Free templates for engineering intake and delivery planning
Four spreadsheet templates for the part of delivery that happens before any ticket exists: taking a request in, sizing it against what the teams can actually give, recording the decisions, and keeping the risks honest. Each is a CSV with a few fictional example rows — open it in Excel or Google Sheets, delete the examples, keep the columns. Each links to the essay that explains the method behind it.
Project request form
The intake form with two required questions and one optional date — a need, and who is affected or what happens if nothing changes. Each row also records who asked and when, and the triage decision that has to follow within the week: plan, discovery, park or decline, with a reason the requester can read.
Columns
- Requested by
- Date received
- What needs to change?
- Who is affected — or what happens if nothing changes?
- Is there a date this is tied to?
- Date context (one line)
- Decision (plan / discovery / park / decline)
- Reason
- Review date
Capacity plan in FTE-months
One quarter, per team and per craft — never an organisational total. A team-line row takes gross capacity down to the honest line; demand rows then walk down it in priority order, each with the figure the delivering team gave and its confidence, until the line is spent.
Columns
- Quarter
- Team
- Craft
- Row type (team line / demand)
- Walk order
- Item
- Gross capacity
- Committed elsewhere
- Interruption reserve
- Honest line
- Demand figure (FTE-months)
- Confidence (high / medium / low)
- Figure given by (delivering team)
- Remaining on the line
- Next step
Decision log
For the decisions that would cost more than an hour to re-make. It records what was decided, what was rejected and why, who decided, and the condition that would reopen it — the field that turns a decision from something defended into something checkable.
Columns
- Date
- Demand it belongs to
- Decision question
- Options considered
- Chosen (what was decided)
- Why, and why the other options lost
- Decided by (a person)
- Revisit if
RAID log
Risks, assumptions, issues and dependencies, attached to a demand on every row rather than to a project document. "Assessed — none known" is a valid status, distinct from a blank nobody looked at; dependencies carry "asked, not confirmed" until the other team agrees; issues name the decision they need.
Columns
- Demand
- Type (Risk / Assumption / Issue / Dependency)
- Item
- Owner
- Status
- Severity (risks only)
- Next review date
When a spreadsheet stops being enough
These templates work in a spreadsheet, and for one person making the commitments on a list that fits on a screen, a spreadsheet is the right tool. What a sheet cannot do is refuse: any cell can be edited, so a row reading "committed" means someone typed it, and the reason a risky commitment was made early is overwritten by the next edit.
DeliverySheet is the same method with the rules enforced:
- A capture floor. A request enters the lifecycle only with a stated need plus an impact or a trigger — the two questions of the request form, checked at the door.
- Enforced states. A demand moves through capture, shape, decide, plan, commit, deliver and learn; parking and declining are recorded decisions with a reason, not rows that stop being updated.
- A readiness verdict. Dependencies, decisions, capacity and risks are read before commitment — each can honestly be "none known", which is different from "not assessed" — and a blocked dependency, an unresolved high risk or an open key decision keeps the demand from reading as ready.
- Recorded exceptions. Committing before a demand is ready requires an exception naming why it is early, what is missing, who accepts the risk, and what condition would change the decision.
Where the sheet genuinely wins, and where it quietly fails, is its own honest page. To check a real request against the need, the impact and why now — no account, nothing stored — paste it into the free request check.
$189/month per workspace, unlimited members. 7-day free trial — card required, cancel before it ends and you're not charged. I answer the support email myself.