Big data and bad decisions

Big data and bad decisions

Now before you go setting your hair on fire, data is good, data is great, I’m a fan. But here’s the thing, we’ve always known that numbers can tell you anything, and that’s back when we were talking about small numbers. So what sort of whopper lies do you think Big Data can tell?

History is littered with individuals who looked at the numbers and went on to make moronic decisions. Napoleon springs to mind. Alexander the Great too. In more recent times the execs at 20th Century Fox looked at the numbers (how much money movies made out of promotional products) and thought it was a steal to give all the rights to George Lucas for a $20,000 reduction in his fee. I think sales of Star Wars promo products passed the billion dollar mark about a decade ago.

Now to be fair, the boffins at 20th Century Fox were looking at historical data, and history was about to change. Same was true of Decca Records rejecting the Beatles in 1962. But isn’t that always happening? Isn’t history always changing? In fact every article you read will tell you that the rate of change is excellerating exponentially, not slowing down. A hockey stick is the angle most likened to the speed of change. So anyone who thinks they can look at data (and ignore history and our blunders throughout it) are kidding themselves.

Hang on, I forgot something. We're so much smarter now aren't we... yes, of course I forgot to factor that in. We don't have to worry about history, just read the data and make your call in your little vacuum chamber - what could possibly go wrong?? 

Actually maybe we're not that smart. That’s why the stock market crashes in nice even cycles, why we believe our politicians and, to the eternal misery of our race, continue to bash each other senseless (and that’s the kindest description I could come up with) over religion, oil, land, water, sex, and well pretty much anything else you can think of. A war even started between Greece and Bulgaria in 1925 over a stray dog (I'm not kidding, look it up). So you get where I’m coming from - we’re still pretty stupid.

Anyway, back to my original point - what I’m trying to get at is that small data, big data, it doesn’t matter. The devil is in the details. How you read the data, along with the landscape of the time, is all that matters. Numbers aren’t going to give you a clear advantage - certainly not when the internet makes the same information available to pretty much everyone. Numbers are an important part of making good decisions, but following numbers blindly makes you blind. And I don’t know about you, but when my sight is limited I tend to bump into things and it hurts like hell. The business equivalent of that is bankruptcy (at least if you do it often enough).

So what to do? You know you need to take notice of the data, but likewise you don’t want to end up passing on the Beatles. My advice would be thus:

1 - Surround yourself with good people. This is my first rule of business, but

2 - Good people don’t always give you good answers. In the same way that good data doesn't always mean good news, good people might mean well, but that doesn’t make them right.

3 - Take in all the available data, but filter it through your prism of what’s important. Others don’t know what you know. Don’t discount your innate ability to smell a rat (or conversely, an opportunity - read Malcolm Gladwell’s excellent book ‘Blink’ for a more thorough and entertaining dissertation on why you should trust your gut).

4 - Your gut might have a use-by date. Everything around you is shifting. So stick to your knitting where possible, but that’s getting harder and harder. You need more external people nowadays than ever before to help you deal with the complexities of business, so like it or not, sometimes you will be wrong simply because the world is changing and you can’t be good at everything (even James Bond can’t order a martini properly, oh and he treats women woefully, and he has anger management issues, and he keeps smashing up Astin Martins which are my favourite car - come to think of it I don’t know why I like this bloke at all). That digression aside, my only advice here is go back to #1 and surround yourself with good people, use your gut to work out who’s on your side and who’s just along for the coin, and then (and this one is critical)...

5 - When someone new wants you to change direction, tack small.

I know there will be people out there who will remind me of my penchant for saying “Bite off more than you can chew and chew like crazy” but not when you’re taking the advice of someone you don’t trust yet. And so often the Big Data consultants are just that - people you don’t know yet (or don’t know well). So float a test balloon and see what happens. Take baby steps if it’s in a direction you haven’t gone before. Make the new member of your team prove their thesis - don’t just bet the whole business on it. These types of bets tend to be 50/50 at best and that’s a tough beat if you lose.

6 - And finally, don’t shy away from Big Data, but at the same time don’t ignore small data.

Big Data can lead to paralysis through over analysis, and in most cases, the job that follows straight after getting the Big Data is the process of culling it into bite-sized small data chunks to inform your decision-making. So working out what the small, elite group of numbers are that give you the most important insights is critical.

An industry I see this a lot is SEO. Big Data is used a lot to justify the continuation of campaigns that should be euthanised. Or to chart a direction that’s exactly where you shouldn’t go. Or to paste over problems, like a website that won’t ever convert leads (or even products nobody wants). And once again, unless you’re an SEO expert, you can’t make these decisions on your own (at least not without burning through bundles of cash). So at the risk of being melodramatic, you wouldn’t represent yourself in court, you wouldn’t do your own surgery, and you wouldn’t sail solo around the world when you’d never been on a boat. UNLESS YOU HAD A DEATH WISH. So take care and get expert help.

SEO specialists may not be surgeons, but they do work in a ridiculously complex and ever changing environment that frankly. YOU CAN'T DO IT YOURSELF AND YOU'LL NEVER BE ABLE TO (at least not without devoting your life to studying it). And let’s be honest - you wouldn’t want to anyway, anymore than you’d like to swap jobs with your accountant or lawyer. If you wanted to be those things, you would have done it years ago (that said I still hold out hope of drumming for Van Halen one day, but some people are just plain delusional).

I know SEO is opaque and some providers make hay out of that. But that doesn’t mean you bury your head in the sand. Keep kissing frogs until you find a prince (thankfully IT frogs aren't any more disgusting than standard ones). Then when you find a good one, hang onto them. Someone who tells you the truth, even if it’s ugly, and takes the blame when they screw up is gold. They shouldn’t be so rare, but they are. Once you find them that will be an important piece of the marketing puzzle, allowing you to release the reins on a little, knowing that the data you’re getting is genuine, the advice you’re getting about that data is worthwhile, and your trusted advisor is just that, someone you can trust.

Back to other posts