A worked example
One email. One record. One lesson, three months later.
Vantia is invented — 45 people, B2B software, invented for this page. The record below is invented with it. What is real is the shape: these are the fields the product records, and the PDF at the bottom is rendered by the same builder that exports a live workspace’s decision.
For decisions that actually happened, with the ones that went badly kept in, read our own ledger.
The workspace
What this team told Decize it cares about.
Values and goals are set once, at setup. They are why a decision can be read against something other than whether it sounded reasonable in the moment.
Values
- Sustainable pace
- Deliver what we promise
- Grow deliberately
- Say the trade-off out loud
Four, in the team’s own words. Decize scores each decision against them and never rewrites them into something tidier.
Goals this quarter
- Deliver every onboarding inside 30 days
- Reach 60 active accounts by year end
A decision is linked to at most one goal, and only when the message makes it obvious. Nothing is filed under a goal to make a chart look fuller.
What somebody sent
The whole input is this email.
Not written for Decize. Sent to the team list on a Monday, with the capture address copied in — which is the only thing anybody did differently.
From Priya Raman
To leads@vantia
Subject onboarding backlog
We are three weeks behind on onboarding and two of the PMs are cooked. Options were (a) hire a third PM now, (b) bring in a contract PM from the agency, or (c) cap new client onboarding at four a month until the end of Q3.
Going with (c). Hiring takes ten weeks we do not have and the agency PM costs more than the revenue we would defer. Pretty confident this is right, though it means telling sales to slow down. Cap starts Monday, I own it, we review it end of Q3.
The same day
What the record held by Monday evening.
Drafted from the message, then confirmed by the person who sent it. Confidence is read from the words “pretty confident” — when a message signals nothing, that field stays empty rather than being guessed.
Cap new client onboarding at four a month instead of hiring a third PM
Onboarding is three weeks behind and two PMs are at capacity. Hiring takes ten weeks we do not have and a contract PM costs more than the revenue the cap defers, so intake is throttled until the end of Q3.
The question
Do we add delivery capacity now, or protect the team by limiting intake for a quarter?
What success looks like
Every onboarding started after 15 June completes inside 30 days, with no PM logging more than 45 hours in a week.
Options considered
- Hire a third PM now - rejected: ten weeks to productive, and the pain is in the next fortnight.
- Contract PM from the agency - rejected: EUR 14k/month against roughly EUR 20k/month of deferred revenue, and no retained knowledge.
- Cap intake at four a month until end of Q3 - chosen.
Assumptions, each with a date to check it
- Pipeline holds at six or more qualified deals a month, so the cap defers revenue rather than losing it. Confidence 3/4. Check at the end of July.
- Two fewer onboardings a month brings PM load under 45 hours a week. Confidence 3/4. Check in the week of 6 July.
- Sales can hold a deal for four to six weeks without it going cold. Confidence 2/4. Check on the first held deal.
Trade-off accepted
Roughly EUR 40k of Q3 revenue is deferred, not lost, and sales has to hold two qualified deals. Accepted because the alternative spends the same money and still leaves the team at capacity through August.
What would make us revisit
Reverse the cap early if the onboarding queue clears before 15 September, or if more than three qualified deals stall on it.
Value alignment
Sustainable pace
+0.9
Deliver what we promise
+0.6
Grow deliberately
+0.1
Say the trade-off out loud
-0.5
Serves Sustainable pace, strains Say the trade-off out loud. An estimate until a person confirms it — and worth remembering when you read the lesson below.
One quarter later
The part a document folder cannot do.
On 30 September the owner gets one email and records how it went. The forecast resolves, and the confidence written in June is finally worth something: it can be compared to what happened.
Outcome
Mixed
Delivery recovered — 11 of 12 onboardings closed inside 30 days. The revenue assumption did not hold: about EUR 25k was lost rather than deferred.
Calibration
Overconfident
Confidence 3 of 4 against a mixed outcome. The verdict stays on the record, and the same comparison runs across every decision as a chart nobody has to keep by hand.
The lesson on record
When we throttle intake, write the revenue being deferred in the same sentence as the capacity being protected - otherwise the trade-off gets re-litigated in month two.
Stored in the team’s own words, and attached to this decision the next time a similar one is weighed.
What preceded it
The three closest earlier decisions in this workspace, surfaced automatically with how each turned out — including the hiring call that ended in regret.
- Pause the partner integration to clear the migration backlogsuccess2025-11-04
- Hire two PMs at once ahead of the Q1 pipelineregret2025-08-19
- Use the agency for overflow design work through Q4mixed2025-09-30
Nothing on this page was typed into a form. The message was sent to a team list, the fields were drafted from it and confirmed, and the outcome was recorded from one email three months later. That is the whole of the work.
Your turn
Start the record with one decision.
Take the same document with you, or make the first entry in your own ledger. The free plan carries 10 decisions a month, for as many people as you like.
The PDF is this same invented decision, in both states, exported the way a real workspace exports one.