In this blog, we’ll explore what happens when you try to move a process to a new location before anyone has actually written the process down, why “we know they’re working, we just don’t know at what” is a genuine sentence a head of procurement once said to me, and how documenting the invisible turned two brand-new nearshore sites into a control tower.
T-Systems asked me in to help move part of their procurement operation out to new sites in Hungary and Slovakia. Straightforward enough, on paper: hire people locally, hand them the existing process, let them get on with it. Except there wasn’t an existing process. There was an existing set of habits, distributed across several buildings in Darmstadt and beyond, and nobody had ever written any of it down.
A company built on three previous companies
T-Systems had been Deutsche Telekom before it was T-Systems, and Deutsche Post before that — and some of the paperwork still in active use hadn’t caught up. I sat with procurement staff working from forms that still carried the Deutsche Post logo at the top. Nobody had reordered the stock in years. Nobody had thought to. It was a small, slightly absurd detail, and it told me everything about the state of the process underneath it: inherited, patched, never properly rebuilt.
The actual work was serious. Procurement had to check every new supplier against T-Systems’ requirements: declarations that they didn’t use forced labour, evidence of sustainable sourcing, sight of their incorporation documents, and a fair amount besides. Necessary checks, all of them, and the requirements varied by category: mobile phones went through one set of hoops, office furniture through another, paper stock through a third.
The trouble was where the checking happened. A colleague in one Darmstadt building would clear one part of an authorisation. It would travel to someone in Munich for the next part. Then back to a different person in Darmstadt for the part after that. The head of procurement, nominally responsible for the whole operation, told me plainly that he had people working for him and no real idea what half of them did day to day. Not because they weren’t working. Because there was no way to see the work.
Getting Darmstadt to admit what it actually did
The interviews were the hard part, and not for the reason people expect. It wasn’t that staff were hiding anything. It was that most of them genuinely believed they already had a process — their process, the one they’d been doing for years — and struggled to see why anyone needed it written down.
Ask someone to describe a routine they’ve performed from memory for a decade and you get a confident, fluent answer that skips half the actual steps, because the missing half is so automatic to them it doesn’t register as a step at all. Getting the real sequence out of someone meant asking the same question three different ways and comparing notes with the person one desk over doing the “same” job slightly differently.
You cannot nearshore a process you haven’t written down
That’s the sentence I kept coming back to, because it’s the entire point of this story. Everyone assumed the plan was: move the work to Hungary and Slovakia, then figure out the control afterwards. I flipped it. You can’t control a nearshore team on a process that only exists as tribal memory scattered across three cities — you’ll simply have exported the confusion somewhere cheaper, which is not the same thing as fixing it.
So the first job was archaeology, not automation. I went through what actually happened, step by step, building by building, person by person, until the invisible process was visible on paper. Only once that existed did it make sense to put it into a service desk tool the new Hungarian and Slovakian teams could actually run against — with a combined dashboard so the head of procurement could, for the first time, look at one screen and see how many authorisations were moving, how fast, and where they were stuck.
What fifty new employees actually got handed
By the end, around fifty people across the two nearshore sites — roughly twenty-five per site — were trained on what proper service management meant for this operation, working from a documented, tool-supported process rather than institutional memory. The transfer from Germany was complete. And the head of procurement finally had statistics instead of guesswork: turnaround time by process type, volume by month, a genuine basis for continual improvement instead of a shrug.
None of this required new technology. The service desk tool itself was nothing exotic. What changed the outcome was the order of operations: document first, tool second, offshore third. Reverse that order — as plenty of nearshoring programmes do, under pressure to show progress fast — and you get a shiny dashboard reporting on a process nobody can actually vouch for, which is arguably worse than no dashboard at all, because it looks like control without being control.
It’s worth being honest about the sequence, because I think it’s the transferable lesson here, not the specific procurement detail. Every nearshoring or offshoring project I’ve seen go wrong has made the same mistake: assuming the receiving team can absorb a process that the sending team never actually documented in the first place. If your own head office can’t tell you what your process is, sending it three hundred miles away doesn’t create clarity. It just moves the confusion somewhere with a different postcode — and adds a language and time-zone gap on top, for good measure.
The Deutsche Post forms are long gone now, I’d imagine. But that detail has stuck with me longer than most of the technical work, because it’s such a precise little symbol of what undocumented process actually looks like from the inside: not chaos, exactly, just habit stacked on habit, inherited from an organisation that stopped existing years before anyone questioned why the paperwork still said its name.
To find out more about PDCA Consulting’s expert consulting services either:
- Book 30-minute conversation directly: here
- Get in touch via the contact form here
- Connect with me on LinkedIn
