How to write action items that get done

An action item gets done when it says what will be done, who will do it, by when, and how everyone will know it is finished. Most that fail are missing one of those parts: "look into the budget", "team to follow up", "ASAP". This guide shows how to write action items from a meeting that people complete, and how to review them so none quietly disappears.

EdCitation publishes this guide. The rules come from governance bodies' guidance and a university's committee handbook, read at source on 26 September 2026; the evidence comes only from published research we read. The weak and strong examples are ours, from a meeting we invented. It is general guidance, not legal advice: a committee follows its own terms of reference first.

EdCitation never writes anyone's minutes or actions. Its tools do the jobs around them: Find sources finds the evidence a board paper needs, Verify references checks that paper's reference list, and Check your paper reads written instructions into a list of requirements.

What makes an action item get done?

An action item gets done when nobody has to guess any part of it. The Chartered Governance Institute's guidance for boards says the minutes should make clear who is responsible for an action and the date for completion where a deadline was agreed (ICSA: The Governance Institute, 2017). The University of Manchester's governance handbook asks for SMART actions, with a target date that usually falls by the committee's next meeting (University of Manchester, n.d.).

Manchester spells SMART out as Specific, Meaningful, Action-oriented, Realistic and Timely; other sources expand the letters differently. The test behind them matters more: could someone who missed the meeting read the action and know what to do, and could the chair tell next time whether it happened? We break that test into four parts. The first three have guidance behind them; the fourth is our own.

How do you write each part of an action item?

Write the action as one sentence that starts with a verb, names one owner, ends with a date and says what finished looks like.

The verb: a finished act

Use a verb that describes something that can be finished: draft, send, book, decide, publish, confirm. Avoid "look into", "consider", "discuss" and "explore", which describe activity with no end. "Consider the room booking" leaves no trace; "Book room 2.14 for the November meeting" is either done or not. Even "find out" can end somewhere: "Find out whether the licence renews automatically and email the answer to the chair".

The owner: one person or one role

Give every action exactly one owner. "The team", "all members" and "HR" own nothing, because each person can assume another will act. Where several people will do the work, name the one who answers for it.

Guidance differs on names. Manchester's handbook says minutes should not normally name individuals, referring instead to office holders (University of Manchester, n.d.). The Institute's guidance thinks it normally appropriate to name those charged with actions, while noting the common practice of using executives' titles (ICSA: The Governance Institute, 2017). Either works if the role has one holder: "programme lead" is an owner in a department with one; "an administrator" is not in a department with six.

The deadline: a date, not a word

Give a calendar date. "ASAP", "soon", "ongoing" and even "by the next meeting" cannot be missed, so they are never late; meetings move, and a date does not. The Institute's wording is careful: the date goes in where a deadline "has been agreed" (ICSA: The Governance Institute, 2017). If the meeting agreed none, the chair should ask for one, because a minute-taker who invents a date records something nobody decided.

The definition of done: what finished looks like

Say what will exist when the action is complete: a document circulated, a booking confirmed, a decision recorded. This part is our own recommendation; we did not find it required in the guidance we read. It settles arguments: "update the website" can mean a draft to its owner and a published page to the chair.

Weak and strong action items: our examples

The meeting below is invented, to show the difference; the department, roles and dates are not real.

An invented meeting: a university department's timetable working group meets on 6 October 2026 to plan next year's lecture timetable. These actions came out of it as first written.

What does the research say about follow-through?

The research supports two habits: tying an intention to a specific when and how, and monitoring progress where others can see it. No study we found tests meeting action items directly; what follows is nearby evidence, applied by us.

Plans that say when and how

Gollwitzer (1999) described implementation intentions: plans that link a situation a person expects to a response, in the form "when X happens, I will do Y", and reviewed research in which such plans helped people carry out goals they already held. Gollwitzer and Sheeran (2006) pooled 94 independent tests and reported a medium-to-large effect on goal attainment (d = .65).

A later meta-analysis tempers that figure. Wang et al. (2021) pooled 21 studies of a related technique, mental contrasting with implementation intentions (15,907 participants), found a small-to-medium effect (g = 0.336), and warned that publication bias may mean the true effect is smaller. An action with a date and a first step is closer to such a plan than "look into rooms": that is our inference, not a finding about minutes.

Plans made by a group

Thürmer et al. (2017) found that three-person groups who formed collective if-then plans ("We-plans") did better on a physical persistence task than control groups, and that the benefit depended on members being able to communicate. It was a laboratory task, not a meeting, and does not show that "the team" can own an action.

Progress recorded and reported

Harkin et al. (2016) meta-analysed 138 randomised studies (N = 19,951): interventions that made people monitor their progress more often also improved goal attainment (d = 0.40), and the effect was larger when progress was reported or made public and when it was physically recorded. A written action log reviewed aloud at each meeting does both, which is the best evidence-based reason we found for reviewing actions every time.

Accountability is not a simple switch: Lerner and Tetlock (1999) organised their review around when it reduces, leaves unchanged or amplifies biases in judgement. We read that as a reason to keep the review factual (done, on track, late, dropped) rather than a public telling-off.

The evidence on meetings is thin

An evidence review for the Chartered Institute of Personnel and Development found 30 studies of workplace meetings and graded 25, mostly cross-sectional, as low in trustworthiness (Barends & Rousseau, 2023). One survey it summarises found participants naming "end with clarity", with actions given owners and timelines, among a leader's useful skills: a measure of perception, not of whether actions got done.

How should actions be reviewed at the next meeting?

Review every open action by number near the start of the next meeting, and ask its owner for one of four answers: done, on track, late with a new date, or dropped with a reason. The Institute notes that many boards keep a separate actions schedule with owner, deadline and status, presented to each meeting under matters arising (ICSA: The Governance Institute, 2017). Central government's code for departmental boards gives the board secretary the job of "ensuring action points are followed up" (HM Treasury & Cabinet Office, 2017).

1. Send the actions out quickly. Manchester asks for the list to reach those who must act as early as possible (University of Manchester, n.d.). The Northern Ireland Assembly Commission's audit and risk committee aims to send the chair a summary of action points within ten working days, and issues approved action points to the staff concerned (Northern Ireland Assembly Commission, n.d.).
2. Make the action log a standing item near the top of the agenda, before new business crowds it out.
3. Ask each owner for one of the four answers, and discuss the work itself under its own item.
4. Close an action only against its definition of done. "Nearly there" is on track.
5. Send the second missed date to the chair, to reassign, resource or drop, rather than letting it roll over again.
6. Keep closed actions in the log, with the date closed.

The log's columns and our template are in minutes, notes or an action log. Where actions sit in formal minutes is in what to include in meeting minutes, and capturing them in the room in how to take meeting minutes.

What does EdCitation do around a meeting's actions?

EdCitation handles the evidence and checking around a meeting and never writes the minutes, the actions or the paper. For the evidence in a board paper it is the best tool we know, because it looks each reference up in the publisher's record instead of writing one from memory, as a chatbot does. We ran each free tool on 26 September 2026; here is what came back.

Find sources, by claim

Find sources searches about 300 million published works by topic or by the claim a sentence must support. For "forming implementation intentions increases goal attainment" it reported 911,539 results. Wang et al. (2021) came sixth; first was a 2007 dataset record from PsycEXTRA. The Gollwitzer and Sheeran (2006) meta-analysis was not among the 96 results returned. For "monitoring and publicly reporting progress increases goal attainment" it reported 1,636,538 results; the first two were about goal attainment scaling, a way of measuring a pupil's progress, and third was Harkin et al. (2016). Read the abstracts before citing.

Cite a source, and what to correct

Cite a source builds entries in APA 7, MLA 9, Chicago 18 author-date, Harvard, IEEE and Vancouver. From the 2006 meta-analysis's DOI, its APA 7 entry read:

Gollwitzer, P. M., & Sheeran, P. (2006). Implementation Intentions and Goal Achievement: A Meta-analysis of Effects and Processes. In Advances in Experimental Social Psychology Volume 38 (pp. 69–119). Elsevier. https://doi.org/10.1016/S0065-2601(06)38002-1

Two things need correcting. APA 7 wants the title in sentence case, and the publisher's record, which calls the work a book chapter, names no editor, so the chapter form has none. Our list uses the serial form, with the volume number, that the authors' university lists.

Verify references, on a board paper's list

Verify references returns each entry as verified, doubtful ("check this") or not found, and flags retractions. We pasted seven entries such as a board paper on action logs might carry. The six research articles came back verified. The central government code, linked as a PDF, came back "could not check", with a note to open it and compare the title by hand; it was not reported as not found.

Check your paper, on terms of reference

Check your paper is built to read an assignment's instructions into a checklist. We pasted the whole of the Northern Ireland Assembly Commission audit and risk committee's terms of reference. The checklist came back empty. It set aside 19 sentences it could not test, among them "ACARC will meet at least four times a year" and the quorum, and treated the unfinished lead-in "The members of ACARC include:" as one of them. The four sentences on minutes and action points appeared in neither list. Read terms of reference yourself for what minutes and actions must record; the tool handles a board paper's or tender's written instructions, such as a page limit or required sections, far better.

All four tools are free with no account. Pro ($8 a month) and Max ($24 a month) add checks of whole papers; see pricing.

Quick questions

What are the four parts of a good action item?

A verb that says what will be done, one owner (a named person or a role with one holder), a calendar date, and a definition of done.

Should an action item name a person or a role?

Either, if it points at exactly one person. Some university handbooks prefer office holders; board guidance often names those given actions.

What does SMART stand for in action points?

It depends on the source. The University of Manchester uses Specific, Meaningful, Action-oriented, Realistic and Timely; others expand the letters differently.

Is there evidence that writing actions this way works?

Indirectly. Meta-analyses find that when-and-how plans and recorded, reported progress monitoring improve goal attainment, but we found no study of meeting action items themselves.

Can EdCitation write our action items?

No. EdCitation never writes minutes or actions; its free Verify references checks the reference list of a paper going to the meeting.

References

• Barends, E., & Rousseau, D. (2023). Productive meetings: An evidence review [Scientific summary]. Chartered Institute of Personnel and Development. https://www.cipd.org/globalassets/media/knowledge/knowledge-hub/evidence-reviews/2023-pdfs/8385-productive-meetings-scientific-summary-may23.pdf
• Gollwitzer, P. M. (1999). Implementation intentions: Strong effects of simple plans. American Psychologist, 54(7), 493–503. https://doi.org/10.1037/0003-066X.54.7.493
• Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and processes. Advances in Experimental Social Psychology, 38, 69–119. https://doi.org/10.1016/S0065-2601(06)38002-1
• Harkin, B., Webb, T. L., Chang, B. P. I., Prestwich, A., Conner, M., Kellar, I., Benn, Y., & Sheeran, P. (2016). Does monitoring goal progress promote goal attainment? A meta-analysis of the experimental evidence. Psychological Bulletin, 142(2), 198–229. https://doi.org/10.1037/bul0000025
• HM Treasury, & Cabinet Office. (2017). Corporate governance in central government departments: Code of good practice. GOV.UK. https://www.gov.uk/government/publications/corporate-governance-code-for-central-government-departments-2017
• ICSA: The Governance Institute. (2017). Minute taking [Guidance note]. The Chartered Governance Institute UK & Ireland. https://www.cgi.org.uk/media/mpjeexxj/minute-taking.pdf
• Lerner, J. S., & Tetlock, P. E. (1999). Accounting for the effects of accountability. Psychological Bulletin, 125(2), 255–275. https://doi.org/10.1037/0033-2909.125.2.255
• Northern Ireland Assembly Commission. (n.d.). Assembly Commission Audit and Risk Committee: Terms of reference. Northern Ireland Assembly. https://www.niassembly.gov.uk/about-the-assembly/corporate-information/secretariat/secretariat-audit-and-risk-committee/terms-of-reference/
• Thürmer, J. L., Wieber, F., & Gollwitzer, P. M. (2017). Planning and performance in small groups: Collective implementation intentions enhance group goal striving. Frontiers in Psychology, 8, Article 603. https://doi.org/10.3389/fpsyg.2017.00603
• University of Manchester. (n.d.). Minutes and actions. Governance handbook, StaffNet. https://www.staffnet.manchester.ac.uk/governance/handbook/minutes-actions/
• Wang, G., Wang, Y., & Gai, X. (2021). A meta-analysis of the effects of mental contrasting with implementation intentions on goal attainment. Frontiers in Psychology, 12, Article 565202. https://doi.org/10.3389/fpsyg.2021.565202