May 19, 2011

...Learn TDD with Codemanship

Check Under The Bonnet Before You Buy A Used Tech Start-Up

Buying used products is a risky business.

Takes used cars, for example. If the previous owner didn't take proper care over the vehicle's operation and maintanance, you could be handing your hard-earned money over for what is effectively a death trap.

Which is why we have laws and entire industries dedicated to protecting use from the worst risks. Anyone selling a car - new or used - has to demonstrate that it's safely roadworthy, and without that all-important piece of paper you cannot get it taxed for the road, buy insurance for it or day any of the other things potential buyers will require.

Even used car salesmen, who have rightly earned a reputation for questionable ethics, have lines they cannot cross without risking huge fines or even prison.

People selling software-reliant businesses have no such constraints placed upon them. While selling cars with MOTs is widely frowned-upon, it seems tech start-ups are actively encouraged to cut corners. The received wisdom is that your code only needs to last as long as you need to complete your exit strategy, and if you make your code any better than that, you're a fool.

That's like encouraging car owners to only do the bare minimum of maintenance to ensure the pwerson who buys it off you can drive it out of the showroom, and maybe get as far as their house, while you laugh all the way to the bank.

Ethically, this puts us somwhere below used car salesmen. And estate agents. I don't know about you, but I feel just a teensy bit uncomfortable with that.

Put the shoe on the other foot. You are in the market to buy a fledgling tech start-up. Undoubtedly there'll be lawyers and accountants involved, and there'll be a process of due diligence. The ambulance chasers and bean counters will go through your business with a fine-toothed comb, looking for potential gotchas or negatives you may have tried to hide from them.

But, it seems, the one place they never look is in your code. Which is odd, because your code is the engine of your tech start-up. If your business was a car, and someone bought it without checking under the bonnet, and didn't even bother to ask for a current MOT certificate, you'd probably be justified in thinking them a fool.

I've only very rarely come across examples of technical due diligence in mergers and acquisitions. I've witnessed some truly massive deals done with absolutely zero attention paid to the software itself. Basically, they've sat in a room and told each other "our software does this, and our software does that", and the deal's been done entirely on that basis. I've seen companies hand over hundreds of millions for businesses that, it later transpired, didn't even have a working build of the software to look at.

And it's this kind of lunacy and gross negligence that leaves the door wide open for the rest of us to say "You know what? Let's not bother" After all, who's counting?

So, yes, in an industry where nobody bothers to check, it is indeed a waste of time and money making your software good enough to last beyond the ppint where it's not your problem any more. But such an industry is truly flawed.

When we're looking at it from the buyer's side, and we're tasked with adapting or integrating software after a merger or an acquisition, we know just what a nightmare of epic and monstrous proportions that can be.

If potential buyers actually did some basic technical due diligence to uncover the reality of what's under the bonnet, and if it became common practice to require that software is "roadworthy" and has enough life left in it to accomodate the buyer's long-term plans, before anyone will buy a business built on it, then I suspect much of our industry would necessarily be transformed.

Of course, nobody who buys tech start-ups is likely to read this blog, so to some extent I'm pissing into the wind on this one. But perhaps if a bunch of us who feel this way got together, pooled our resources and got organised, we could educate a few buyers about the risks they're taking.

Posted 10 years, 6 months ago on May 19, 2011