0 · Opening

Eight processes, taken apart to the end.

For 25 years it's come down to the same question: where is the time leaking, and why did nobody notice before us. The answer changes from business to business. The method doesn't.

The eight examples below always follow the same structure: the situation as we found it, what we identified, what we measured, how we untied the knot, and what we followed up afterwards. Two come from measured work, with the numbers exactly as they came out; the other six describe typical situations, with the same rigour in the method and no invented number to close the sentence.

2 · The eight examples

The eight examples.

01 Distribution

Two days of reconciliation, down to ten minutes

Two people checked by hand, every month, what the bank already knew.

1 · Situation: A client in the irrigation and pools sector, eight stores. Every month, two people checked, one by one, whether every bank statement transaction matched an issued document.
2 · Identified: The process wasn't badly done — it was done by hand because there had never been time to design it any other way. Each store generated its own rhythm of documents, and reconciliation pulled it all together manually at month-end, with no written rule for what to do when a figure didn't match at first pass.
3 · Measured: We timed the process as it actually happened: two full days, for two people, twelve times a year.
4 · Untied: We designed the reconciliation path with explicit rules — what matches automatically, what gets flagged for human review, and why. The same task dropped to ten minutes.
5 · Followed: The time saved is measured and confirmed. We're not publishing an annual euro figure here — not because it isn't there to measure, but because this client had two optimisations in the same period, and we still can't split with certainty how much belongs to each. We'd rather not state a number we can't trace back to its source.

02 Distribution

Three channels, one lost order

Phone, email and the online shop fed the same stock without talking to each other.

1 · Situation: A distributor with its own warehouse. Orders come in through three channels — phone calls straight to the sales reps, email to the office, and a more recent online portal — and all of them end up, sooner or later, reserving from the same physical stock.
2 · Identified: Each channel had its own confirmation rhythm. A phone order was reserved immediately, in the head of whoever took the call; an email order only entered the system once someone read it and logged it; a portal order reserved automatically. There was no single clock deciding who got there first.
3 · Measured: We followed a real day of incoming orders across the three channels, from the moment they arrived to the moment they showed up in available stock — and logged how long each channel took to "appear" to the other two.
4 · Untied: There was no need to switch systems. The portal already had a stock-reservation API that had never been connected to the phone and email entries — it became the single point deciding availability, even for orders that keep coming in by voice.
5 · Followed: We followed a week of orders across the three channels again, to confirm the lag between channels had disappeared, rather than just moved somewhere else.

03 Maintenance

The job sheet nobody reads twice

The technician's paper and the office spreadsheet told different versions of the same day.

1 · Situation: A maintenance company with technicians in the field. Each technician fills in a paper job sheet — hours, materials, incidents — and, at the end of the day, someone in the office copies that sheet, by hand, into a spreadsheet that feeds invoicing.
2 · Identified: Between what's written on paper and what gets typed in, there's always a small gap — a hard-to-read letter, a forgotten material, a rounded hour. Nobody was measuring that gap, because the two versions had never been compared side by side.
3 · Measured: We took one closed month and compared, job by job, the original paper sheet against what had been typed in — counting how many lines disagreed, and on what (hours, materials, or both).
4 · Untied: We replaced the paper sheet with a form on the technician's phone, filled in on site — no paper travelling between the field and the office, no second transcription.
5 · Followed: In the month after the change, we repeated the same comparison — this time between what the technician logged on the phone and what reached invoicing — to confirm the gap had actually disappeared, not just changed shape.

04 Services

The appointment nobody cancelled

No-shows only surfaced at the end of the day, too late to refill the slot.

1 · Situation: A clinic with phone bookings and a system diary. When a client misses an appointment without notice, that slot sits empty — but it's only noticed once someone, at the end of the day, reviews the diary by hand.
2 · Identified: There was no automatic reminder or advance confirmation — the diary depended entirely on the client remembering by themselves. And when they missed it, nobody was alerted in time to call the waiting list.
3 · Measured: Over one month, we counted how many bookings ended in a no-show, and how much time passed between the booked slot and the moment someone at the desk noticed it was free.
4 · Untied: We switched on an automatic SMS reminder 24 hours ahead — a feature that already came with the booking system, but had never been activated — and built a waiting list that gets notified as soon as a slot is confirmed free.
5 · Followed: We counted no-shows again the month after the change, and the time between the booked slot and the slot being reused, to see whether the reminder was changing behaviour or just delaying the problem.

05 Construction

The quote built from scratch, every time

Every quote was built from scratch — until the client asked for his own customers to be able to build one themselves.

1 · Situation: A construction company that quotes job by job. Every request meant starting over: looking up material prices, adding labour, checking an old similar quote so as not to drift too far, and writing it all out by hand. Prices lived scattered across spreadsheets, supplier catalogues, and the memory of someone who has been quoting for twenty years. A quote took as long as it took — and while it did not come out, the customer waited.
2 · Identified: The work was not deciding the price. It was going to find it. The part that took experience — judging the job, anticipating what goes wrong, knowing what is not in the drawing — took a fraction of the time. The rest was searching, copying and adding up.And because every quote was a fresh document, nothing learned in one was available in the next.
3 · Measured: We timed the building of a real quote, from request to delivery, separating time spent looking up prices from time spent deciding. And we counted how many requests were sitting in the queue in a normal week.
4 · Untied: We built two pieces. First a configurator, where material and labour prices live in one place, editable by the people who know them — with nobody to depend on to change them.Then a quote builder that feeds off it: you pick what the job needs and the quantities, and the quote comes out, landing within roughly 95% of the final figure. The missing 5%is what takes an eye: the difficult access, the surprise inside the building, the condition you only see on site. It is still human work — but only that.\n\nA month later the client came back with a request that was not in the plan: if the tool gave him an approximate quote in minutes, why not let his own customers do the same? We opened a public version of the builder. People looking for work done can now configure what they want and get a preliminary quote on the spot, without waiting for anyone.
5 · Followed: Quoting time dropped to a fraction of what it was. But the effect nobody had predicted was a different one: with quotes arriving in minutes instead of days, they started winning more work. A quote that takes three days competes with two rivals; one that arrives in minutes arrives first. The optimisation started as internal time saved and ended up bringing in sales.

06 Services

The same data, written twice

One entry in the management system, another in the invoicing one — and nobody checking they matched.

1 · Situation: A professional services office. Every client and every service delivered exists in two systems: one for project management, one for invoicing — with no automatic link between the two.
2 · Identified: Whoever enters the data knows they're duplicating work, but nobody had measured what that duplication cost in mistakes — only what it cost in time.
3 · Measured: We picked one closed month and compared, client by client, what was entered in one system against what was entered in the other — counting how many records didn't match, and in which field they diverged.
4 · Untied: We didn't replace either system — both had a reason to exist. We connected an automatic export from the management system to the invoicing one, so the same data is only written once and travels on its own into the second system.
5 · Followed: We repeated the comparison a month after the automatic link went live, to confirm the mismatches had dropped to zero, not just moved somewhere else in the process.

07 Industry

The expensive ERP, and the spreadsheet in charge

The system was bought and paid for; the side spreadsheet was what actually decided what went to the factory floor.

1 · Situation: A small manufacturer with an ERP installed for several years. The production and stock modules are active and paid for — but alongside them lives a "just to be sure" Excel sheet that is, in practice, the document that decides what goes into production next.
2 · Identified: Nobody had formally decided which of the two sources was the truth. Each person writes first wherever they trust most — and the ERP, which should have been the single source, was reduced to a historical record that gets updated later, when there's time.
3 · Measured: Over one week of production, we compared the orders sitting in the Excel sheet against those logged in the ERP — counting how many existed on only one side, and how many had different quantities between the two.
4 · Untied: There was no need to buy anything new — the ERP already had a production-planning module that had never been configured. We configured it with the rules that, until then, had only existed in the head of whoever kept the spreadsheet, and the spreadsheet stopped being needed to decide what went to the factory floor.
5 · Followed: We compared planned production against actual production again a month later, to confirm the configured ERP could hold up on its own the decisions that only the spreadsheet used to make.

08 Manufacturing

The morning lost to copying

Phone orders were written out three times before they became an invoice.

1 · Situation: A food producer with its own factory and customers who order by phone. Each order was noted down in an Excel file at the office. That file was shared with the factory, which filled it in as things were produced and shipped, and sent it back. Then, at the office, someone picked it up and entered the orders one by one into the ERP, for records and invoicing. It took up an employee's entire morning. Every day.
2 · Identified: The same order was written three times: in the office Excel, in the factory Excel, and again in the ERP. None of those added new information — they just carried the same data to the next place. And between each step there was waiting: the factory did not know what had come in until the file reached them, and the office did not know what had gone out until the file came back. The time was not being spent deciding anything. It was being spent copying.
3 · Measured: We timed the daily ERP entry with the person doing it, and counted how many times the same piece of data — customer, item, quantity — was rewritten between the phone call and the invoice.
4 · Untied: We built an orders module inside the system the client already used for the online shop, where phone orders are now recorded directly. For the factory, a panel of its own: the most frequent actions as buttons, so people working on their feet with their hands full do not have to fill in forms. The office now sees the state of every order as it happens, with no file to wait for. That left the ERP, and two roads: integrate through an API, or export a file it could import in one go. The integration was more elegant and more expensive, and it depended on the vendor's schedule. The client chose the export — same problem solved, a fraction of the cost, and nobody left waiting on a third party.
5 · Followed: Four hours a day that stopped being spent copying. The employee who spent every morning entering orders got the morning back. And the information stopped travelling in files between the office and the factory — it now lives in one place, updated by the people who produce it.

3 · Final CTA

We haven't seen yours yet.

Each of these eight examples started the same way: someone describing how the work happens today, without yet knowing where the leak was. That's the starting point of any new conversation.

Thirty minutes, online, no strings attached. Tell us how the process works now; we'll tell you if there's something there worth mapping, and how we'd do it.

Book thirty minutes to talk

online, no strings attached