Here’s a question we get asked a lot:
How do you start a tech company with no money?
Well - let’s just kick things off by saying the average survival rate for a startup is around 10%. That’s right -- 90% of startups are shuttered before they grow enough to sustain themselves.
As research from a large database of startup post-mortems shows, failure often centers around a few major mistakes:
Almost half of failed startups shut the doors because they didn’t make a product people actually needed. Some failed because they ran out of cash, others because of an ill-equipped team.
While the stats paint a bleak picture, knowledge is on your side. By knowing the biggest traps to avoid, and having a solid gameplan in place to build, market and sell an app people actually want to use, your startup has a fighting chance for survival.
So, where do you start?
In this article, we’ll cover an actionable 7-step plan to start a successful tech company:
- Build an MVP the market wants
- Validate the app with early adopters
- Iterate to meet product-market fit
- Build a skilled and unified founding team
- Get the funding you need to grow
- Develop and practice an agile methodology
- Generate funding and scale team
This guide is rooted in lean startup methodology, so let’s look at an introduction to how winning startups think before diving into the step-by-step guide.
Don’t Overcomplicate, Start Lean
For a tech company to make money, you’ll need a market that’s ready to buy, the right feature set to solve the pains of those users, pricing your market can afford, and a hundred other things that you can’t possibly know at the outset without doing painstaking research and spending a truckload of money:
That’s why startups perform best when they’re grown iteratively, based on user feedback from a constantly-improving app. Eric Ries outlines this successful methodology in his book The Lean Startup. This radical guide to getting a low-cost technology business off the ground before you run out of money has become the startup bible.
To run a lean startup, don’t get bogged down in creating an LLC straight away. It’s unnecessary legal baggage.
Truth is: you can go into business with a general partnership and co-founder. This keeps your costs down, and saves you dealing with issues like organizational structure and taxes before you even have a product to sell.
In the lean startup methodology, the tech is far more important than the structures that support it. Your product is what’s going to make you money, after all, so start with the build.
If you have the skills or are willing to learn to code, you can start laying the foundations for a new tech startup as a side hustle, allowing the day job to pay the bills. This will give you an early prototype you can take to other potential co-founders.
As a non-technical founder, however, you need to sell your vision to a prospective CTO (chief technology officer) -- this way, the technology costs nothing but time.
Now that you have the bare business essentials in place, and at least one technical team member, it’s time to set the wheels in motion.
7 Steps to Build a Successful Tech Company
#1. Shortlist the core features of an MVP
Most great startup ideas are the result of a founder being unable to find a good solution to their pain. In the words of Paul Graham:
“The verb you want to be using with respect to startup ideas is not ‘think up’ but ‘notice.’ At YC we call ideas that grow naturally out of the founders’ own experiences ‘organic’ startup ideas. The most successful startups almost all begin this way.”
If you’re developing a product based on your own pain, you might be the most qualified person to list its core features. But, assuming that others share this pain and are looking for a solution, you can bring on early adopters to help refine the value proposition.
Nailing down the features, collecting feedback and validating the concept is critical to ensuring your product roadmap is pointed in the right direction.
Since 42% of failed startups close down because of poor product-market fit, making informed choices at this stage is critical to survival.
You can use a simple framework like this to define user goals and see how different pain points map to core product features:
- What is the overall idea?
- Who are the customers?
- Who are the end users?
- Why would they want it?
- Why are we building it?
You have to be very selective when building an MVP.
Focus on providing the minimum set of features users need to achieve a goal or realize value; your project management app needs attachment uploading more than it needs support for custom emoji.
3 mistakes to avoid when building an MVP
Overcooking your MVP
The idea behind starting small is to start quickly.
Your MVP doesn’t need to serve multiple audiences and use cases; it just needs to be validated by one niche. Get a small, highly-targeted segment of your market to buy into the idea and you get the cash needed to expand an MVP and take it to a wider market.
Choosing the wrong market segment
Knowing who to sell to is just as important to building an MVP with market fit. Getting an engaged community of early adopters to provide feedback doesn’t happen if you’re going after the wrong segment, and without those early adopters you won’t get the time and insight needed to iterate and grow.
Refusing to pivot if the MVP won’t stick
Even though you’ve done your research on the audience and feature set needed to compete in the market, it might be that your MVP is simply being marketed to the wrong customers or missing features that persuade customers to buy.
“Failure in validating your MVP only invalidates a small way to acquire customers — not the whole business model.” -- Swarnendu De
#2. Pre-sell the MVP
A lead capture page for the GrowthHacker Projects MVP, which funneled qualified companies into a concierge demo.
One way to combat the problem of running out of cash is to bring paying customers on board as fast as possible.
This gives customers a financial stake in the success of the product -- they’ll be willing to help “co-develop” the product in return for getting the features that they ask for (and will want to pay for) built.
Sales = validation!
One of the biggest inflection points for any company is finding out people will actually pay for what you have, or plan to build.
Here are the different types of MVP pre-selling strategies you can use to quickly validate a product concept:
- Single-feature MVP. Focus on nailing one user goal, to validate the need for the feature and build early adoption.
- Piecemeal MVP. Keep costs low by combining existing products and services to come up with a unique offering. For example, your product could be the result of some Zapier zaps or a code-free prototype built with Bubble.
- Concierge MVP. Do some or all of the software’s work manually and work closely with customers to understand how best to improve. With that information, automate the manual work with technology.
- Wizard of Oz MVP. An MVP that seems like a software service, but its results come from manual work. This is a way to validate the results your app will provide and whether customers will pay for them.
- Crowdfunded MVP. A lot of successful startups came from a video of a one-off prototype uploaded to Kickstarter -- with crowdfunding, you can generate buzz and get feedback on a product before you’ve put it into production. This keeps costs down while giving you money to survive on.
- Smoke test MVP. Validate demand for your idea by sending paid traffic to a pre-release signup page. The number of signups will help you gauge the level of interest.
#3. Source talent with equity
Not technical? No problem.
You’ll be able to find an excited technical co-founder if you have the right startup idea and equity offer.
You don’t have to spend money sourcing talent -- instead, use a platform like VentureStorm to get connected with your perfect CTO or CEO.
There’s a lot of debate over the matter, but startups typically offer technical co-founders around 10-35% in equity. This is because future funding will further dilute equity, so giving away a chunk close to half can be a risk for CEOs looking for investment.
#4: Acquire customers
Once the MVP has been validated and you have iterated with user feedback, the next step is to release the product into a wider segment of your target market.
You can meet all the right user goals with all the right features, but if no one knows your product exists you will lose out to a competitor.
In the world of SaaS, it’s a land grab. You need to take the oxygen out of the room.
For startups that don’t have enterprise budgets, large-scale paid ad campaigns and a well-oiled sales team are largely unattainable.
Lack of budget is definitely a disadvantage, but not a game-ender. It just means you have to get scrappy, and do hard manual promotion that doesn’t scale.
This often involves finding where your customers hang out online, and getting involved in the conversation.
Generating early buzz with online communities
Every startup remembers their first Hacker News home run or Product Hunt launch -- probably because the amount of traffic it blasted at the servers took them offline.
A popular piece of content or well-received product launch that shoot to the top of sites like Reddit, Hacker News, and Product Hunt can send tens of thousands of visitors to your site. It’s a tactic used as part of regular content promotion and bigger product/feature launches.
For example, below is a 2007 Hacker News relic from Dropbox founder Drew Houston:
Not only did this post get the attention of product experts, it also got thousands of users to weigh in on Dropbox.
For small sites that could mean a 60%+ increase in sessions overnight. Check out the Show HN category to see what’s hot, and try to replicate the formula.
Reddit is another site that consists of thousands of communities - everything from SaaS to data visualizations. It’s well-known that Reddit hates self-promotion, but submitting content that’s genuinely relevant and useful to a specific community can be a way to get vital attention in the early days:
This post links off to a promotional resource but gives the community the full text in a Reddit post -- this makes it easier for users to read and comment, and is seen as less spammy than simply dropping a link in the subreddit.
Four startups that grew by doing things that don’t scale
Apps like Tinder live and die by the size of their networks. It’s a hyperlocal app, so it needed to onboard a lot of users in close geographical proximity for the pilot period.
Tinder launched a series of popular parties in California and made it mandatory to have the app installed to gain entry.
People attending the parties installed the app, and Tinder’s user base grew to 15,000 overnight. The “network effect” baked a viral loop into the product, helping to retain existing users and attract waves of new ones.
In 2019, it was announced that Tinder had 5.9M paying users.
Developing a community-driven product without an actual community is a tough ask. In the early days of the site, Quora founders would start and engage in threads to provide an illusion that the community was a lot larger than it actually was.
Herd mentality took hold, and soon thousands of new users were flocking to the site to engage in the conversation. Eventually, it hit a critical mass and the founders no longer had to seed new conversations.
Generating a wide variety of question-answer content around different topics is also a way to rank in the search engines for specific long tail keywords. By seeding Quora with a range of frequently asked questions in the startup/tech/coding worlds, it became wiki for information that could be expanded with different community perspectives.
It’s vital at the start to provide a way for users to see how they should use the platform and why it’s beneficial to them – the Quora staff did this by leading by example.
Bonus resource: Check out step #12 in this guide to learn how to mine Quora for a high-traffic threads. This is a great way to engage in the conversation, provide value, and generate targeted referral traffic to your tech startup.
It might not be a game-changer for a mature product, but 72 signups can make a massive difference for an early-stage startup.
Twoodo managed to attract 452 unique visitors which converted at a rate of 16% with a strategy that involved the founder making targeted comments on 40 blogs. The whole project took just 6.5 hours, but helped the company land its first customers.
Stripe founders, now a hallowed duo in startup circles, got their product used by early customers with very hands-on tactics. The founders visited leads in person, where it was easy to gain access to their computers and do custom set-up work for the Stripe API (remember what we were saying earlier about how an MVP can be partly manual at first?).
“If you were interested in Stripe, they would set it up for you on the spot. They didn’t go back to their office and send you a link by email hoping you’d sign up. There was no choice. If you said, “yes” to try Stripe, they would grab your laptop and set it up for you.” -- Mikael Cho
Bonus resource: Want more inspiration? Check out this growth hacking guide that looks at the specific strategies and tactics 77 hyper-growth companies used to land their first paying customers.
#5. Look at the numbers
By pushing new users into the product funnel, you get data on how well the app performs in the hands of real customers -- do they stick, or churn out quickly? What changes need to be made?
While signups are great, but they only tell part of the story. If you can’t get users to adopt the product, you’ll leak revenue and find it near impossible to get funding.
Here’s a shortlist of the most important metrics every tech startup should be measuring:
Active users regularly return to use your product. They didn’t log in and leave never to be seen again -- they get real value from the product, and are one of the most valuable assets for your business.
How to calculate active users
Active users are usually analyzed by time period. So, your monthly active users (MAUs) are those who have logged in within the last 30 days. Every app promotes different usage frequencies.
Twitter hopes you use the product every day, because there’s always something new to see and it makes money from ad impressions. Something like Sumo is a tool users might need to use much less frequently, because once you’ve deployed a pop-up form you don’t need to log in and adjust it every single day.
To calculate active users, you’ll need a free tool like Google Analytics or a paid, dedicated customer data platform like Amplitude or Mixpanel. The calculations are done with user IDs and event tracking.
Retention rate is often low for apps. Small improvements make a big difference.
Retention rate is the metric you need to improve if you want to grow your active users. It measures the percentage of users that log back in after signing up. If your retention is low -- e.g., no one that signed up a month ago still uses your product -- you’re better off putting money into retaining customers than generating leads because it costs 6-7x less to retain an existing user than it does to attract a new one.
What’s the point in attracting new users if they’re only going to stick around for a month?
How to calculate retention rate
Retention rate can be calculated with this formula:
Customer Retention Rate = ( ( Number of customers at the end of a period - New customers created in that period ) / Number of customers at the start of that period ) * 100
Like the active users metric, 30 days is a good guideline. Diving deeper into the data, you can also find correlations between customer profiles and behaviors that promote retention.
For example, a user that invites their team might be more likely to retain than one who just uses the product by themselves.
Net Promoter Score (NPS)
How happy are users?
The NPS metric gives you the answer:
As the image hints, users that participate in this survey are divided into one of three buckets depending on their response. With feedback of 9 or 10, a user is an asset to your company -- they’re doing marketing work for you - likely sending high-converting referrals your way each month.
Users scoring 1-6 are detractors who are unlikely to have much good to say about your product, whereas the neutral scorers (7-8) could go either way.
What do you do with this data?
Well, your overall NPS score is a solid measure of user happiness. The higher the score, the healthier your customer base. You can also market differently to each NPS segment -- for example, launch campaigns to convert neutral fence-sitters into product evangelists!
How to calculate NPS
The first step is to deploy an NPS survey. This can be done over email, during support sessions, or inside the app itself. Tools include Delighted, Promoter.io, and Wootric:
Each data point you get back from a survey is simply a number between 1 and 10.
To calculate NPS, use this formula:
Net Promoter Score = (% of respondents scoring 9 or 10) - (% of respondents scoring 1 to 6)
#6. Stay Agile
The Agile Manifesto and the world of literature that surrounds it is always essential startup reading!
Collecting metrics and crunching numbers is great, but at this scale it’s easy to get bogged down in data that might matter less than just improving your product.
With user feedback and reports on key metrics like retention and activation, it’s time to make product, feature, and marketing iterations that impact your bottom line. This means making decisions based on limited data and voices of a few dedicated customers -- it can feel like a leap of faith, but the old Facebook mantra of Move Fast and Break Things is still a valuable lesson for startups.
Your team and product will grow as it gains traction, but you need internal processes and structure to support that.
For many successful startups, that means:
- Working in sprints. Iterate on the product in 2-week bursts, evaluating performance at the end of each sprint and having open discussions about what went wrong and what went well.
- Taking user feedback logging seriously. User feedback is so vital to startup success, but it’s often mismanaged. Create specific tags in your support ticketing tool that describe the feature or aspect a user is talking about (#feedback-sidebar-ui), or at least diligently upkeep a Google Sheet with the latest snippets of feedback.
- Killing features or campaigns that don’t move the needle. You have such limited resources that you can’t afford to press on in the wrong direction. Does the data say that a beta feature you’re pushing hard isn’t getting much adoption?
#7. Fund and Scale
The mistake a lot of startups make is thinking that step #7 is actually step #1.
Getting funding to scale should come later in the startup journey. First, you need to establish product-market fit, get early adopters, and iterate fast before implementing a large scale go-to market strategy:
“Funding is appropriate for products that have some traction in a large market. For the 99 percent of companies that don't fit this bill, external capital can be a recipe for disaster. If you're at the invention stage, just remember that necessity is the mother of invention, not money.” -- Dave Bailey, VC and serial founder
Scaling too early can hurt a young startup. Boris Wertz of Version One VCs advises that there’s nothing damaging about scaling up things that serve many users for little cost -- like better server infrastructure or self-serve acquisition -- but scaling up on things that burn through money without producing reliable results, like big sales and marketing initiatives, is often a factor in failure.
For these reasons, seek out funding when you’ve already got a validated MVP and a list of active paying customers. This will separate your business from the many others vying for investor capital.
Here are some of the funding options available for early-stage startups:
Alongside accelerators and angel investors, crowdfunding is one of the few funding options available in the early stages before a product generates real revenue.
Services that can help crowdfund your startup include Kickstarter, IndieGogo, and Fundable -- the options available are either pre-selling an MVP, or selling percentages of company stock to investors.
Angel investors are often ex-founders that have a combination of startup experience and tons of cash from their last big exit. They tend to invest in companies that have valuations of around $3m, and will typically invest around $150,000.
Accelerators don’t just offer money. They offer connections, mentoring, and opportunities that would otherwise be out of reach for most startups.
When a startup’s application to an accelerator is accepted, they will usually give up around 7-10% of their company equity in exchange for $25-$125k.
While most accelerators take a sizable chunk of equity, MassChallenge takes a 0% cut and still offers $3m in cash annually to startups as part of its accelerator programs.
MassChallenge is a not-for-profit community of investors with access to angel groups, law firms, marketing agencies, PR firms, venture capital firms, and corporate executives -- everything a startup needs to get set up and go to market.
Founding teams also have access to staff training sessions and workshops on strategy, marketing, fundraising, and team alignment.
Interested? Register here.
Ready. Set. Build.
It’s obvious from the stats that starting a tech company is no walk in the park. With something as complex as software and as unpredictable as team chemistry, there is bound to be a large margin for error.
However, if you follow these steps you’re set with a battle-tested framework that successful startups like Dropbox, Uber, and Buffer have all used to build, validate, and market.
- Define what your MVP will and won’t be. List the features, and think about the lightest way to solve one important user problem.
- Pre-sell your MVP. Leverage connections, crowdfunding, and smoke tests to build buzz and raise cash.
- Offer equity to build a great founding team. Tech startups need a CEO and CTO at the bare minimum. Use talent-sourcing sites to find a co-founder with the right skills, energy, and vision.
- Hustle to reach your first few customers. Without user feedback, software can be built in the wrong direction, wasting resources and ultimately killing the company. The money your first few customers pay is important, but their input on what features to build next, and why, is priceless.
- Analyze your data and pivot (or don’t). Use customer analytics tools to capture the ways users use your app, measure retention, and other success metrics. If something’s broken, find out what it is and fix it.
- Practice agile methodology. As your team and product scales, it gets too complex to manage informally. Agile helps startups build products in quantifiable sprints of work, and in short enough cycles to adjust to customer feedback.
- Get funded, scale your team. If you’ve made it this far, you have a validated idea. To make money and grow, you just need to double down on what works. That’s what funding gives you the freedom to do.
It’s a long road, but when you think about how fraught with pitfalls it is, the fact that 10% of startups survive is a good thing, because with the right action plan you’ll position yourself ahead of the majority of other businesses trying to build the tech company the wrong way.
This post has been updated form the original that was published in Febraury 2019.