measure-experiment-results

Name: measure-experiment-results
Author: product-on-purpose

486stars64forks

How to install measure-experiment-results?

product-on-purpose/pm-skills/measure-experiment-results

$npx -y skills add product-on-purpose/pm-skills --skill measure-experiment-results

Installs into the current project.

Summary

Documents the results of a completed experiment or A/B test with statistical analysis, learnings, and recommendations. Use after experiments conclude to communicate findings, inform decisions, and build organizational knowledge.

Files · 1

View on GitHub

SKILL.md

1	<!-- PM-Skills \| https://github.com/product-on-purpose/pm-skills \| Apache 2.0 -->
2	# Experiment Results
3
4	An experiment results document captures what happened when you tested a hypothesis, including statistical outcomes, segment analysis, learnings, and clear recommendations. Good results documentation turns individual experiments into organizational knowledge that improves future decision-making.
5
6	## When to Use
7
8	- After an A/B test or experiment reaches statistical significance
9	- When an experiment is ended early (for any reason)
10	- To communicate findings to stakeholders who weren't involved
11	- During decision-making about whether to ship, iterate, or kill a feature
12	- To build a repository of learnings that inform future experiments
13
14	## When NOT to Use
15
16	- The experiment is not designed or run yet -> use `measure-experiment-design`
17	- The results demand a direction decision -> use `iterate-pivot-decision`; this skill reports the evidence, that one decides
18	- You want the transferable learning banked for the organization -> follow up with `iterate-lessons-log`
19	- Your data is survey responses, not a controlled experiment -> use `measure-survey-analysis`
20
21	## Instructions
22
23	When asked to document experiment results, follow these steps:
24
25	1. Summarize the Experiment
26	Provide context: what was tested, when it ran, how much traffic it received. Link to the original experiment design document if one exists.
27
28	2. Restate the Hypothesis
29	Remind readers what you believed would happen and why. This frames the results interpretation.
30
31	3. Present Primary Results
32	Show the primary metric outcome clearly: what were the values for control and treatment? Include statistical significance (p-value), confidence intervals, and sample sizes. Be honest about whether results are conclusive.
33
34	4. Analyze Secondary Metrics
35	Present guardrail metrics that ensure you didn't cause unintended harm. Note any secondary metrics that moved unexpectedly.both positive and negative.
36
37	5. Segment the Data
38	Look for differential effects across user segments (platform, tenure, plan type, etc.). Sometimes overall results mask important segment-level insights.
39
40	6. Extract Learnings
41	What did you learn beyond the numbers? Include surprising findings, questions raised, and implications for the product hypothesis. Negative results are valuable learnings.
42
43	7. Make a Recommendation
44	Be clear: should we ship, iterate, or kill? Support the recommendation with the evidence. If the decision is nuanced, explain the trade-offs.
45
46	8. Define Next Steps
47	Specify what happens now.engineering work to ship, follow-up experiments, metrics to continue monitoring, or documentation to update.
48
49	## Output Format
50
51	Use the template in `references/TEMPLATE.md` to structure the output. A complete readout fills every template section: Summary; Hypothesis Recap; Results; Segment Analysis; Visualization; Learnings; Recommendation; Next Steps; and Appendix.
52
53	## Quality Checklist
54
55	Before finalizing, verify:
56
57	- [ ] Statistical methods and significance are clearly stated
58	- [ ] Confidence intervals are included (not just p-values)
59	- [ ] Segment analysis checked for differential effects
60	- [ ] Secondary/guardrail metrics are reported
61	- [ ] Learnings go beyond just the numbers
62	- [ ] Recommendation is clear and actionable
63	- [ ] Negative or inconclusive results are reported honestly
64
65	## Examples
66
67	See `references/EXAMPLE.md` for a completed example.

People who liked this also viewed

6 picks

Type

measure-experiment-results

486stars64forks

View on GitHub

How to install measure-experiment-results?

product-on-purpose/pm-skills/measure-experiment-results

$npx -y skills add product-on-purpose/pm-skills --skill measure-experiment-results

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	<!-- PM-Skills \| https://github.com/product-on-purpose/pm-skills \| Apache 2.0 -->
2	# Experiment Results
3
4	An experiment results document captures what happened when you tested a hypothesis, including statistical outcomes, segment analysis, learnings, and clear recommendations. Good results documentation turns individual experiments into organizational knowledge that improves future decision-making.
5
6	## When to Use
7
8	- After an A/B test or experiment reaches statistical significance
9	- When an experiment is ended early (for any reason)
10	- To communicate findings to stakeholders who weren't involved
11	- During decision-making about whether to ship, iterate, or kill a feature
12	- To build a repository of learnings that inform future experiments
13
14	## When NOT to Use
15
16	- The experiment is not designed or run yet -> use `measure-experiment-design`
17	- The results demand a direction decision -> use `iterate-pivot-decision`; this skill reports the evidence, that one decides
18	- You want the transferable learning banked for the organization -> follow up with `iterate-lessons-log`
19	- Your data is survey responses, not a controlled experiment -> use `measure-survey-analysis`
20
21	## Instructions
22
23	When asked to document experiment results, follow these steps:
24
25	1. Summarize the Experiment
26	Provide context: what was tested, when it ran, how much traffic it received. Link to the original experiment design document if one exists.
27
28	2. Restate the Hypothesis
29	Remind readers what you believed would happen and why. This frames the results interpretation.
30
31	3. Present Primary Results
32	Show the primary metric outcome clearly: what were the values for control and treatment? Include statistical significance (p-value), confidence intervals, and sample sizes. Be honest about whether results are conclusive.
33
34	4. Analyze Secondary Metrics
35	Present guardrail metrics that ensure you didn't cause unintended harm. Note any secondary metrics that moved unexpectedly.both positive and negative.
36
37	5. Segment the Data
38	Look for differential effects across user segments (platform, tenure, plan type, etc.). Sometimes overall results mask important segment-level insights.
39
40	6. Extract Learnings
41	What did you learn beyond the numbers? Include surprising findings, questions raised, and implications for the product hypothesis. Negative results are valuable learnings.
42
43	7. Make a Recommendation
44	Be clear: should we ship, iterate, or kill? Support the recommendation with the evidence. If the decision is nuanced, explain the trade-offs.
45
46	8. Define Next Steps
47	Specify what happens now.engineering work to ship, follow-up experiments, metrics to continue monitoring, or documentation to update.
48
49	## Output Format
50
51	Use the template in `references/TEMPLATE.md` to structure the output. A complete readout fills every template section: Summary; Hypothesis Recap; Results; Segment Analysis; Visualization; Learnings; Recommendation; Next Steps; and Appendix.
52
53	## Quality Checklist
54
55	Before finalizing, verify:
56
57	- [ ] Statistical methods and significance are clearly stated
58	- [ ] Confidence intervals are included (not just p-values)
59	- [ ] Segment analysis checked for differential effects
60	- [ ] Secondary/guardrail metrics are reported
61	- [ ] Learnings go beyond just the numbers
62	- [ ] Recommendation is clear and actionable
63	- [ ] Negative or inconclusive results are reported honestly
64
65	## Examples
66
67	See `references/EXAMPLE.md` for a completed example.

People who liked this also viewed

6 picks

Type