Run a 15-Minute Weekly Improvement Meeting

A small team often knows where work gets stuck, but observations made during a busy day are forgotten by the next week. A short improvement meeting gives those observations a place to become one test. Fifteen minutes is enough to review one previous action, select one current problem, assign one person and agree how to tell whether the change helped. It is not enough to resolve every operational dispute. The meeting works when its output is a specific experiment, not a collection of complaints or an extra layer of reporting.

Prepare before the clock starts

Choose a consistent time when the people closest to the work can attend. A five-person cleaning business might meet before the first Monday visits or immediately after Friday jobs. Keep a visible list of observations gathered during the week: missing access codes, late supplies, repeat visits or confusing customer instructions. Record each issue in one sentence with a real case. Do not ask staff to prepare a presentation. The owner or facilitator brings last week's measure and whether the promised trial happened. If the meeting requires hours of preparation, the format defeats its purpose.

Select a measure before you debate solutions. For a problem with missing keys, count visits delayed because access information was unavailable. For repeat cleaning, count jobs that needed a return visit. For a recurring handoff, count the minutes from finished work to customer confirmation. Also choose a balancing measure if a faster step could cause a worse result, such as complaints or missed safety checks. The Institute for Healthcare Improvement's improvement model asks how the team will know that a change is an improvement and encourages measures over time, not impressions alone.

A timed 15-minute agenda

Minutes 0–3: Check last week's experiment. Did the named owner do it, for how many cases, and what happened to the chosen measure? If it was not done, ask what prevented it without assigning blame. Minutes 3–6: Select one issue from the observed cases. Agree on its start and end and make sure the team is solving a problem it can influence. Minutes 6–11: Choose one small change to try on a few cases. Decide who will do it, what information or permission is needed, and what would make the team stop. Minutes 11–14: State the measure, baseline and date of review. Minute 14–15: Repeat the decision aloud and record owner, action, measure and deadline in one line.

Protect that timetable. A complex staffing dispute, safety event or customer complaint needs the correct separate route and more than a rushed final minute. Put it on a follow-up list with an owner and time, then return to the meeting's limited purpose. Avoid using the meeting to announce a solution nobody doing the work has tested. A person who sees the problem firsthand should help describe the actual sequence, even if the owner retains the final decision.

One improvement cycle in a real setting

Assume a cleaning firm completes 20 similar visits per week. Over one week, four visits started late because the team's access instructions lacked a current door code. The team proposes a single field in its booking record for access information, verified with the customer before the first visit. One coordinator owns the field; the cleaner confirms it is available before leaving. For the next ten visits requiring access codes, the team records whether the code was present, whether a late start occurred and whether customers reported a problem with the confirmation request. These numbers are illustrative, not typical industry results.

At the next meeting, suppose nine of ten records contain a verified code and none of those nine visits started late for that reason, but the tenth lacked a code and was delayed. The team should not claim universal success. It learns that the rule worked when followed and asks why the tenth case was missed. Perhaps a last-minute booking entered through another channel. The owner can test a change to that one exception in the next cycle. If customers object to an insecure way of sharing codes, stop and improve the information-handling method before expanding it. This is a practical plan, do, study and act cycle, as described by the Institute for Healthcare Improvement.

Keep the record small and honest

A meeting log needs only date, observed problem, proposed change, owner, sample, measure, result and next decision. Mark an experiment “completed” only when it was tried and reviewed, not when someone suggested it. Over a month, count experiments completed and how often the chosen measure moved in the intended direction without worsening the safeguard. That does not prove the meeting alone caused the change, but it shows whether the team is turning observations into action. Note unusual cases and seasonal effects that may explain a short-term movement.

If the same issue returns, inspect the work itself. Map a Workflow in 30 Minutes helps locate the handoff or wait, while Find and Remove Double Work focuses on repeated entry and requests. Avoid reading all three guides aloud at the meeting. Bring one specific case and ask what changed between steps. A clear question often saves more meeting time than a more elaborate agenda.

When the format needs adjustment

A tiny owner-operated shop may not need a standing group meeting; a 15-minute weekly review with one colleague may be enough. A larger service team might rotate attendance while collecting observations from everyone. If there is no completed trial for several weeks, the proposed changes may be too broad or ownership unclear. Shrink the experiment to one shift, one customer type or five cases. If the chosen measure does not reflect the problem, define it again before claiming improvement. If staff cannot safely report a mistake, the meeting becomes a performance ritual instead of a useful way to learn.

Do not remove a compliance, safety or independent approval check through a quick vote. Clarify its purpose and consult the right specialist. Employment consultation and personal data rules vary by location. A meeting is also not a substitute for immediate escalation of a serious customer harm, equipment failure or security incident. The goal is to manage modest operational improvements while urgent issues follow their proper route.

Next step: put fifteen minutes on next week's calendar and ask each participant to bring one observed delay with a real example. Leave with exactly one owner, one small test on the next five suitable cases and a measure to review the following week.

Sources

• Institute for Healthcare Improvement: Establishing measures. Guidance on using measures to assess improvement.
• Institute for Healthcare Improvement: Testing changes. Guidance on planning, trying and learning from small changes.