The Four Tenets of Bootstrapping a Tech Startup

Plunging yourself into the world of entrepreneurship and developing an idea into a thriving, profitable business is one of the most exciting and rewarding adventures anyone can embark on. But walking the road to startup success can also be one of the most stressful, nerve-wracking, and exhausting experiences too. One that, more often than not, ends in defeat and disappointment.

This stress can be even greater if you’re Bootstrapping, which a lot of entrepreneurs decide to do. Bootstrappinggrowing a business without external funding — with the right mindset can make the difference between your startup joining the 25% that succeed, or joining the 75% that don’t.

The 4 tenets of bootstrapping successfully

Your startup is unique and comes with its own opportunities and challenges. I can’t offer you a one-size-fits-all template for success. That said, the key aims when launching a tech startup are generally the same for all: Rapid delivery of a product that effectively meets the needs, of an eagerly awaiting audience, at relatively low cost to the startup.

In this article, I’ll describe three tenets that, if adhered to, will help you achieve just that. I’ll explain how to test your idea before you build it, generate some hype before you launch it, and deliver a product that your customers actually want to use—without blowing your budget.

Tenet 1: Test your idea before you build it

Surprisingly, this is one of the main reasons startups fail. A recent study of failed startups by CB Insights showed that 42% of those startups failed because there was no real desire for their product.

Forty-two percent! Mindblowing!

Type caption for image (optional)

I’m willing to bet that few of this 42% did any real testing of their idea before they started building it. As a consultant, I see this regularly when I speak to new clients — the desire to start building a completely untested idea seems to be endemic among entrepreneurs.

Before you go pouring your life savings into an idea, it’s prudent to first spend some time and money testing your concept with a real audience to see if anyone is interested in what you have to offer.

Yip, I know, this doesn’t sound as sexy as the usual “go with your gut”, “shoot from the hip” entrepreneur rhetoric. But this approach is based on reality, and will reduce the risk of you really wasting a ton of real cash. Until you have some real data, your decision to launch a business is based on your assumptions. You actually know very little. It’s your first job, as the bold and intrepid entrepreneur, to close the gaps between what you assume is true, and what you know is true about your product in your marketplace.

There are lots of good ways to test your assumptions and validate your idea. The best way really depends on the nature of your business. In general, you want to find a way to present the gist of your product to real, potential customers, and gauge their initial thoughts and reactions.

Set up a launch page

What is it you’re actually offering people? Do they want it? Even before you start developing your product, you can start to answer these key questions.

One quick and easy way to test your idea on a portion of your audience is to set up a launch page. Since you don’t have a functioning product yet, you won’t be showing screen-grabs of it in action, but a catchy headline that explains what you’re offering, and a gist of its key benefits is a good place to start.

Ensure your launch page has a prominent and clear call-to-action. This could be a “Leave your email address” form, or a “Follow us on Facebook” link. Both have their pros and cons.

Email subscription gives you a direct means of contacting nearly 100% of your signups, while Facebook offers great analytics on the age, gender, and location of everyone who’s liked your page.

Choosing the right option for your product is a judgement call, but I tend to prefer email because it gives a more reliable means of reconnecting with your audience.

If you opt for an email signup form, don’t go overboard by asking your visitors for their name, gender, shoe size, and GPS coordinates. People are already reluctant to provide their contact information, and adding more fields to a form is directly correlated with a reduction in the number of people who fill it in.

If you push for likes on a Facebook page, try to offer your followers something worthwhile by posting relevant and interesting content to your Facebook page a few times per week.

Conserve the attention of your audience

Avoid the temptation to add multiple calls-to-action for Twitter, Facebook, email, to your landing page. After someone has read your pitch, it should be undoubtedly clear to the reader what they should do next. Adding more calls-to-action will just confuse them and they’ll be less likely to take any action at all — other than leaving your site, that is.

Use conversions as your benchmark

Apart from the obvious benefit of being able to re-engage with potential customers further down the line, calling your visitors to take action also acts as a great benchmark. By tracking who converts (who signs up and who doesn’t), you can start to get a clear idea of the number, and the type of people who show an interest in your product.

Would your business plan change if only 0.01% of under 30s from New York like the sound of your recipe sharing service, but 12% of over-30s from Yorkshire love it and want to know more?

If you’re inundated with people eager to hear more, that’s a great indication that you’re moving in the right area. If nobody shows interest then it means either you’re not pitching it to the right people, your sales pitch isn’t great, or your idea isn’t great.Address those concerns in that order.

For a quick and easy landing page, check out a service like UnBounce or LaunchRock. To track which of your links are getting the most clicks, share a different link each time using Bitly.

Test multiple sales pitches, simultaneously

The traditional approach to building a launch page is to have just one landing page — the homepage. You post links to your website all over the internet with an attractive headline and hope that your homepage is effective enough to persuade them to sign up.

The problem with this approach is that everybody who you reach out sees the same thing; they hear the exact same message. There’s no way for you to vary your pitch and test what works better, and what doesn’t. It’s a one-way conversation.

Web real-estate is really, really cheap. Instead of creating just one landing page and hoping for the best, create multiple landing pages; each one with a different headline or offer, that highlights the benefits of your product in a different way, or to a different type of person.

Imagine for a second: You’ve set up two separate landing pages, each describing your service, but with a different pricing structure. After a month or promoting each landing page, your signup data shows that only 2.5% of visitors were interested in paying a fixed rate for your service, but 9% wanted to hear more about a freemium version of your service with micro-purchases to unlock additional benefits? Would your plans to build a fixed–rate service change at all?

Using Google Analytics and Google URL tagging, you can track not only which product offerings were most effective, but also which audience was most interested, and which traffic source was the best for bringing visitors. Did you get more sign-ups from Mumsnet compared with Facebook or Twitter? Was direct email to a specific group more effective? Was there a difference in the conversion rates of headlines between the various traffic sources?

Running multiple landing pages simultaneously, gives you the chance to start learning about your audience and hone your sales pitch before your product has even been developed; with real data that shows you what works, and what doesn’t. This sort of preliminary market research specific to your product is invaluable and should give you a much clearer idea of what your product is, and who you’re developing it for.

Ensure you have enough validation

How much validation do you need before you’ll be confident enough to proceed? 100 email addresses? 1000 likes? This number is really dependent on your particular idea, your business model, the market, how you’re testing it, and how confident you want to be in your idea. For landing page conversions, a minimum acceptable conversion rate is about 4% — and no, that can’t be 1 sign up from the 25 of your mum’s friends you invited. Aim to pitch to at least 2000 people per landing page. This will reduce the likelihood that one of your pitches seems more effective than the others by pure chance.

Key takeaways:

• Before you start developing your product, take steps to validate your idea in the context of a real audience
• Learn as much as you can about your audience, those who converted and who didn’t. Not only will this reduce the risk of you developing a product nobody wants, it will give you extra practice developing your idea, as well as your pitch
• Test your sales pitches with a significant sample of your audience to avoid false positives appearing as a result of having too little data

Tenet 2: Connect with your audience before launch

Now that you have a means of keeping in touch with your audience, why not engage with them during development and keep them excited about your product’s launch? After all, you’re aiming to launch to an eagerly awaiting audience, right?

There’s a subset of your target market known as the “early adopters”. They are the people who love to try out a new idea, and they get excited at the prospect of having or being involved in something new.

Early adopters typically make up about 15% of a given market, but since your signups have signed up to a product that doesn’t exist yet, you can assume that most, if not all of them have the early adopter mindset.

These people are your new best friends.

Keep a fun and open dialogue with them. Reward them with things like an exclusive sneak peek behind the scenes; post regular updates of your team working on the product; ask them to help choose the final logo from a short-list; have them help name your features or mascots; offer invite-only access to your product and ask them for feedback. Actually listen to their feedback.

By engaging with your audience during development, your customer becomes a part of your design process. It’s a lot harder to develop a product nobody wants if you base its design on real user feedback.

(Incidentally, the movie industry has been doing this for decades).

Key takeaways:

• Get to know the early adopters who have shown an interest in your product and keep them as excited about the launch as you are. Appeal to their inner narcissist by rewarding them with exclusive access which they can brag about to their friends.
• Involve them in the design process by asking for their input as you develop the product. This will not only make them more excited to see the finished article, but increase the likelihood that they’ll stay loyal to your service, and recruit their friends.

Tenet 3: Build an MVP, quickly

Even having a well-validated idea, and engaging with your customers through development, you don’t really know how effective your product is, or how much you’ll need to change it until you launch it. So do that as soon as you can.

Here at Katana, we always encourage clients to focus on launching an MVPfirst, and we guide them through the process. Your MVP — Minimum Viable Product — is the bare minimum you need to launch, in order for this product to be a success. Pay special attention to the second letter, V. I’ve seen some startups become too focused on the minimum part, and not enough on the viable part. Your MVP still has to sufficiently meet the needs of your customers, it just doesn’t have to fill their bath-water and tweet about it too.

Here’s a great example: Would Facebook still be popular without the “Poke” feature? Probably. Would it still be popular if you couldn’t upload photos? I’d guess no.

Start with one platform

Choose one platform to launch your product on first. Sure, your end–goal might be “complete world domination”, but you can add the rest later, once you have the budget to do so and your product has matured a little. If your product is primarily a web-based service, build a web app (an interactive website) and leave the native mobile apps until later. Web development is typically faster than mobile app development, so you can test your idea sooner, and cheaper. Also, websites can be made responsive, which means they can offer a great experience across most mobile devices too.

If your product fits better as a mobile app, build for iOS first and leave other platforms until later. Android development typically costs around twice as much and takes twice as long as iOS app development. When you couple that with the fact that Android users are less likely to pay to download apps and less likely to make in-app purchases, then iOS first becomes a no-brainer. Here’s a great post about a startup that opted for Android first — add it to your reading list.

What goes into your MVP?

Unfortunately, there are no hard and fast rules for what should go into your MVP. It’s a judgement call, and it really depends on your product, your audience, and your competition. Getting an MVP right takes time and a bit of experience. Here’s a great exercise that I’ll do with clients, to help them clearly define their MVP:

Grab a piece of paper and draw a line down the middle, creating two vertical columns. In the left column, note down all of the features you’d like your product to include. Be sure to list everything, not just the obvious, main features. Seemingly trivial features like a “Share on Facebook button” and an “Invite friends link” will still add to the overall development time, so it’s important that you list them too.

The column on the right will become a list of all of the things you should include in your MVP. Starting from an empty slate, add the most obvious and essential feature and then ask yourself: “Could we launch it with just this?”. If the answer is no, choose the next most obvious feature and add that to the list, then ask yourself again. Rinse, and repeat.

Once you reach a point where the list on the right truly covers the bare essentials your app will need, you have the features of your MVP. The other features in the left-hand column are those you might add further down the line. A lot of them will never be added.

Stay focussed on your plan

As an entrepreneur, waiting patiently on the completion of your product can be frustrating. While development is underway, you’ll still be actively engaging with your audience and coming up with new ideas on how your product can be improved even more. Developing a product for a constantly evolving business model is a tricky line to toe. I’ve seen a few startups fail simply because they keep updating the product again and again with new improvements. They ran out of money before they were able to launch anything. When that happens, the feeling of regret is crushing.

If you do come up with something that you think is essential to your product that you missed before, pass it through the same filtering process as the other features. Add it to the left-hand column of your features list, and look at the right-hand column. Ask yourself again “Could we launch it with _just_this?”. If the answer is still yes, then don’t add the new idea yet — keep it in the left-hand column and review it later. If the answer is no, then add it.

A final note on your MVP: Don’t worry about your product being perfect — it will never be perfect. Focus on creating something good enough, that your customers will love. Perfectionism kills startups.

Key takeaways:

• Aim to launch an MVP; focus on building only the features you need. The nice-to-haves can come later—when the time and budget allow
• Don’t meander too much during the development process
• Launch

Tenet 4: Keeping the launch costs low

Launching a new tech product can be expensive, but it’s not the only expense you’re going to incur. You’ll also need to market the product, pay salaries, and pay for future feature development and updates too — so don’t spend your entire budget on the initial product development.

Keeping the overall cost of launching your app to a minimum depends on a few factors, most of which are outside the scope of this article. But I can provide some tips on keeping your software development costs to a minimum.

Choose the best tools

Almost every new web startup you’ve heard of in the past few years, including Twitter, and Groupon, was built using a web framework called Ruby on Rails — there’s a reason for that: Rapid development.

There are countless frameworks you can use to build websites in these days. Some are old and outdated, Rails is constantly being updated; some are obscure, Rails is extremely popular; some are slow and clunky to work with, Rails is fast and painless.

I’ve covered the benefits of Ruby on Rails in another post. The short version is that Rails is build using the Ruby programming language — a really elegant and concise language that boosts developer productivity. Rails also comes with a set of smart configuration defaults that remove a lot of the tedious set-up decision making that comes with other frameworks. Rails is also scalable, and trusted by some of the largest companies in the world.

If you can’t find a Ruby on Rails developer, use Django as the next best option.

If you’re building a mobile app, RubyMotion is a great tool. Like Rails, RubyMotion lets developers build using the Ruby programming language, a much more elegant and flexible language than the alternatives for mobile development.

Since autumn 2014 though, RubyMotion is also the first tool to allow developers to build native applications for iOS and Android — which will save you a considerable amount of time when you decide to launch the Android version of your app.

Being a very young product, RubyMotion doesn’t have the same legacy as Ruby on Rails yet, but smart companies including Basecamp and Jimdo have already jumped on board and are taking advantage of the competitive edge RubyMotion offers.

If you can’t find a RubyMotion developer, Apple’s Swift is the next best option.

Offload as much as you can to 3rd party services

Don’t re-invent the wheel, unless it’s cheaper and quicker to do so. Your developers can save a lot of time by using 3rd party services and code libraries to handle some of the usual tasks within your application. Emails, SMS, card payments can and should be offloaded to 3rd parties who specialise in doing those things really well.

A lot of services, like Mandrill, also offer a freemium option to small companies with a low volume of transactions. They’re free when you start out and will scale along with you, only charging more when your usage supports it. Sometimes, though, it can take longer to integrate a 3rd party service than it would to build it from scratch or use an open source alternative.

Again, this is something you should discuss with your developer.

Update: As of July 15th, Mandrill no longer offers a freemium option :(

Cut corners, sensibly

Clients often ask me if there are any corners we can cut to launch sooner. And the answer is “yes”. There are a ton of best practices that developers should do, but can sometimes mean things take longer to build. Automated testing is a perfect example.

There are few things more frustrating than launching a new update to an application, only to find out a day later that this change has completely broken another part of the system. Murphy’s law says it’s usually the sign-up page or the check-out process. Tests prevent scenarios like that. Automated tests are short lines of code that automatically check the features of your application are working as they’re supposed to. Having a test suite in place is great!

The downside to testing is that it takes extra time to write and maintain each test.

Another example is accessibility. Accessibility, in a software context, means making sure that your website or app can be used easily by people with certain disabilities or special requirements. Offering the best experience possible for all of your users is an important issue, and it’s sometimes a legal one.

However, as with automated testing, making an application more accessible requires extra time and money. It’s not necessarily something that you need to do right away, or before launch. These days, 3rd party accessibility tools are becoming far more effective, removing the urgent need for web developers to fill in the accessibility gaps.

Be prepared to repay your technical debt

So yes, you can save yourself time and launch sooner by skipping these and other best-practices. But this comes at a price. Un-tested, rushed code has to be checked manually each time you launch a new update. The time required to make changes to your application inflates over time and the occurrence of bugs will start to increase, as the codebase grows in complexity.

Eventually, you’ll have to fill in the corners you cut, and often the overall price you pay may be considerably higher. There’s a reason why we call corner-cutting “technical debt”.

But, like financial debt, a little technical debt can a sensible trade-off if it means you can start making money sooner.

Key takeaways:

• There are short-cuts you can take to expedite the development process, these shortcuts will need to be addressed at some point
• In the long run, cutting corners is rarely the cheaper option
• If your developer tells you they don’t write tests, find a new developer.

Putting it all together

By employing the principles I’ve described, and getting creative with your own variations, it’s entirely possible to validate your idea with real customers; to learn about your marketplace and improve on your idea with real feedback; and to grow an eager and loyal fanbase — before you’ve even started to build anything. By making the right decisions on the technology you use, and cutting the right corners, you can launch a great product on time, and in-budget.

Rapid delivery of a product that effectively meets the needs, of an eagerly awaiting audience, at relatively low cost to the startup.

That sounds like an ideal launch to me!