save time, money, and sanity

How to save money on your next project

9 January, 2018: 800 words, 3-minute read

Companies can burn through obscene quantities of cash building IT systems. Consider any government-led project: how many are successfully delivered on-time and on-budget?

Fortunately, there are ways to minimise your initial outlay and get the best-value solution in the shortest possible time. But it will involve some effort on your part…

Plan your project

Who knows more about your business? Is it you or a random developer you contacted on Twitter?

if you fail to plan, you are planning to fail Benjamin Franklin (probably)

Projects often fail because no individual in the client organisation fully understood the problem, the target audience, the solution, or the complexities before coding started. The idea is vague so the resulting application is a poorly-conceived, unusable mess of random functionality.

The solution: write a high-level requirements document before you consult developers. Ensure it identifies core essential functionality and the nice-to-haves. It's a pain and no one will want to write it, but it will save you considerable time and money. Nothing is set in stone and some ideas may be impractical, but it'll give you a solid starting point for discussions and estimates.

…but stop thinking about technologies!

Unless you're a developer with up-to-date knowledge, don't try to describe the technologies your project requires. That's what you are employing a developer to do.

we want an app! everyone since 2007

The best system architects will recommend cost-effective technical implementation options based on your requirements. Suggesting they use DatabaseX or LibraryY is a distraction at best and a costly mistake at worst.

"How much?" is a question for you!

Would you enter a car showroom and ask to buy a vehicle without revealing how much you want to spend? You could, but while you may be considering a Ford, the sales person is talking about a Ferrari. Revealing your budget saves considerable time for everyone.

The same is true for IT projects. A five-page website could cost £500 or £50,000; both do much the same thing, but the more expensive version could be slicker, is faster, has more functionality, has bespoke imagery, works on more devices, can be installed, works offline, is heavily promoted, has after-sales service etc. Your budget will absolutely determine what can be achieved.

the time it takes for someone to ask "how much" is inversely proportional to the amount of hassle they cause Craig Buckler's law

If you want the best-value fixed-price solution, it's up to you to state that price. A developer can then suggest what you will get for your money based on your specification. There will still be caveats because it's an estimate. Every project is a step into the unknown because your system has never been developed before.

Of course, no one does this. Every day, developers are bombarded with "how much?" requests from people with vague or non-existent specifications. The answer will always be "it depends?". If pushed, I will think of a obscenely high number, double it, add 500% contingency then multiply by Pi because it's impossible to quantify. You certainly won't get the best price.

Forget revenue share options

I certainly would make millions if I had a pound for every approach like this!

I have a fantastic idea: you develop it and we'll share the millions! doomed start-ups

Perhaps you do have an amazing idea but, if that's the case, why aren't you investing your own money or seeking investment? Why offer a percentage of your business to a random developer who may not be capable and certainly cannot be committed?

You're actually asking a developer to invest their time: a far more valuable commodity than money. While they're working for you for free, they cannot charge for other projects or work on their own amazing applications. It's a double hit.

Only consider a revenue share if you know the developer well and they become a director of your company. Even then, they should receive a salary and the bulk of all profits until their commitment is repaid.

Perfection is futile

No system is perfect. Applications can always be better, faster, easier, more attractive etc. Even if you reached perfection today, tomorrow's technologies can improve your system further … or make it obsolete.

Let's presume your application could be 80% completed within one month. The core functionality was implemented and the application was usable. Would you wait another year for the last 20% of features? That's a year without users, feedback or revenue? This is precisely what many organisations do.

The solution:

  1. Identify essential functionality. Develop a Minimum Viable Product (MVP) and launch as early as possible.
  2. Assess user praise and criticisms then evolve the application as required.
  3. Profit!

If you do nothing else…

Stop reading this, clarify your idea, write that requirements document, then contact us to discuss it further!

Best of luck.