Spend capture
Company cards fire an event the moment they are used. The freelancer photographs the receipt in the app. A model reads it, then files it against the right production.
Every production lived in its own spreadsheet, so nobody knew the real position until the shoot wrapped. Costs now land categorised, and a hot budget line surfaces the same week.
Hugo takes the project apart on screen: the bottleneck we found, the system we shipped, the number it moved.
No walkthrough yet. The written case below carries the same numbers.
Vela Studios makes commercials and branded content out of Madrid, with a second office in Barcelona. Twelve people on staff, up to sixty freelancers in a busy month, eight to twelve productions running at once.
Each production got a spreadsheet, copied from the last one and edited by whoever had it open. Card receipts arrived as photographs in a WhatsApp group. Some never arrived at all.
The real numbers were built at the end, from bank statements. By then the overspend had already happened. On two productions in 2025 it reached eleven percent of budget.
Month close took three weeks of one person, which meant Vela was always closing the month before last.
Two days of interviews and one week of their own data. Here is what was actually going wrong, told the way the people living with it told us.
Every production started as a copy of the last one. The sheet then drifted, edited by whoever had it open that afternoon, until two shoots described the same cost in two different words.
That left the producer flying blind through the part of the job where the money actually moves. Real figures arrived weeks after the wrap, from a bank statement, long after the decision they should have shaped.
Freelancers felt it at the other end. Their invoices waited behind an approval nobody owned, and two of the best ones quietly stopped picking up.
Built inside the tools the company already paid for. Every workflow, credential and prompt was handed over at the end.
Company cards fire an event the moment they are used. The freelancer photographs the receipt in the app. A model reads it, then files it against the right production.
Every approved budget becomes a live sheet that reloads each night. A line crossing 80 percent sends the producer a message naming the line, the shoot day and the amount.
Freelance invoices are matched to the purchase order, queued for one approval and posted to Holded. The close pack builds itself on the second working day.
Before figures come from the company's own records, pulled during the audit. The after column is the ninety day average once every system was live.
| Metric | Before | After |
|---|---|---|
| Month close | 3 weeks | 3 days |
| Chasing receipts, weekly | 14 h | 20 min |
| Budget position | 5 weeks late | Same day |
| Productions over budget | 4 of 11 | 1 of 13 |
| Freelance invoice to payment | 34 days | 12 days |
| Producer admin hours, weekly | 26 h | 4 h |
I used to find out we had overspent when the money was already gone. Now I get a message on the Tuesday and there is still time to move something.
Every project has them. Writing them down is how the next one gets shorter.
We built the reporting before the category list was agreed, so the first month of data had to be recoded by hand.
Two hours in a room with the bookkeeper would have prevented all of it.
We also assumed freelancers would adopt the card app on their own. They did not, until a producer sat down with each of them for ten minutes.
First system in production inside a month. Everything after that is iteration with the client in the room.




Enquiries fully handled: 68%
Read the caseFirst response, was 4 h: 11 min
Read the caseThe AI and Operations Audit finds it, prices it, then hands you a plan in two working days. Your fee comes off the first thing you build with us.