The short version
- Handoffs fail for four reasons: they describe activity instead of state, they omit the unknowns, they name no one, and they live somewhere nobody looks.
- Four fields fix all four: current state, open questions, blockers, next actions. Each open question and blocker carries a name.
- The most commonly missing field is open questions, and it is the most valuable, because it tells the next zone where the work is fragile.
- A handoff that cannot be questioned overnight will still lose you a day. The document has to be answerable, not just complete.
A handoff succeeds when the next person can continue the work without contacting the previous one. Judged that way, most handoffs fail, and they fail in four recognisable patterns that have nothing to do with effort. People write long handoffs that fail and short handoffs that work, because what matters is which questions the handoff answers.
"The handoff was a Slack message nobody read."
The four ways handoffs fail
- It reports activity. "Worked on the migration, fixed two bugs, reviewed Anna's PR." True, and useless. The reader cannot tell where the work now stands.
- It omits the unknowns. Handoffs are written to sound competent, and unknowns feel like admissions. So the reader inherits the uncertainty without being warned about it.
- It names nobody. "Waiting on infra" gives the reader no action. "Waiting on Marek, ticket INF-2210" gives them one.
- It lives in a scroll. Posted into a busy channel at 18:30, buried by morning. The receiving zone does not find it, so they ask, which is the cost the handoff existed to remove.
The four fields, and what each one prevents
| Field | Answers | Prevents |
|---|---|---|
| Current state | Where does this stand? | Re-deriving what was already established |
| Open questions | What do I not know? | Inheriting uncertainty unknowingly |
| Blockers | What is stuck, and on whom? | Silent waiting, chased days later |
| Next actions | What happens next, and who does it? | Two people starting, or neither |
That is a brief. Four fields, ninety seconds, written at the point you stop. The full template and a worked example are in the end-of-day handoff.
Activity is not state
This is the most common and most fixable error. Activity describes what you did. State describes where the work is now. They feel similar to write and they are completely different to read.
"Spent the afternoon on the retry logic" is activity. "Retry logic is written and passing locally, not deployed, and the 409 case is unresolved" is state. The second tells the next person exactly what they are inheriting and exactly where it is incomplete.
A useful discipline: write the current state as if the reader has to decide whether to deploy it tonight. That framing forces out the qualifiers that matter and removes the narrative that does not.
Why every line needs a name
Open questions and blockers without names produce a specific and expensive failure: the next zone reads the handoff, understands the situation, and has no idea whether it is their job to do something about it. So they assume someone else has it, and the item sits for a day, which is exactly the outcome the handoff was written to prevent.
A name converts information into an action. "Blocked on the security review, Anna, raised Tuesday" tells the reader who to chase, that it has been chased, and how stale it is. Three pieces of information in nine words. If you cannot name anyone, that is itself the finding, and it belongs in the handoff as a line saying so.
Complete is not the same as answerable
Here is the limit that no template solves. A handoff answers the questions the author thought of. The next zone reliably has one they did not, and at that moment the team is back where it started: a capable person, blocked, with the only source of the answer asleep.
What removes the last gap is making the handoff itself answerable. With StandIn, the brief you confirm in ninety seconds is what your StandIn speaks from while you are off. It answers questions in your words, with a source under every answer, and it never guesses: if the answer is not in your brief, it says so and names who to ask. So the next zone can interrogate the handoff at 03:00 your time and either get the answer or find out immediately that this one genuinely has to wait for you.
That is the difference between a document that hopes it anticipated everything and a handoff that holds up when it did not. See how StandIn works.
Common Questions
How long should a handoff be?
Ninety seconds to write, and usually four short paragraphs to read. Length is a poor proxy for quality: a two-line handoff that names the open question beats a page of narrative that omits it.
Should handoffs be per person or per piece of work?
Per person is easier to sustain and per piece of work is easier to read. Start per person, since habits beat structure, and add a work-specific handoff for anything genuinely being continued by someone else overnight.
What if the next zone does not read it?
Usually it is in the wrong place or it arrives after they have started their day. Put it where they already begin, and post it at the end of your day rather than the start of theirs. If they still do not read it, ask one of them what they look at first thing, and move it there.
Do handoffs replace the standup?
They replace most of what a standup carries across a seam, which is state and blockers. What a standup adds beyond that is the live conversation that follows, so keep a short one where it earns its slot. See replacing the daily standup.
When you're off, your StandIn is on.
It answers your teammates' questions from work you've already done, in your words, with a source under every answer.