Data gravity is weighing down your business

Recently, I was struck by a question from a client when working through an enterprise data strategy. As I explained how their data should flow through the martech stack from sources and platforms, they stopped me: “Yeah, but where are the centres of gravity?”

This was an interesting way of phrasing it, but it was a question that made a lot of sense the more I thought about it. Unsurprisingly, the idea of data ‘gravity’ already existed, and it becomes more relevant every year as the volume of data within businesses continues to grow.

Illustrator: Hugh Osbourne

‘Data gravity’ was first introduced as a metaphor on engineer Dave McRory’s blog, Data Gravitas. It describes the phenomenon where, as the volume of data grows within an organisation, platforms and systems become attracted, even enmeshed, in the data.

It’s an interesting term, and it’s only becoming more relevant as the volume of data being produced doubles every two years. 

Functionally, in an organisation, this means wherever your data generative sources are, your platforms and technology will cluster. Platforms and systems have their own gravity as well, but this is generally borrowed from the quantity of data the system holds. 

Think of a customer relationship management system. Whether it’s Salesforce, Hubspot, or something else, the extensions, additional software, add-ons, and integrations you build out from it are probably a consequence of that system being where you update and store your customer relationship data. It also becomes really difficult to make the switch to newer, cheaper, and more efficient pieces of software should they become available, because everything is so interconnected.

This is a relatively simple example, but for something like a supply chain, requiring complex, integrated systems that gather data at hundreds of touch points – from factory to customer – data gravity becomes a much more pronounced problem. 

If your business has an unhealthy reliance on massive spreadsheets full of historical data, you’ve encountered another data gravity problem – everyone knows there’s probably a better way to do it, but nobody wants to put in the time or effort to move to a better system. And why does no one seem to switch from iOS to Android? Because moving everything to a new operating system is a pain in the neck.

Illustrator: Hugh Osbourne

Being led by technology isn’t unusual for many businesses, especially as they grow, but if you don’t define these centres of gravity for your data, your technology will end up defining it for you. It’s a feedback loop, too: the more a piece of technology is used, the more data produced by that piece of technology, the harder it is to change, and so an organisation keeps using the tech.

This presents several problems when you wish to use data from multiple sources and start working with larger datasets. If you really want to unlock the value of AI and machine learning tech on your own data, you’ll also need to have all the data somewhere central.

If you’ve worked in marketing, you’ve probably seen a version of the Martech 5000 landscape by Scott Brinkler. It’s an absolute mess. There are now over 7,000 providers of marketing technology across dozens of categories.  

There are now over 7,000 providers of marketing technology. Source: chiefmartec.com (Click to enlarge)

But if you did an audit of all the platforms and services currently being used by your brand (in 2017, the average was 91), you might just find something similar. Instead of a solar system revolving around a central source, such as a data warehouse, you end up with a complex, tangled web of technologies and data flows, with rich, valuable data stored in silos you might not even own, such as media data owned by partners or suppliers. It is data that effectively holds legacy systems and software hostage, preventing a company from evolving.

I’ve had plenty of frustrating conversations when searching for the data I’ve needed to do my job. “Who’s got the login?”, “Who’s the point of contact?”, “Can you pull this for me?”, are all common refrains in the workplaces I’ve been in. What if there was just one central source, a centre of gravity, where all relevant data could be collected, stored and used across many applications? 

These centres of gravity don’t necessarily look the same for all companies; one-size-fits-all models rarely fit well. Rather, your organisational data strategy needs to examine where your data is generated and where it ‘naturally’ gravitates towards. 

By taking the time to define an organisational centre of data gravity, your data is unlocked as a strategic asset. It’s also why the big enterprise battleground between Amazon Web Services, Microsoft, and Google is now in cloud storage and computing. In order to be most useful and efficient, enterprise data needs a platform-agnostic environment in which to live, where it can be combined with other data to produce better insights.

Practically, this also takes shape in a ‘big bet’ backing one platform provider. Collecting, moving and using data becomes more efficient when it doesn’t need to be transformed between platforms. Even if you’re already making a big ecosystem bet, your technology mightn’t be used to its full potential; in the ‘crawl-walk-run’ model of implementing new platforms, lots of organisations never make it past a crawl.

Illustrator: Hugh Osbourne

Woolworths’ latest Salesforce transformation is a great example of both big bets and unlocking platform potential. The supermarket giant had seven parallel instances of Salesforce running, with customer data being generated from all, often being duplicated, and not being used effectively. While the system ‘worked’, it was inefficient, costly, and couldn’t unlock the true value of the data by synthesising it across the organisation.   

Despite being made of gas, Jupiter weighs about 300 times that of Earth, and gravity is about 2.5 times as strong as we experience it on Earth. If you could find somewhere to stand (which you couldn’t, as it’s made of gas) it would be much harder and require far more energy to move your body around than on Earth. It’s much the same with data gravity; the more data mass you accumulate around a specific technology, the harder it is to move anywhere meaningful, both in terms of technical effort and organisational willpower.

By taking the time to define where your data will be housed and operated on, and placing big bets around one technology ecosystem, you can ensure your organisation doesn’t fall into a data gravity trap. Data shouldn’t be kept in departmental silos, and platforms or technologies should not dictate what you can and can’t do with your valuable data. And remember –  the longer you leave it, the harder it’s going to be to free yourself from the gravitational pull.

Regan Kerr is a strategist at Mutiny