Remote teams did not invent the daily standup. They inherited a meeting that assumed a shared room, a shared morning, and a shared fifteen minutes. When the team stretches across timezones, that assumption breaks. Someone is always joining too early or too late. Someone is always half-present. The meeting still happens, but the reason for the meeting — a shared picture of progress and impediments — often does not.
Async standup tries to keep the signal and drop the calendar tax. People write what they finished, what they are working on, and what is blocked. The channel fills with updates. On paper this looks healthy. In practice the thread becomes a wall of text that nobody reads carefully, especially the lines that matter most: the blockers.
This post explains a practical Done / Doing / Blocked roll-up for small remote teams, why blockers must be forced to the top, how that format relates to the Scrum Guide without pretending to be Scrum orthodoxy, and where a writing aid like Standup AI helps versus where it cannot help.
Older versions of Scrum popularized three questions: what did you do yesterday, what will you do today, and what is in your way. Many teams still use a Done / Doing / Blocked variant. The point was never bureaucracy. The point was inspection: can the team see progress toward the goal, and can impediments be named early enough that someone can help?
The 2020 Scrum Guide removed the prescription of those exact questions. Developers may choose any structure that inspects progress toward the Sprint Goal and produces an actionable plan for the next day of work, within a fifteen-minute Daily Scrum when the event is live. That is an important honesty check for tool vendors. A Done / Doing / Blocked template is one useful format among several. It is not sacred, and it is not required by the current Guide.
Walking the board, reviewing aging work, or asking a single focus question about the biggest obstacle can work better for some teams. If your team already has a strong visual board and talks item by item, a pasted status dump may be redundant. If your team is small, remote, and drowning in chat, a structured roll-up often restores the property live standups used to provide: blockers get airtime.
In a live standup, social pressure and facilitation push impediments into spoken time. In async chat, people bury the hard line at the end of a paragraph: shipped the docs PR, reviewing billing copy, also waiting on staging secrets for three days. Readers skim the first clause. The blocker never becomes a team problem; it stays a private sentence.
Timezones make this worse. By the time the last timezone posts, the first timezone has already started deep work. Nobody schedules a five-minute reaction window. The thread becomes a museum of status rather than a decision surface. Teaching Agile and Scrum Alliance materials both warn that unread async updates are not a functioning Daily Scrum — they are a checkbox.
A roll-up does not invent missing discipline, but it changes the default presentation. When blockers are sorted to the top of one shared summary, the cost of ignoring them rises. Leads and teammates see the risk list before the done list. That single ordering change recovers a lot of what the live meeting accidentally enforced.
Keep the input cheap. Ask people to write short lines, not essays. Prefer attributable notes when names matter: Alex finished API auth; working on rate limits; blocked on staging keys. If your culture already uses tickets, referencing a ticket id is fine. If it does not, plain language is fine.
Collect the paste once per day from wherever people already write: a Slack channel, a Notion page, a linear comment dump, or email. Do not invent a second system of record if you can avoid it. Collection is a workflow problem; roll-up is a reading problem.
Then produce one summary with three sections, blockers first:
Close with an optional next-step prompt for humans: which blocker needs an owner today? The tool should not invent owners. People should.
Pick a daily cut-off that matches how your team actually works, not a theoretical UTC midnight. Publish the roll-up at a predictable time. Ask at least one person to spend five minutes reacting: claim a blocker, schedule a huddle, or say no action needed. Without that reaction step, async standup fails quietly.
Protect deep work. Async is not an excuse to demand instant replies from every timezone. The roll-up creates a shared artifact; follow-ups can be opt-in huddles for the people named in blockers.
Keep the artifact short. If the summary is longer than a careful person will read before coffee, you will recreate the wall-of-text problem inside a prettier shell.
Standup AI takes pasted updates and merges them into one Done / Doing / Blocked summary with blockers first. It can format for Markdown, Slack-friendly text, or email paste. It is a writing and roll-up aid.
It does not poll Slack on a fixed schedule. It does not force people to post. It does not replace Jira, Linear, or any sprint board. It does not guarantee that your team will stop holding live meetings, or that participation will improve. Those outcomes depend on habits you own.
Compared with classic standup bots, the pain we target is reading and merging, not collecting. If your pain is that nobody writes updates at all, start with facilitation and expectations — a cleaner summary of empty input is still empty.
This is not a claim that async is always better than live. Co-located teams that enjoy a short ritual should keep it. This is not a claim that Done / Doing / Blocked is the only valid Daily Scrum format under the Scrum Guide. This is not project-management software. This is not legal, HR, or management advice.
If your standup is mainly a political status ritual for management, cleaner text will not fix the politics. Fix the purpose of the meeting first.
Day 1: agree on a paste location and a cut-off time. Day 2–4: generate a roll-up and force a five-minute review of blockers only. Day 5: measure two things — how many blockers were named, and whether any of them got an owner the same day. Keep the experiment if both numbers move in a useful direction. Drop it if the team only produces unread archives.
Honesty beats theatre. A short shared risk list that people act on beats a perfect transcript nobody opens.
Teams that operate across three or more timezones should treat the roll-up as a shared newspaper, not as a chat that demands instant replies from every member. The newspaper metaphor clarifies ownership: someone publishes, someone reads the front page (blockers), and follow-up interviews (huddles) happen only when needed.
When you paste updates into Standup AI, keep secrets out of the input the same way you would keep them out of a shared Google Doc. Access tokens, customer PII, and unreleased pricing should stay in systems designed for them. The roll-up is for work status, not for credential sharing.
Leaders sometimes ask for daily narrative length as a proxy for diligence. Resist that. Longer paragraphs rarely improve planning quality. Prefer crisp lines that a teammate can act on, and let the tracker hold detailed history. The summary should be shorter than the input paste whenever possible.
If your organization still requires a live status meeting for political reasons, you can still use an async roll-up as the pre-read. People arrive already knowing the blockers. The live slot becomes decision time rather than discovery time. That hybrid pattern is honest about power while still reducing waste.
Document the cut-off time in the team README. New hires should learn where to paste and when the summary appears without asking in chat. Process that lives only in someone's head will collapse the first week that person is out sick.
For contractors and part-time contributors, async roll-ups reduce the tax of joining a live call at an awkward hour. They can paste once during their working window and still appear in the shared risk list without pretending they share the same morning as the core team.
Design the paste format so a new teammate can succeed on day one. A three-line example pinned in the channel beats a long policy doc. Show one good update and one weak update side by side so people learn by contrast.
When blockers involve vendors or other teams, include the external dependency name and the last contact date. Vague blockers like waiting on them waste the next day. Specific blockers create a natural follow-up action for whoever owns the relationship.
Older versions of Scrum popularized three questions: what did you do yesterday, what will you do today, and what is in your way. Many teams still use a Done / Doing / Blocked variant. The point was never bureaucracy. The point was inspection: can the team see progress toward the goal, and can impediments be named early enough that someone can help?
In a live standup, social pressure and facilitation push impediments into spoken time. In async chat, people bury the hard line at the end of a paragraph: shipped the docs PR, reviewing billing copy, also waiting on staging secrets for three days. Readers skim the first clause. The blocker never becomes a team problem; it stays a private sentence.
Keep the input cheap. Ask people to write short lines, not essays. Prefer attributable notes when names matter: Alex finished API auth; working on rate limits; blocked on staging keys. If your culture already uses tickets, referencing a ticket id is fine. If it does not, plain language is fine.
Pick a daily cut-off that matches how your team actually works, not a theoretical UTC midnight. Publish the roll-up at a predictable time. Ask at least one person to spend five minutes reacting: claim a blocker, schedule a huddle, or say no action needed. Without that reaction step, async standup fails quietly.
Standup AI takes pasted updates and merges them into one Done / Doing / Blocked summary with blockers first. It can format for Markdown, Slack-friendly text, or email paste. It is a writing and roll-up aid.