Watch Item

A watch item is something the team agrees to keep an eye on after a decision is made, with a clear line that, if crossed, means the decision should be looked at again.

For example, the team decides to roll out the points progress bar. They agree to watch rewards redemptions, and if redemptions fall below 20%, they will reopen the progress bar decision. That agreement is a watch item.

Why it matters

Most decisions are made and then forgotten. The world changes, the numbers drift, and nobody notices until a quarterly review, or until a customer complaint lands on a leader's desk. By then the fix is expensive.

A watch item keeps a decision honest after it's made. It names the number to watch, the line that matters, who is watching and how often. When the line is crossed, the team already knows which decision to revisit, so they can act early without reopening everything.

For every important decision, name what to watch, set the threshold in advance, give it an owner, and agree how often to check.

How we write one

Every watch item is written as one sentence that names the metric, the threshold and the decision it can reopen:

Watch a metric; if it crosses a threshold, reopen a decision.

Watch rewards redemptions; if they fall below 20%, reopen the progress bar decision.

Writing it this way sets the trigger before anyone sees the numbers, so the team reacts to a line they agreed on rather than to a bad week.

How it differs from similar ideas

  • A risk is something that could go wrong. A watch item is a specific number being monitored, with a line that triggers action; a risk often gets a watch item to track it.

  • A UX metric is a measure of experience. A watch item uses a metric, adds a threshold, and ties it to a decision.

  • A benchmark is a comparison point, such as a baseline. A watch item's threshold is the line that reopens a decision.

  • A decision is the call that was made. A watch item is how the team knows when that call needs another look.

A good watch item

  • Names one metric

  • Sets a clear threshold in advance

  • Names the decision it can reopen

  • Says how often it will be checked

  • Has an owner

  • Comes from findings or a bet

What happens to it

  1. Raised. Findings or a bet surface something worth monitoring.

  2. Set. The owner agrees the metric, threshold, decision and check frequency.

  3. Watching. The metric is checked on schedule and stays within the line.

  4. Tripped. The metric crosses the threshold and the linked decision is reopened.

  5. Closed. The watch ends when the decision is settled, replaced, or no longer needs monitoring.


For Ray

This part tells Ray how to find a watch item in an input, such as a findings readout, a decision memo, a launch checklist, a dashboard alert setup or a retro note, and how to store it.

How to find one

  1. Mark statements about monitoring. Read the input and mark every statement about keeping an eye on a number or behavior after a decision.

  2. Find the metric. Pull out what is being watched. Link it to an existing UX metric or KPI if there is one.

  3. Find the threshold. Look for the line that triggers action, such as "below 20%" or "rises above $50". If none is given, leave it blank.

  4. Find the decision. Identify which decision or bet it can reopen. If none is named, leave it blank.

  5. Find the frequency and owner. Note how often it will be checked and who is responsible.

  6. Check the current value. If the latest value is in the input, compare it to the threshold.

  7. Write the sentence. Use the format: Watch [metric]; if it [threshold], reopen [decision].

  8. Fill the record, score it, link it and store it using the rules below.

Cues to look for

  • Monitoring language: "keep an eye on", "we'll watch", "set an alert if", "revisit if"

  • Thresholds: "if it drops below", "anything over", "as long as it stays above"

  • Follow-up sections in findings readouts and decision memos

  • Dashboard alerts and guardrail metrics in experiment plans

What doesn't count

<table xmlns="http://www.w3.org/1999/xhtml" style="min-width: 50px;"><colgroup><col style="min-width: 25px;"><col style="min-width: 25px;"></colgroup><tbody><tr><th colspan="1" rowspan="1"><p>If the statement...</p></th><th colspan="1" rowspan="1"><p>Store it as</p></th></tr><tr><td colspan="1" rowspan="1"><p>Describes something that could go wrong, with no metric or threshold</p></td><td colspan="1" rowspan="1"><p>Risk</p></td></tr><tr><td colspan="1" rowspan="1"><p>Reports a metric value with no threshold or decision</p></td><td colspan="1" rowspan="1"><p>UX Metric</p></td></tr><tr><td colspan="1" rowspan="1"><p>Sets a baseline to compare against</p></td><td colspan="1" rowspan="1"><p>Benchmark</p></td></tr><tr><td colspan="1" rowspan="1"><p>Suggests a next action from testing</p></td><td colspan="1" rowspan="1"><p>Recommendation</p></td></tr><tr><td colspan="1" rowspan="1"><p>Reports a measured change after launch</p></td><td colspan="1" rowspan="1"><p>User, Product or Business Outcome</p></td></tr></tbody></table>

The record

<table xmlns="http://www.w3.org/1999/xhtml" style="min-width: 75px;"><colgroup><col style="min-width: 25px;"><col style="min-width: 25px;"><col style="min-width: 25px;"></colgroup><tbody><tr><th colspan="1" rowspan="1"><p>Field</p></th><th colspan="1" rowspan="1"><p>What to fill in</p></th><th colspan="1" rowspan="1"><p>Values</p></th></tr><tr><td colspan="1" rowspan="1"><p>Statement</p></td><td colspan="1" rowspan="1"><p>The one-sentence watch item</p></td><td colspan="1" rowspan="1"><p>Watch [metric]; if it [threshold], reopen [decision].</p></td></tr><tr><td colspan="1" rowspan="1"><p>Metric</p></td><td colspan="1" rowspan="1"><p>What is being watched</p></td><td colspan="1" rowspan="1"><p>Link to UX Metric or KPI</p></td></tr><tr><td colspan="1" rowspan="1"><p>Threshold</p></td><td colspan="1" rowspan="1"><p>The line that triggers action</p></td><td colspan="1" rowspan="1"><p>Value with direction, such as "below 20%"</p></td></tr><tr><td colspan="1" rowspan="1"><p>Decision</p></td><td colspan="1" rowspan="1"><p>The decision or bet it can reopen</p></td><td colspan="1" rowspan="1"><p>Link to Decision or Bet</p></td></tr><tr><td colspan="1" rowspan="1"><p>Check frequency</p></td><td colspan="1" rowspan="1"><p>How often it is checked</p></td><td colspan="1" rowspan="1"><p>Daily, Weekly, Monthly, Quarterly</p></td></tr><tr><td colspan="1" rowspan="1"><p>Owner</p></td><td colspan="1" rowspan="1"><p>The person responsible for watching it</p></td><td colspan="1" rowspan="1"><p>Name or role</p></td></tr><tr><td colspan="1" rowspan="1"><p>Latest value</p></td><td colspan="1" rowspan="1"><p>The most recent reading and date</p></td><td colspan="1" rowspan="1"><p>Value and date, or blank</p></td></tr><tr><td colspan="1" rowspan="1"><p>Raised from</p></td><td colspan="1" rowspan="1"><p>Where it came from</p></td><td colspan="1" rowspan="1"><p>Link to Findings, Bet or Risk</p></td></tr><tr><td colspan="1" rowspan="1"><p>Source</p></td><td colspan="1" rowspan="1"><p>The input and exact passage</p></td><td colspan="1" rowspan="1"><p>Link to Input</p></td></tr><tr><td colspan="1" rowspan="1"><p>Status</p></td><td colspan="1" rowspan="1"><p>Where it stands</p></td><td colspan="1" rowspan="1"><p>Raised, Set, Watching, Tripped, Closed</p></td></tr><tr><td colspan="1" rowspan="1"><p>Confidence</p></td><td colspan="1" rowspan="1"><p>How clearly it was stated</p></td><td colspan="1" rowspan="1"><p>High, medium, low</p></td></tr></tbody></table>

New watch items start as Raised. A watch item can't move to Watching until it has a threshold, a decision and an owner.

Scoring confidence

Confidence measures how clearly the watch item was set up, not how likely the threshold is to trip.

<table xmlns="http://www.w3.org/1999/xhtml" style="min-width: 50px;"><colgroup><col style="min-width: 25px;"><col style="min-width: 25px;"></colgroup><tbody><tr><th colspan="1" rowspan="1"><p>Level</p></th><th colspan="1" rowspan="1"><p>When to use it</p></th></tr><tr><td colspan="1" rowspan="1"><p>High</p></td><td colspan="1" rowspan="1"><p>Written down with a metric, a threshold, a decision to reopen and an owner</p></td></tr><tr><td colspan="1" rowspan="1"><p>Medium</p></td><td colspan="1" rowspan="1"><p>The metric is clear, but the threshold, decision, owner or frequency is missing</p></td></tr><tr><td colspan="1" rowspan="1"><p>Low</p></td><td colspan="1" rowspan="1"><p>A vague intention to "keep an eye on it" with no metric or threshold</p></td></tr></tbody></table>

If any part of the record can't be filled from the source, leave it blank and lower confidence by one level.

Linking

  • Check for duplicates first. If an existing watch item tracks the same metric for the same decision, add the new source to it instead of creating another.

  • Link to the findings or bet that raised it and its source input.

  • Link to the decision it can reopen and the metric it watches.

  • Link to any risk it is tracking.

  • When it trips, link it back to Define and Collecting so the reopened decision starts with fresh inputs.

When to store and when to ask

  • Store automatically when confidence is high.

  • Add to the review digest when confidence is medium or low, when it has no threshold or decision, or when it has tripped.

When to flag

  • A watch item's threshold has been crossed

  • A watch item has no threshold or no decision to reopen

  • A watch item has no owner

  • A watch item hasn't been checked within its stated frequency

  • A bet went Live with no watch items set

Worked example

Input: the follow-up section of the value test findings readout.

"After rollout, watch redemptions. If fewer than 20% of new cardholders redeem within 60 days, we should revisit the progress bar. Analytics lead to check monthly."

What Ray finds:

  • "Redemptions" is the metric.

  • "Fewer than 20% of new cardholders redeem within 60 days" is the threshold.

  • "Revisit the progress bar" names the decision to reopen.

  • "Analytics lead to check monthly" gives the owner and the check frequency.

Stored record:

<table xmlns="http://www.w3.org/1999/xhtml" style="min-width: 50px;"><colgroup><col style="min-width: 25px;"><col style="min-width: 25px;"></colgroup><tbody><tr><th colspan="1" rowspan="1"><p>Field</p></th><th colspan="1" rowspan="1"><p>Value</p></th></tr><tr><td colspan="1" rowspan="1"><p>Statement</p></td><td colspan="1" rowspan="1"><p>Watch redemptions; if they fall below 20%, reopen the progress bar decision.</p></td></tr><tr><td colspan="1" rowspan="1"><p>Metric</p></td><td colspan="1" rowspan="1"><p>Rewards redemptions within 60 days, new cardholders</p></td></tr><tr><td colspan="1" rowspan="1"><p>Threshold</p></td><td colspan="1" rowspan="1"><p>Below 20%</p></td></tr><tr><td colspan="1" rowspan="1"><p>Decision</p></td><td colspan="1" rowspan="1"><p>We decided to test a points progress bar because new cardholders can't see value; owner product lead.</p></td></tr><tr><td colspan="1" rowspan="1"><p>Check frequency</p></td><td colspan="1" rowspan="1"><p>Monthly</p></td></tr><tr><td colspan="1" rowspan="1"><p>Owner</p></td><td colspan="1" rowspan="1"><p>Analytics lead</p></td></tr><tr><td colspan="1" rowspan="1"><p>Latest value</p></td><td colspan="1" rowspan="1"><p>Blank</p></td></tr><tr><td colspan="1" rowspan="1"><p>Raised from</p></td><td colspan="1" rowspan="1"><p>Findings: progress bar raised task success to 78% on the value test</p></td></tr><tr><td colspan="1" rowspan="1"><p>Source</p></td><td colspan="1" rowspan="1"><p>Value test findings readout, follow-up section</p></td></tr><tr><td colspan="1" rowspan="1"><p>Status</p></td><td colspan="1" rowspan="1"><p>Set</p></td></tr><tr><td colspan="1" rowspan="1"><p>Confidence</p></td><td colspan="1" rowspan="1"><p>High: metric, threshold, decision and owner all stated in writing</p></td></tr></tbody></table>

Next step: stored automatically; Ray will check the analytics export monthly and raise a flag if redemptions fall below 20%.

Edge cases

  • No threshold. "Keep an eye on redemptions" has no line. Store it at low confidence and prompt the owner to set one.

  • Threshold tripped. Mark the watch item Tripped, flag it in the review digest, reopen the linked decision, and send the new reading back to Define.

  • Several metrics in one. "Watch redemptions and support tickets" is two watch items. Split them, each with its own threshold.

  • Threshold changed. Don't overwrite the old line. Record the new threshold, who changed it and why, and keep the old one in history.

Identify where decision quality breaks down

The Glare Design Assessment helps teams spot weak validation, stakeholder friction, alignment gaps, and assumptions that scale without measurable learning—so you have a clearer starting point for improvement.

About 5 minutes · Team-based · Diagnostic snapshot you can act on

Take the Design Assessment