these are actually true. If you’re
selling your hardware to consumers, it
needs to be certified. And even if your
module is already certified, you still
need to certify your build of it. Now,
what do I actually mean by certifying my
hardware? Is there some special place
where you take your hardware and they
give you a big rubber stamp of approval?
Well, not really. CE is
self-certification. You, the
manufacturer, certify that your hardware
complies with various standards. Testing
labs give you the evidence to
confidently back this up. So, when you
self-certify, you are basically saying
to someone else, if you test my hardware
against these standards, it will pass.
Some things you can be pretty confident
of. You need to comply with restriction
of hazardous substances. This should be
as simple as making sure your
manufacturer does it in a compliant way.
Things like EMC, electromagnetic
compatibility, and RV radio equipment
directive are more difficult. My advice
is to get your board tested by a proper
lab. Now, depending on where you go,
this can get expensive quickly,
especially if you keep failing and
having to retest. I got some really
useful advice using the simple start
service from Smander. Based on my
schematics and intended use case, they
were able to tell me exactly what
standards I should be using. This meant
that when I spoke to the test lab, I
actually had a chance of understanding
what they were talking about. One really
useful tip was that unless I was
actually using the wireless capabilities
of the ESP32, I didn’t actually need to
get red certification. This made things
a lot cheaper. I was recommended a very
affordable test lab in China. Now, the
downside of using a lab in China is you
have to ship your hardware to China. And
of course, if the hardware fails, you
can’t just pop into the lab and try and
fix it. If this is a concern, then you
should find a local lab.
Now, how do you even get people to come
to your campaign page? Crowd Supply help
with this. There are already many people
who are back campaigns and received the
Crowd Supply newsletter, but it really
is up to you to market yourself. I’m
very fortunate. I got a semi-popular
YouTube channel, a newsletter, and a
blog. This gave me an existing audience
that I could tap into. I also created a
newsletter for this particular project
before I even launched on Crowd Supply.
That let me build my audience before I
was accepted onto Crowd Supply. So, I
had a readym made bunch of people
already keen for the project. I made
sure to keep this audience in the loop
on everything that was happening in the
project. So, updates go out on Crowd
Supply and also go out to my own list of
people. If you don’t have an audience
already, then you really need to get
yourself out there and do some
marketing. Social media has to become
your friend. Now, we also realize
referring to people as the audience
sounds absolutely awful, but I’m not
sure community sounds any better. Now,
I’m sure you’re thinking, “Where can I
buy one? This sounds great.” Well, it’s
still up on Crowd Supply. Please order
one. They’ve got stock right now, and if
they keep getting orders, I’ll do
another production run. Would I use
Crowd Supply again? Well, there are some
really good benefits. They do provide a
platform for doing crowdfunded pre-sales
campaigns, and they also add on their
own house order, so you get decent
volume, and you get the money up front
for a big production run. This is very
valuable. I would have had to seriously
think about whether I wanted to do such
a large production run without knowing
there was sufficient demand. Would have
been a significant investment on my
part. Almost £9,000 may not seem much
much to some people, but it’s a lot of
money to me.
You get all the distributor benefits.
So, crowd supplier deal with all the
complexities of selling to individual
customers. I don’t need to worry about
collecting sales tax, dealing with
chargebacks, all the problems that come
with selling to individuals. Nice thing
about having a distributor, I just ship
a bunch of big boxes to Mouser and they
handle it all. There’s a lot to be said
for being able to hand that off to
someone else.
Accountability. Crowd Supply have run a
lot of campaigns. They know what makes a
good campaign and what makes a bad
campaign. things like regular updates to
backers. Even if you think you’ve got
nothing to say, send out an update so
people know you haven’t just run off
with the money. They also really maybe
do realistic shipping days. They were
very careful to make sure expectation
with backers were set appropriately. See
my international shipping complaints
later. The downside is it costs money to
use the distributor. They’re not doing
it out of the kindness of their hearts.
They want to make a profit. This impacts
the amount of money you can make from
your products and has a knock- on effect
to how much you need to charge. If I was
selling direct, I could have probably
charged quite a bit less and still made
a profit. You have to do things
properly. You’re selling something
through a distributor. They care about
their reputation. Things like
Certification become important. You have
to ship with proper documentation and
safety information. It needs to be in
multiple languages. Most products on
Tindi don’t worry about these things.
Everything is shipped to the US and then
sent out again. It’s great if the
majority of your customers are US-based.
Not so great for the rest of the world.
Shipping becomes a significant amount of
money. And then related to the above,
tariffs. Shipping everything to the US
so it could be sent out makes sense in a
low tariff world. Starts to make a lot
less sense when tariffs are high. For
people outside the US, this could be
very bad. You could end up losing a lot
of money. So what went well? Working
with crowd supply was really useful.
Having someone keeping you accountable
and on track is invaluable. You could
definitely do this without them, but
you’d need good discipline.
Certification turned out to be a lot
less complex than I expected, mostly
because I spent some money and got some
advice. Don’t be afraid to do this and
make sure you allow for this in your
pricing. Now, there’s a really great
community forming around the project and
that’s really nice to see. Making
everything open source is the way to go.
Don’t be afraid of showing your terrible
code to people.
What went better than expected? I was
very lucky. My hardware just worked.
Every project I made moved things
forward. The firmware just seemed to
fall into place without a huge amount of
effort. AI tooling definitely helped
with the website and building out all
the supporting code. There’s no way I
could have built everything without the
help of AI.
What didn’t go so well? Oh my god,
international shipping. I had no idea
this would be one of the most
frustrating, opaque, and mysterious
parts of the project. I have posted
things internationally before, but
nothing of this value. For over a month,
I spent my time looking at the tracking
page, wondering why the boxes hadn’t
moved for days. I’d ra support requests,
and that seemed to magically unstick
things, but I had no idea why they were
stuck in the first place. Now, if you’re
in the UK, make sure you get yourself an
EOI number. There’s nowhere to put it in
any of the shipping forms, but you need
it on the paperwork. So, would I
recommend this? It’s definitely an
experience. Pricing is so difficult,
though. You need to make sure you can
sell at a high enough price to make
things work, but a low enough price that
people actually want to buy it. Lean on
crowd supply. They’ve got lots of
experience with crowdfunding and
delivering products. They know what
works and what doesn’t. Was it
financially worth it? As with most fun
things, if I’d worked those hours for my
normal consulting rates, I’d have made a
lot more money. I think I made a small
profit, so I can’t really complain.
Would I do it again? Honestly, I
couldn’t really say for sure. I’ll need
to have a very compelling idea with a
very large customer base. Now, I’m not
going to pretend I come out of any
newfound wisdom or some great
understanding of the world. In the end,
I didn’t get rich, but I did build
something that real people are using and
having fun with, and that’s worth a lot.
So thanks for watching. Hope this is
interesting or useful. Full details are
in the blog.