19 Jul

The internet, magazines, and white papers are ablaze with how the internal organisational conflicts big data and BI platforms cause can be rectified with models, change agents and\or data scientists, resulting in a decision making utopia. Unfortunately this advice can be misguided at best and downright harmful to a business at worst. The problem can be simply defined and is best explained by a quick look at the natural world.

Everyone will have read or heard stories and reports of how the absence or addition of a single species or plant no matter how small or seemingly insignificant can completely transform ecology. Any addition of a new or returning species into an eco system needs to scientifically and legally demonstrate that it must first do no harm, which is where the burden of proof lies. Take Scotland for example which is currently reintroducing species such as the beaver and lynx. The reintroduction takes an experimental form in natural conditions as a very small number of the animal in a restricted area are gradually added to the ecology. Each change and adaption is carefully monitored, each benefit and risk to the ecology logged. If any animal does escape it’s in small numbers and this can actually create another natural experiment- how this new wildlife adapts and affects another area of the ecology. This is decoupled risk and experimentation in a complex environment, and only when all the risks and benefits are known, anticipated and mitigated for will the reintroduction widen.

An organisation is an ecology in itself and also part of a huge global economic ecology with billions of interdependencies. But people don’t treat organisations like an ecology they treat them like bland linear text book models. If you add something to an organisation you should be aware of the harm it could do- what good practices could it wipe out? Which key staff could be compromised? Will the upside outweigh the downside, and of course, do we know the downside? Most organisations have no real idea why they are successful; success is more to do with tacit knowledge and heuristics than it is to do with complex models and procedures. But these tacit strengths get ignored and then wiped out by the next craze, and this is what’s happening with big data and analytics.

Use big data and analytics but introduce it to your ecology with care. Focus on a small team and let them use analytics intuitively and tacitly, experiment and tinker. Keep tabs on the heuristics and tacit abilities they develop and the results and insights they produce. Ask questions such as- what data cues are the team spotting and what outcomes do they produce? Is there a consistency and pattern in this relationship? Not only will this experimental approach mitigate widespread ecological damage it will also drive insight and innovation. Gradually widen the application across the organisation and the guidance your able to pass on to new users will be based on natural use and intuition, and I’ll be clear here- technology fails dramatically when it doesn’t dovetail with how people make decisions naturally.

There are simple, easy to adopt ways of maximising the use of data and BI platforms, but organisations seem to be seeing the adoption of big data strategies as an end in itself and wreaking ecological havoc in the process. There are huge benefits to the new technology available and the key to unlocking these benefits is an experimental approach.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: