Before TeamTrek
- Friday went on the timesheet. Six hours across the team, every week, spent remembering which client Tuesday afternoon belonged to.
- Unlabelled hours reached invoices. Roughly one hour in five arrived with no project on it, and the account lead guessed or wrote it off.
- Retainers overran quietly. Nobody could see a retainer was 80% spent until the month closed and the overage had already happened.
The agency was not short of data. Hours existed in a spreadsheet, work existed in a project tool, and the two agreed with each other once a week, briefly, on a Friday afternoon. What they did not have was one record that was true as the work happened, which is a different thing from a record that is assembled afterwards by people doing their best.
What they changed
One decision carried the rollout: a tracked hour has to arrive with its client and its project already on it, or it does not count as tracked. Everything else followed from that. The team start a timer against a task rather than against a day, the clients and projects were set up once at the start, and the Friday review stopped being a data-entry session.
The rule that made it work
No untagged timers. A timer started without a project prompts for one before it runs, so an hour cannot become unlabelled later without someone having chosen that.
Putting it to work
Two weeks, and most of it was agreement rather than software. The order mattered: the structure went in before the people did, so nobody spent their first week typing hours into a shape that did not exist yet.
- Day 1 Clients and projects in. Twenty-two active projects across eleven clients, each with a rate and a retainer ceiling where it had one.
- Day 2 The team told, before the software. What gets recorded, what does not, and that every person can open their own record in the same detail their lead sees.
- Week 1 Timers on, reports off. A week of tracking with nobody reviewing anything, so the habit formed before the numbers were used for anything.
- Week 2 The Friday review, on the record. The first review read the week rather than rebuilt it. It took 50 minutes.
What it changed
The review is the headline, but the invoice is the one the owner noticed. Work that used to sit for a week and a half waiting for someone to work out whose it was now goes out in four days, because by Friday there is nothing left to work out.
Weekly timesheet review −88%
Unlabelled hours on invoices −87%
Work done to invoice sent −64%
The argument stopped being about whether people were working and started being about what the work was taking. That is a different conversation, and a much shorter one.
Where they go from here
Two things are open. Retainer ceilings are watched by a person rather than by an alert, which works until someone is on holiday. And the estimate on a new project is still set from instinct, although there are now nine months of finished projects to set it from instead.
Try it on your own clients
Fourteen days, no card. Set up your clients and projects and see what a Friday looks like.