Why New Hardware Startups Fail and How to Make Sure Yours Doesn’t

Why New Hardware Startups Fail and How to Make Sure Yours Doesn’t

Learning from your own mistakes is usually just a part of the process of developing and selling a new product. But what’s much more efficient is if you can instead learn from the mistakes of others.

Studying the failures of others can actually ensure that you don’t make the same mistakes they did.

When I brought my own hardware product to market I definitely made a lot of mistakes. However, I gained so much valuable knowledge from both the things I did right and the things I did wrong. Helping others avoid these mistakes is my primary focus now.

Get your FREE Ultimate Guide - How to Develop and Prototype a New Electronic Hardware Product in 2024

When it comes to developing and selling a new hardware product, there are so many possible mistakes to be made. There’s a reason behind the cliché “Hardware is hard!”

You would be wise to also study in detail how other new hardware businesses have gone wrong.

Over the years I’ve worked with a lot of hardware entrepreneurs, and I’ve been one myself. I’ve also spent considerable time reviewing autopsy reports for hardware startup failures.

The top reasons new hardware businesses fail can broadly be divided into three categories: technical, financial, and sales/marketing.

“I’ve found learning from failure is one of the most powerful tools we have to help hardware companies prosper.”
– Ben Einstein, Founder of Bolt.io (Venture capital firm specializing in hardware startups)

Technical Reasons Why Hardware Businesses Fail

Technical issues generally occur during the development and early manufacturing stages. The more complicated the product, the more challenging it will be to avoid mistakes during these stages.

Underestimating Product Development

Product development is complex, expensive, and almost always takes longer than anticipated. Even big tech companies developing new products always struggle to estimate the time and cost required.

I’ve worked on many large product development teams and it’s extremely rare for any project to go better than expected. It’s always the other way around.

There will always be unexpected challenges regardless of the experience level of the developers. Creating something completely new is never simple. If it were easy, then everyone would do it and every market would be saturated with products.

Underestimating the complexity of scaling to mass manufacturing

Entrepreneurs tend to completely neglect the complexity, cost, and time needed to scale their product from a prototype to mass manufacturing. In fact, I’d say this is the most forgotten step in launching a new hardware product.

A group of chefs in a kitchen Description automatically generated with medium confidence

From a very early development stage you need to be thinking about how to design your product so it’s easy to manufacture. This is called Design-For-Manufacturing (DFM).

The sooner you begin incorporating DFM practices into your design the easier and less costly it will be when it comes time to scale manufacturing.

For most electronic hardware products the plastic enclosure is the most complicated part to scale to mass manufacturing. This is because completely different technologies are used for making plastic prototypes (3D printing) versus production units (injection molding).

Injection molding requires adherence to very strict design rules, so most plastic pieces require considerable modifications to prepare them for injection molding.

Insufficient quality testing

You’ve put in enormous amounts of time, energy and money to get your product ready for market. Don’t ruin your opportunity by shipping faulty products to your customers.

In the early stages of manufacturing it’s definitely best to err on the side of too much quality testing. This is true even if it means you won’t make a profit initially.

A sure way to kill your startup is to ship a bunch of defective units. Of course, a small percentage of flaws will always make it past your testing, no production process is ever 100% perfect. But you should strive to reduce the defect rate to about 1%.

Product is too complex

You may have heard the phrase feature creep before. It’s a common mistake made by a lot of entrepreneurs.

You’re passionate about your product, so you want it to be perfect. This means not only must everything work perfectly, but your product must also have every conceivable feature.

This can be a great strategy if you have unlimited capital and unlimited time. Maybe in ten years you’ll have your perfect product. Sorry for the sarcasm but I just wanted to drive home the point. Obviously, no one has unlimited money or time so this strategy is almost always a bad idea.

Instead, I encourage the strategy known as Minimum Viable Product (MVP) which was made popular by Eric Ries in his bestselling book The Lean Startup.

Start by developing the simplest version of your product with only the core required features. Get that version on the market, so you can begin to gather feedback from real customers.

Then, use that feedback to create the new and improved version 2.0 of your product. This is a strategy commonly employed when developing software, but it also works great for hardware.

I find that it’s also helpful to begin by listing every feature you’d like to include. Then, do the work to estimate the cost for each feature. Based on your results, you can then make a better decision on which features you should include.

I recommend that you work with experienced product developers to help simplify your product before you begin full development. This strategy of product simplification will allow you to get to market much faster and cheaper with a higher chance of success.

Since simplifying your product lowers the engineering costs, these developers should be independent from your primary developers in order to avoid any conflict of interest.

In fact, I always suggest you work with independent developers to provide you unbiased advice. They can also review the designs from the primary developers.

Get your FREE Ultimate Guide - How to Develop and Prototype a New Electronic Hardware Product in 2024

Overpromising to customers

Don’t promise something until you have it in your hands. Regardless of when your development team says something will be ready, it’s best not to promise anything until you have tested the product yourself.

On the other hand, it’s good to have customers involved as early as possible. I’m not advocating that you ignore customers until you have inventory ready to ship.

Keep them updated on your progress, and perhaps give some approximate dates. Just don’t overpromise on what you can deliver.

Financial Reasons Why Hardware Businesses Fail

Many entrepreneurs think if only I had more money I could quickly and easily launch my product. There’s no doubt about it, money is a huge obstacle for most new businesses, especially those developing new hardware products.

Fortunately, not having lots of money can actually be a positive. This is because it forces you to make smarter choices. That being said, you still need money to get started.

Generally, you’ll need to fund at least the initial stages of development. You should take the product far enough that you can get others interested in investing (unless you plan to bootstrap yourself all the way to market).

Text Description automatically generated

Ran out of money

This may be the most common reason why hardware startups fail, although it’s rarely the fundamental reason.

For example, if you run out of money because you underestimated the development cost, then I’d say the problem wasn’t that you didn’t have enough money, the problem was you underestimated how much money you needed.

Most reasons for failure eventually lead to running out of money. Perhaps you develop the wrong product that no one wants, and you don’t have enough money to create the product people actually want.

Your problem wasn’t a lack of money, your problem was not knowing your customers.

You absolutely must be strategic about this whole process so you can avoid or delay many of the major costs. This means don’t start by spending thousands of dollars on a patent or on developing an expensive custom prototype.

Instead, you need to focus on understanding what lies ahead, so you can plan out how to surpass your financial obstacles. To succeed you absolutely must be strategic.

Insufficient profit margins

When doing small batch manufacturing in the beginning your profit margins will be low or even non-existent. This is because you’re not yet getting the advantage of economies of scale. As your manufacturing volume increases, your cost per unit will decrease significantly.

Once you have manufacturing (and sales) flowing smoothly, you’ll likely begin spending a considerable amount of time striving to lower your production costs without sacrificing product quality. This is how you increase your profit margins.

In the early days, a low profit margin may be acceptable. But you have to eventually make a profit to be able to sustain your company, yourself, and your family.

Even in the beginning having low profit margins will make growing your startup much more challenging because you won’t have any significant profit to invest back into your company.

You need to know your Cost Of Goods Sold (COGS) at various production levels in order to forecast your profit margins. COGS is the total cost to manufacture and ship your product. It’s key to determining your profit margin.

Without a good handle on your production costs you’re more likely to run into problems in the future with low profit margins.

A picture containing text, building, warehouse, scene Description automatically generated

Underestimated production/inventory cost

Underestimating the production cost is one of two elements leading to the low profit margins I just discussed. You only have two knobs to turn in order to adjust your profit margin – either decrease your cost of goods sold (COGS), or increase your sales price.

If your profits are lower than expected then you’ve either underestimated the production cost, overestimated how much people are willing to pay for your product, or both.

Also, if you underestimate your manufacturing cost then you’ll also underestimate the cost of inventory. Once you have manufacturing setup, inventory costs will be your biggest obstacle.

Overestimated sales price

This is the other side of the coin from underestimating the production cost. If you overestimate your sales price then your profit margins will be lower than expected.

Unless you can make up for this by lowering your cost of goods sold, a lack of profit margin will kill your startup. A company with low margins is very difficult to push forward toward sustainability.

Cash flow

Cash flow is a huge problem for hardware startups. It’s likely the biggest problem after getting manufacturing up and running. Most manufacturers require a partial payment upfront with the remaining amount due before shipping.

If you plan to sell your product directly to consumers via your website then the time it takes to get paid for those units depends on how fast you sell your inventory. Keep in mind that it will most likely take months for you to sell those units.

If you instead plan to sell your product through retail stores, then you can expect to wait a minimum of 30 to 90 days to get paid.

In both of these scenarios there is a length of time between when you spend money on inventory and when you finally get paid. This creates a serious cash flow problem. You will need a considerable investment to keep your hardware startup afloat during these early days.

For example, let’s say a big retailer wants to purchase $1 million dollars of your product. This means that you will need to finance around half that, $500k, for 2 to 4 months.

Purchase order financing and invoice factoring are two common ways for you to finance this amount.

Although my favorite inventory financing option is called manufacturer financing, which is what I used for my own product (read the details here). Basically, the manufacturer agrees to let you pay them after you get paid by your customers. This can essentially eliminate your early cash flow issues.

Sales/Marketing Reasons Why Hardware Businesses Fail

Most entrepreneurs focus all of their effort on product development while neglecting the most important part of launching a new hardware product: selling it. Without sales nothing else you do really matters.

For many entrepreneurs the idea of actively selling is downright scary, especially for more technical entrepreneurs. I know, because it was terrifying for me, and it forced me outside of my comfort zone.

But to be a successful entrepreneur, you absolutely need to embrace going outside of your comfort zone. If an introvert like me can muster the courage to call on large retailers, manage a team of 20 sales people, and run a tradeshow booth, then so can you!

Failed to know their end customer’s needs/wants

Don’t mistake positive comments from potential customers as proof the product will sell. Just because someone says they like your product, or that they would purchase it, doesn’t really mean they will purchase it. Nothing speaks the truth like actual sales.

This is one of the biggest challenges with hardware products. You need sales feedback as soon as possible, yet it takes considerable time and money to get a hardware product ready to sell. It’s a catch-22 situation.

Fortunately there are strategic ways to get this valuable feedback without spending a fortune.

My favorite solution is crowdfunding where you can get people to “vote” for your product using actual money before you have it ready for market. However, you still need to do considerable work on your product before you’ll be ready for a crowdfunding campaign.

Do you know what the easiest and cheapest, yet least common method for getting customer feedback is?

Shockingly, it’s to simply talk to them! Unfortunately, talking with customers seems to be something new entrepreneurs seek to avoid.

You need to talk to customers as early as possible, and it’s not something you should delay until after you have something ready to sell.

Didn’t build a audience around their product

Stop what you’re doing and begin building an online audience around your product right now! This allows you to grow your audience slowly in the background while you develop your product.

This is what the founder of Pebble Technology did. Eric Migicovsky built up an email list of around 6,000 subscribers during the years he was developing the first Pebble smartwatch.

Having this email list was key to raising over $10 million from his first Kickstarter campaign. Without an email list of interested customers you will never have a successful crowdfunding campaign.

You need this audience to help you develop a product that people actually want to buy and to help raise money through crowdfunding if you go that route. Later on, you can then sell your product to this audience.

You should strive to collect as much feedback as possible as soon as possible. There is no better way to do this than by building an audience interested in buying your product.

See this blog article and this course on building an online audience.

Get your FREE Ultimate Guide - How to Develop and Prototype a New Electronic Hardware Product in 2024

Didn’t begin sales/marketing soon enough

So many of the entrepreneurs I work with focus solely on developing their product. The more technical the founders are the more this is true.

If you are a technical founder, your best bet is to bring on a co-founder with marketing experience. Startups with two founders, one with a technical background and one with a marketing background, is an ideal combination.

If bringing on a co-founder isn’t an option for you, then you will need to embrace online marketing yourself.

“A maker, a hacker, and a hustler”


“A maker, a hacker, and a hustler”
Brady Forrest, head of the hardware accelerator Highway1 and co-author of the excellent book titled The Hardware Startup: Building Your Product, Business, and Brand, describing the ideal early team for a hardware startup


I highly recommend that you bring on an independent sales representative from your industry as soon as possible. Independent sales representatives work on a commission basis so you normally only pay them once you’ve been paid by the customer.

This is an ideal situation that really lightens the load on your cash flow. This is especially true for solo founders who don’t have a marketing expert on their team, or for teams without marketing experience in their particular industry.


It’s a foregone conclusion that you will make some mistakes in route to developing and selling a new product. But you can avoid many of these mistakes by learning from the mistakes of others.

This requires analyzing how other startups have failed and working with people that have already developed and sold new products.

There is nothing more valuable than being able to learn from the mistakes (and successes) of others. That’s how you make the fastest progress!

Other content you may like:

5 4 votes
Article Rating
Notify of

Oldest Most Voted
Inline Feedbacks
View all comments

Developing hardware isn’t an easy process, I feel you


Now I know the nature of my cross Hardware development, indeed it hasn’t been easy


Great insights and sure it will help the upcoming entrepreneurs. Looking forward to more such informative blogs. Kudos to the author.


Great insight!

jane fraser

Startups now days should create a plan so that they don’t face any failure for these reasons

Carina Rogers

Thats the great reasons which would help startups to take precautions without failing


Great list.

I’d add “picking the wrong manufacturer, or involving a manufacturer too early”. A typical OEM or ODM supplier in China wants to get to mass production ASAP and they quickly lose patience when a new product development is slow, the customer changes direction, etc. A manufacturer can be the wrong choice for many reasons (e.g. they are actually a trading company, they are an ODM and they are interested in actually getting their hands on your IP, they said they have a MFi license but they actually don’t, and so on).

And I would not mention “quality testing”. That’s very general. I’d distinguish between compliance testing and reliability/performance testing. For certain products there is overlap, for others it is really very distinct. The most likely to be overlooked is reliability testing.

Elmar Joydt
Elmar Joydt
Reply to  Renaud

In a university course we learnt that “quality” is defined as the degree of requirement satisfaction by your product’s implementation. These requirements very much depend on your specific product. It’s not just performance or “reliability” (service lifetime or correct functioning I suppose?) but also energy, safety/security, real-time and product-specific requirements (“it should do X after Y time…”) etc. They can be very different depending on high-end tech, low-end tech or the degree of digitalisation in your product. The technical term actually is “Quality Engineering” or “Quality Assurance”. In our course we mostly learnt tools for proving (verifying) the correctness of an implementation considering a formal specification, for hardware and software. And yes, you cannot verify customer fitness.

I think, this article helped me a lot to recap on common advices I got to know and focus on important issues.

John Weiss

“A maker, a hacker, and a hustler”

What’s the difference between a ‘maker’ and a ‘hacker’?

John Weiss
Reply to  John Teel

Thx. I’m hearing hardware hobbyists in SF use the term ‘hardware hacker’ for a few years now. Also ‘social hacker’. Basically, repurposing, upcycling, or even pranking systems, including hardware, seems the current usage. So that quote confused me. Cheers!

Ravindra Moholkar
Ravindra Moholkar

Profound insights John. Thank you. May i add another important point (that i am likely to encounter), statutory/regulatory requirements and certifications. This may look like a given, but when it comes to sales and operations in markets not domestic for you.

Dave Millman
Dave Millman

This is the best single summary of failure modes for hardware startups on the web. Founders tend to know hardware design, as opposed to manufacturing, marketing, sales or distribution. Great post!

mohamed elaouni

hi John tell
thanks for your advices , your site is full of wisdom and knowledge.
i wanna tell you that iam doing these kind of electronic hardware devlopement in a far way marginalised small tawn in morocco north africa ,
Iam stragling alone, without any results or progress in my strat up , no funds, no market,
i built some electronic devices but selling just fews units, like alarms systems, gates and portail remotre control systems, Queeing system control…
Its difficults my friend…
i didn’t found what doing to push things ahead.
nexet moths i will get crazy….
thanks mr John.


Great article. Wish I found this earlier.

I made all these mistakes with our tech start up. In truth, it’s much harder to pivot as a hardware tech start up. We made a product for one particular market, but in rolling out, realised the market that was more interested was the medical space. We tried to “hack” both the software and hardware to suit the market, but though there was traction, the initial experience wasn’t great. But it gave us validation, but we ran out of money.

I’m now starting again, and hoping to learn from my mistakes.

One thing I’m exploring; is it possible to purchase/find used plastic moulds that suit our casing needs (maybe with just a small modification), rather than completely start from scratch? Plastic tooling was one of the things that caught me completely off-guard (and we didn’t design it with manufacturing in mind). Something I’m hoping to learn from again.

Luke Stavrowsky

On the subject of building a community (email lists) around my product to increase success on a future crowd funding campaign, I was wondering how I do so with a new technology that can easily be stolen or marketed by someone who reads my blog or website information? Could I build a subscription base without detailing the technology i’m using, or would people tend to dismiss my concept as unworkable?


Hi, I totally agree with Joel the introduction of IoT has forced the hardware guys to learn software tools and connectivity tools as well. What I am finding difficult is the low shelf life of technologies, like say if we opt a technology and want to design a product based on that, then by the time we are finished with the development and other work new technology comes into the market, this is creating problems for the startups as generally startups do not ve a big technical team. the other problem I myself facing is to decide on what basis a product has to be manufactured? Quality or Quantity, Latest or simple technology. as far as my second point is concerned I personally feel that a product has to be simple to use no matter what technology is used, but certain people still feels we have to introduce latest technology. what is your opinion on this, please do share your views.

Chad Sayers
Chad Sayers
Reply to  John Teel

John, What a GREAT article! We have seen and made so many of those very mistakes you have documented from yourself and the failures of others! The great thing about “hardware being HARD”, is that fact that those who master it, typically become behemoths of industry such as Apple, Belkin, HP, Samsung, IBM and others. We hang in there because we know we are the best at product features and are honing our specific products as we prepare for launch. But what an epic battle over the years to review, analyze, modify and then do it all over again… again and again and again to get it right. Exhausting in every sense of the word, but worth every learning and changing moment. We know that true champions in any discipline or industry are marred by loss, leveled by disappointment and devastated by disaster until they learn to rise from their own ashes a phoenix. So, on we march to our rise with our hardware mantra of do it right and do it well… until its done.


Number 3 has an error.

That your sales rep became a stockholder is a bad thing in general.

Marvin Motsenbocker

I am a pure “technical” guy who is scared of marketing and just starting. This answers my basic questions with extremely good advice. thanks

Joel Carter

This list is spot on. There’s nothing easy about hardware design, development and scaling to mass production. I’ve recently seen two really good products from entrepreneurs who failed to take into consideration long lead components and availability. They prototyped with parts from Mouser or Digi-Key but failed to consider sourcing and locking down large quantities of key components. They failed to plan for success! There are a number of bear traps one can step into — a whole lot more than software. On that note, hardware has now become even more complex because of the critical nature of IoT and connectivity software. I used to think of myself as a hardware guy but now realize I need the whole end-to-end connected solution including cloud integration and apps. Well written article full of truth.

Copyright 2024 Predictable Designs LLC.  Privacy policy | Terms
  10645 N Oracle Blvd, Ste 121-117, Tucson, Arizona 85737 USA