Cleaning up with data analysis

Spic ’n‘ span!

What do digitalization and data have to do with vacuum cleaning, mopping up floors and emptying wastebaskets? Jan Bockholdt spent weeks and months pondering that question. Quite a lot, he decided – and set out to prove it, too.

Cleaning up after others runs in the Bockholdt family. His grandfather founded a small commercial cleaning company in 1959 and called it ‘Blitz-Blank’ (‘spick ’n‘ span’). Jan Bockholdt grew up “breathing hoovered air”, he jokes. Today, he employs more than 6,000 cleaners working out of 18 locations all over Northern Germany, making him one of the largest industrial cleaners in Germany.

But cleaning is mainly a manual job: cleaners push their carts full of mops, brooms and cleaning equipment through the corridors, stopping to complete the same set of routines in every cubicle. But some offices are empty because the occupant is sick or on vacation, so why wipe floors that nobody has walked on? Bockholdt went to work with a dedicated team of software developers and created an app that today is installed on tablet computers fitted to every cleaning cart and displaying a detailed list of tasks the cleaner is expected to perform in each room. Customers can access the system and specify exactly what they want the cleaner to do where and when.

But Bockholdt wanted to take to the idea a step further: Clients can also grade the cleaner’s job performance. “This gives us a good idea of how good our own people are”, he says, “but even more important is that we know exactly how happy our customers are at all times.”

Once started, Bockholdt didn’t look back. There must be even more ways to put existing data to work, he thought. It occurred to him that many of his cleaners have long commutes to their workplace, often one or two hours twice a day. Inefficient, he concluded. So his programmers started comparing addresses of cleaners and customers and trying to match skill sets with the needs of their clients. Maybe that way they could find something closer by for their people. But in many cases, it turned out that wasn’t possible: The cleaning profession in Germany is highly regulated, and employees need special qualification in order to be allowed to perform certain jobs. And in some instances, special training is required. Cleaning the blades of wind turbines, which is mandated at certain intervals, requires ‘cleaner climbers’ who can rappel themselves up to dizzying heights like mountaineers. These specialists are in huge demand, and finding them on the job market is tricky.

Instead, Bockholdt asked his computer people to scan the backgrounds of his employees to see if any of them had previous job experience that could qualify them to become climbers. If someone used to work in high-rise construction, for instance, he obviously has a head for heights, so why not offer to retrain him and offer him a job that is closer to home and pays more to boot.

“There are still hundreds of ways we can make better use of our data”, Bockholdt believes, “and if we can clean up by finding them, then so much the better!”


Follow me on Facebook and Twitter or have a look at my YouTube channel

Dieser Beitrag wurde unter IoT abgelegt und mit , , , verschlagwortet. Setze ein Lesezeichen auf den Permalink.

Kommentar verfassen