# More visitors, a lower rate, and a misleading headline

Companion to /learn/measurement/monthly-reporting/

## Outcome
Write a monthly report that ends in decisions, owners and explicit uncertainty.

Fictional example, not a client result.

## Scenario
StudySpark is a fictional revision app. In an illustrative month it receives 1,000 visitors and 200 defined actions. In the next it receives 2,000 visitors and 300 actions. The action rate falls from 20% to 15%, while the action count rises by 100. Either “conversion collapsed” or “ASO is winning” would omit important context.

## Reasoning
Start with comparable metric definitions and segment the audience where useful. Investigate whether a campaign or seasonal demand brought different visitors. Place actual listing changes beside the timeline, not in a disconnected activity appendix. A monthly report should distinguish what changed, what may explain it, and what remains unproven.

## Action
Report both the rate and count, then investigate source and market composition before recommending creative changes. Summarise the strongest evidence, list competing explanations and assign the next check. End with a small decision register: retain, change, investigate or defer. Readers should know who is responsible and what evidence will be reviewed next.

## Trap to avoid
Do not turn the monthly report into a catalogue of tasks. A large keyword export, a meeting count and a set of design drafts describe effort. They need to connect to a decision or completed deliverable before they explain progress.

## Advanced considerations
Preserve last month’s unresolved questions and revisit them explicitly. Otherwise the report can become a series of fresh narratives that never checks prior assumptions. Keep a compact decision log recording what was chosen, why, and what later evidence showed. This creates institutional learning even when the same people are not present at every review.

## Exercise
The only work completed this month was research and a screenshot brief. How should the report discuss performance?

## Suggested answer
Report delivery separately from outcomes. Explain what was learned and what is ready for implementation, then describe observed performance without claiming that unpublished work caused it. Name the release dependency and next review.

## Your worksheet

### Reporting period and metric definitions
Answer:
Evidence:
Uncertainty:
Owner:
Next action:

### Comparable baseline and current counts
Answer:
Evidence:
Uncertainty:
Owner:
Next action:

### Source or market composition changes
Answer:
Evidence:
Uncertainty:
Owner:
Next action:

### Published changes and concurrent events
Answer:
Evidence:
Uncertainty:
Owner:
Next action:

### Evidence, competing explanations and uncertainty
Answer:
Evidence:
Uncertainty:
Owner:
Next action:

### Decision, owner and next review
Answer:
Evidence:
Uncertainty:
Owner:
Next action:

## A one-page decision report for StudySpark
This completed report outline uses the fictional example above: actions increase from 200 to 300 while visitors rise from 1,000 to 2,000. It deliberately avoids attributing the change to ASO before source and release evidence has been reviewed.

Fictional report example. It distinguishes observed counts, hypotheses and unpublished work.
Report section | Completed editorial example | Evidence or next step
Headline observation | Defined actions rose by 100; action rate fell from 20% to 15% | Retain comparable counts and source definitions
Interpretation | Audience composition may have changed | A hypothesis, not an established cause
Delivery | Screenshot brief completed; publication still awaiting approval | Do not attribute performance to unpublished creative
Decision | Keep the current creative while checking source composition | Analyst reviews comparable source segments
Dependency | Publishing owner needs product approval of revised claims | Schedule review; record actual publication date
Next review | Revisit the explanation after source analysis and release verification | Carry unresolved questions into the following report

### Write a headline that survives scrutiny
“More defined actions from a larger audience, with a lower overall rate” describes the evidence more accurately than “ASO won” or “conversion collapsed.” Follow it with the decision that matters. Include the source, periods and filters beside the underlying chart so the reader can check whether the comparison is valid.

### Close last month’s actions before inventing new ones
Begin the review by checking whether previously assigned work was completed, blocked or superseded. Link each outcome to its evidence. If the source analysis was never done, retain the uncertainty rather than replacing it with a fresh narrative. A small unresolved-questions register makes reporting cumulative and prevents the same investigation being proposed every month.

### Use a stable appendix for reproducibility
Keep metric definitions, source exports, release versions and detailed calculations outside the executive narrative but accessible from it. State any changes to the reporting method. If a chart’s meaning changes between months, mark the break instead of preserving a visually smooth series. This makes the report useful to someone who was not in the meeting.

## Continue reading
- /learn/measurement/aso-kpis/
- /learn/measurement/metadata-change-log/
- /learn/aso-roadmap/