5 Tech Tips for Startup Founders

A collection of technology pitfalls I’ve seen non-technical startup founders make, and how to avoid them.

Don’t wait

Let’s kick things off with an easy one. If you want to start a company where users all sign up to use Service X but the technology for Service X isn’t going to be ready for 1 month, what do you do in the meantime?


Desert by Moyan_Brenn, on Flickr

Answer: get the users anyway. Getting users is hard for pretty much every startup in its early days, so don’t delay doing it. Keep a spreadsheet of email addresses, write down phone numbers of everyone you meet who expresses an interest in what you’re doing, post about what you’re doing in forums and get contact details from everyone who replies.

When your technology is ready to go, you have hundreds of accounts you can create from day one and avoid that standing-start empty-desert feel new sites sometimes have.

Your startup’s technology will never be finished

Just as a journey of a thousand miles begins with a single step and Rome wasn’t build in day, so your startup’s technology will never be finished. The danger here for non-technical co-founders is to view the technology side of your startup as a hurdle to be overcome, and postpone other activities until the technology is “finished”. The reality is that your technology is only ever “finished for now” but will need improving, updating and maintaining as you get more users or your users get more involved or your idea grows. Technology grows as your company does; don’t let the technology become an excuse for delaying (see above).

Your technology doesn’t need to start perfect

Remember that your technology can always be changed. It will never be finished, so don’t assume it will start off being perfect. In six months, your concept might have changed or you might have 100x as many users as now – so your technology just needs to be good enough, not perfect. It’s a waste of time and money to make something that works for millions of users when you only have 1 thousand today. I see a lot of startups worrying about whether they’ll be able to serve 1 million users before they’ve even got 1 thousand. Save your time, save your money: get the technology good enough for now and focus on getting to 1 thousand users before you think about getting to 1 million.

Beware traps

Some startup ideas are simply traps. By traps, I mean ideas that sound attractive or easy but are actually incredibly difficult to execute well.

Traps can be things that lots of people are starting right now. An example? Dating sites: there seems to be a new one every week because a lot of people think they see why existing dating sites are failing and they think they know how to fix it. Even if you build the perfect dating site (and, let’s face it, you probably haven’t), you now have the challenge of recruiting thousands of users away from all the other sites in what is already a saturated market. I suggest asking yourself where your idea came from: were you reading about another startup or talking to someone who uses an existing service? Or did you meet someone who has a problem which there just isn’t a solution for at the moment? You’re much more likely to succeed if you have access to a group of people whose problem isn’t being solved than if you’re trying to improve on flaws in existing solutions.

Dating sites are also a great example of another trap I call symbiotic markets. Symbiosis refers to living things that have evolved together and are dependent on each other for survival. Check out the Wikipedia page for some cool examples. Symbiotic markets refers to a business idea where you need two distinct groups of users to make a success of the site. E.g. eBay needs both buyers and sellers, dating sites for heterosexuals need both men and women. If eBay had no buyers, the sellers wouldn’t visit. And if the sellers don’t visit then the buyers won’t list. The danger of symbiotic markets is that you get yourself into a chicken-and-egg conundrum where it seems impossible to get one group of users without the other, and in the end neither comes. It takes double the effort to get enough users for your site to pass the tipping point in a service with symbiotic markets. If I was going to build a dating business, I’d focus on gay and  lesbian people.

Trap number three: platforms. A platform is a service for other people to build on to make their own services. There are famous lucrative examples, like the Facebook’s platform for building Facebook-based apps, or even the Google and iPhone platforms. But platforms are really hard to get right. You can’t do a platform properly until you can anticipate every single facility that will be wanted by someone who builds on it. The famous platforms everyone has heard of started out by building a whole raft of apps for themselves before letting other companies try out the platform, to avoid this problem. Make several applications first THEN extract the platform from that and you stand a much better chance than if you try to build the platform first.

Hire a CTO or consultant you trust


Trust by Joi, on Flickr

An obvious point but one that bears repeating. If you hire a CTO, you’re likely to be working closely with them for a long time. If you hire either a CTO or a consultant, they have the potential to either ruin your business or make it incredible.

Years ago, I joined my first startup as CTO. We were looking at changing the way clothes are recommended and bought in the fashion industry, and we had a sound business model. But it was a terrible idea for me to be their CTO; I’m not at all enthused about fashion and was very open with them about this. I should never have asked to join, and they should never have wanted to have me. I also made clear that I wanted to have nothing to do with running the business, I just wanted to be left in isolation to make their tech. This should have been another warning sign: be wary of separating the technology from any other part of the business. It’s vital for planning the business to know which parts of the technology are hard/expensive/slow, and which are easy/cheap/fast. Similarly, it’s vital for the technology side to be made with the future direction of the business in mind. The technology can always be changed but early decisions can make later changes easier or harder.

I left that startup after only a few months and take a much stronger role in the business side of startups I work with now. Last I heard, they’d secured investment and were in a London startup accelerator programme; hopefully they went on to find a CTO who was a better fit with what they were doing.

A good CTO / consultant will care about what you’re doing, and will tell you how the technology affects your business, help you plan and think of new directions and will help you make the best decisions. They’ll take half the work off your hands and help you get things done.

A bad tech-person will just do whatever you tell them, even when it’s not the right decision for the business. This will seem easier in the short term but will harm you in the long term when you’ve spent all your cash on technology you didn’t need and bad technology decisions come back to bite you.