AI assistants like ChatGPT, Gemini, and Grok can be powerful analysis partners — when properly structured. Because Godcaster exports include structural nuances (attribution fields, action types, device indicators, session reconstruction limits, timestamp precision, and play-counting rules), how you prompt the AI directly affects accuracy. To produce consistent, board-ready reports, Godcaster recommends using our developer tested and authenticated prompts. Computation is separated from narrative to improve reliability across AI tools.

How to Use This
• Export your monthly performance CSV from the Godcaster dashboard
• Open your AI assistant (ChatGPT, Gemini, etc.)
• Upload the CSV
• Copy prompts and run it
• Do not shorten the prompts, as the structure is what ensures accuracy

Purpose

Analyze only the uploaded Godcaster Digital Station Statistics export or exports.

Every reported metric, observation, recommendation, and conclusion must be supported directly by the uploaded export or exports.

Do not use:

• Aptabase
• Internal analytics systems
• Previous reports
• Previous conversations
• Websites
• Outside knowledge
• Historical information not contained in the uploaded export or exports

Even when the customer uses The Station App, report only activity directly contained and attributed within the Godcaster export.

Do not report or infer app listening hours, app sessions, deep-link visits, attributed app installations, install conversion, app retention, car connections, or any other Aptabase-only measurement.

Godcaster Mobile App listening and engagement may be reported only when directly supported by fields in the Godcaster export, such as Action, App, Platform, Device, Referer, Program, or Episode.

When multiple Godcaster exports are uploaded, compare reporting periods only when their fields, definitions, and methodologies are consistent.

If the data does not support a conclusion, omit the conclusion. State a limitation only when it materially affects leadership's interpretation of an important finding.

Reporting Objective

Write the report as though presenting it personally to the station's General Manager and leadership team.

The report should help leadership understand:

• The most important performance story
• What is working particularly well
• How listeners are using the Digital Station
• Where a practical opportunity exists
• What should continue
• What leadership should pay attention to next

The report is not an inventory of every available statistic. It is an executive interpretation of the findings most useful to the customer.

Governing Editorial Principle

Favor relevance over completeness.

Before including any metric, table, observation, limitation, or recommendation, determine whether it does at least one of the following:

• Demonstrates a meaningful accomplishment or encouraging trend
• Reveals an important audience, listening, platform, or content pattern
• Identifies a material change from a comparable prior period
• Supports a practical customer decision
• Identifies a constructive opportunity for growth
• Prevents leadership from misunderstanding a major result

If it does none of these, omit it.

Do not include a section or interpretation merely because a field exists in the export. Do not attempt to demonstrate that every available data point was analyzed.

Tone and Framing

The report should be:

• Executive
• Professional
• Warm
• Encouraging
• Objective
• Evidence-based

Accentuate genuine accomplishments and positive trends without exaggerating them.

Use a positive-first narrative order:

• Lead with the strongest meaningful result.
• Explain why that result matters.
• Describe the most useful supporting patterns.
• Present the clearest opportunity constructively.
• Finish with a practical next step and forward-looking direction.

Never conceal or minimize a material negative result. Give it proportional attention, place it in the larger performance context, and describe the practical opportunity it creates.

Do not treat a small decline, zero value, or absent action as a headline unless it is strategically important, materially changed, or directly supports a useful recommendation.

When zero activity is strategically important, describe it as an opportunity to strengthen use of an existing capability, not as customer failure.

Distinguish 'no recorded action' from proof that listeners did not respond.

Avoid promotional language, criticism, alarmist wording, technical jargon, and unsupported explanations of listener motivation.

Do not speculate about why something happened unless the uploaded data directly supports that conclusion.

Reporting Definitions

Play Events - Action = play.

Resume Events - Action = resume.

Total Listening - Play Events + Resume Events.

Total Engagement — Every recorded Action in the Digital Station. Total Engagement includes direct audience activity such as listening, browsing, sharing and following, as well as system/distribution activity such as RSS and initialization. It measures total activity around the Digital Station; it is not a count of listeners.

When system/distribution activity represents a substantial portion of Total Engagement, explain that composition so leadership understands what is driving the total.

Listening Activity - Play + resume.

Navigation Activity - Guide, feed, latest-content, search, and browsing actions.

Response Activity - Share, support, ministry-link, and insert activity.

Ownership Activity - Follow and Progressive Web App activity.

System / Distribution Activity - Initialization, RSS, feed requests, and other technical activity.

Support Classifications

• support:program = Program Support
• support:player = Station Support
• support:player_rss = Station Support

Interpretation Rules

• Support Actions are recorded clicks, not confirmed donations or completed gifts.
• Progressive Web App activity is not an installation unless installation events exist.
• PWA activity may be meaningful because listeners can activate and use the Progressive Web App directly from a browser or desktop. However, do not infer how or why PWA activity occurred unless the export establishes it. Include PWA activity when its volume, trend or relationship to other station use provides a meaningful leadership insight. If its source or significance cannot be determined, it may remain in a KPI table or definitions without receiving executive emphasis.
• RSS or feed requests indicate content distribution activity and are not listeners, listening events, or downloads.
• Technical identifiers are not verified people or verified listeners.
• Plays are already de-duplicated to one Play Event per listener per hour. Do not perform additional de-duplication.
• Reach estimates based on IP addresses or IP + User Agent combinations are directional estimates, not verified audience counts.

Internal Data Validation

Before writing the report, verify internally:

• Reporting period
• File structure and column headers
• Distinct Action values
• Program and Episode availability
• Device, Platform, App, and Referer fields
• Timestamp precision
• Presence or absence of Session IDs
• Missing or unexpected values that materially affect interpretation
• Whether comparison-period definitions are consistent
• Whether totals and percentages reconcile

Determine whether reliable session reconstruction is possible using IP + User Agent with a 30-minute inactivity timeout. If timestamp precision or Session IDs are insufficient, do not reconstruct journeys.

This validation is an internal quality-control step. Do not reproduce the checklist or routine validation commentary in the customer report.

Content Selection Rules

• Build the report around two to four primary performance stories, not around every available field.
• Give the greatest space and emphasis to the findings that matter most to the customer.
• Synthesize related metrics into one executive observation rather than discussing each metric separately.
• Discuss each major finding once. Do not repeatedly restate it in the Executive Summary, KPI commentary, Takeaways, Observations, Recommendations, and Conclusion.
• Omit immaterial fluctuations, insignificant counts, routine technical activity, and unavailable metrics unless they change the interpretation of a major result.
• Do not create commentary to explain every row of a table.
• Do not describe every country, state, device, operating system, program, episode, or Action value. Select only those that reveal a meaningful pattern.
• When a decline is reported, compare it with the larger performance picture and explain its practical significance without dramatizing it.
• When prior-period data is not methodologically comparable, do not calculate or imply a trend.

Executive Report Structure

Adapt the report to the available data. Do not create a section solely because it appears in this outline.

1. Executive Summary

Write approximately 250-400 words as a coherent narrative.

Address these questions in order:

• What was the strongest overall result?
• Why does it matter?
• What is the station doing particularly well?
• What is the clearest constructive opportunity?
• What should leadership watch or act on next?

Use selected KPI values only when they support the narrative. Do not turn the Executive Summary into a list of statistics.

Lead with the strongest meaningful customer result, not with total technical activity, data limitations, or minor declines.

2. Performance Overview

Provide one concise KPI table containing only the measures needed to understand the month's principal performance story.

Core measures may include, when available and useful:

• Total Station Activity
• Total Listening and Play Events
• Live Listening
• On-Demand Listening
• Shares or other meaningful Response Actions
• Directional Listening Reach, when the methodology is consistent and the figure is useful

When comparable prior-period data exists, include the prior period and percentage change. Generally limit the primary table to approximately 6-10 meaningful measures.

Follow the table with one concise observation explaining the most important relationship among the figures. Do not explain every KPI individually.

3. Key Performance Findings

Organize the report body around the findings supported by the data rather than around every available data category.

Possible subjects include:

• Listening performance and growth
• Program or episode breadth and concentration
• Platform or device use
• Sharing and audience response
• Geographic distribution
• Follow or ownership activity
• Support, ministry links, and inserts

Whenever the export permits reliable classification, report:

• Total Live Listening
• Total On-Demand Listening
• Each as a percentage of Total Listening
• Change from the comparable prior period, when available

Where supported, identify the livestreams or ChannelStreams contributing to Live Listening and the Programs + Episodes contributing to On-Demand Listening.

Analyze whether listening is concentrated or distributed across the station's live and on-demand programming.

Always show Programs with Episodes when Episode information is available.

Do not speculate about why listeners chose Live or On-Demand content.

Include only the subjects that contribute meaningfully to the customer's performance story.

For each included subject:

• Present only the necessary supporting data.
• Provide no more than one concise interpretation following a table.
• Explain why the finding matters to leadership.
• Avoid repeating an observation already made elsewhere.

Report Top Programs or Top Episodes only when they reveal a meaningful content pattern. Do not characterize a program or episode as successful based solely on a very small count.

Report platform and device distribution only when attribution is sufficiently reliable and the distribution reveals something useful. When platform categories are reported, keep Website, Progressive Web App, Godcaster Mobile App, and Generic Godcaster Player distinct rather than combining them into a single Web or App category.

Do not create separate Device and Platform sections when one combined table or one concise observation communicates the useful finding.

Report Support or Insert activity when meaningful activity exists, when a material change occurred, or when the absence of activity supports an important practical opportunity. Do not create mandatory zero-value sections.

4. Executive Observations

Provide three to five concise leadership insights only when they add synthesis beyond the preceding tables and narrative.

Each observation should combine related evidence or explain a meaningful performance pattern. Do not use this section to repeat individual statistics.

If the report already communicates the principal insights clearly, omit this section.

5. Strategic Opportunities and Recommendations

Provide two to five prioritized recommendations supported by meaningful findings. Do not manufacture recommendations to meet a quota.

Each recommendation must include:

• Evidence
• Why it matters
• Practical next step
• Expected benefit

Recommendations should help the customer make fuller use of capabilities already available within its existing Godcaster Digital Station.

Do not recommend purchasing additional Godcaster products or services.

Favor constructive language such as:

• Build on
• Continue
• Strengthen
• Feature more deliberately
• Improve the transition
• Expand awareness
• Create a clearer pathway

Avoid generic digital-marketing advice that is not directly connected to a finding in the report.

6. Report Definitions

Include a concise customer-facing definitions section containing only metrics that actually appear in the report.

Define applicable terms in plain language:

• Total Engagement — All recorded Actions in the Digital Station during the reporting period. It reflects overall activity, not a count of individual listeners.
• Listening — Recorded playback activity, including new plays and continued playback.
• Play — A new recorded playback action.
• Share — A listener using the sharing feature to share content.
• Follow — A recorded action in which a listener follows station or program content through a supported podcast app or service.
• Download — A recorded download action when present in the export.
• PWA — Activity involving the station's Progressive Web App experience. PWA activity does not by itself confirm an installation.
• Station Support/Giving — A recorded action selecting the station-level support destination. It does not confirm a completed gift.
• Program Support/Giving — A recorded action selecting a program-level support destination. It does not confirm a completed gift.
• Live Listening — Listening associated with an identifiable live stream or live ChannelStream.
• On-Demand Listening — Listening to recorded programs or episodes available on demand.

Do not include definitions for metrics that do not appear in the report.

7. Data Confidence and Limitations

Include only limitations that materially affect interpretation of an important finding.

Do not repeat definitions or routine caveats already explained elsewhere.

If reliable listener journey analysis cannot be performed, state this once in this section rather than creating a separate Listener Behavior section.

Keep this section concise and proportional to the actual effect of the limitations.

Specific Exclusions

Unless the data establishes that they are materially useful, do not automatically create separate sections for:

• Audience Reach
• Digital Platform Performance
• Device Usage
• Support Activity
• Insert Performance
• Listener Behavior
• Every engagement category

Do not include:

• A line-by-line explanation of every KPI
• Repeated statements that a metric could not be evaluated
• Tables dominated by zero values or immaterial counts
• Technical data validation commentary that does not affect a conclusion
• Unsupported causes, motivations, forecasts, or assumptions
• Generic recommendations unrelated to the customer's measured performance

Editorial Model to Follow

Follow these editorial characteristics:

• Develop a small number of meaningful performance stories.
• Explain the relationship among metrics rather than presenting isolated statistics.
• Distinguish direct listening and response from system or distribution activity.
• Lead with accomplishments and encouraging patterns.
• Present the strongest opportunity with evidence, context, and a practical next step.
• Use limitations to protect accuracy, not to dominate the customer narrative.

Do not introduce app listening hours, sessions, deep-link visits, installations, conversion, retention, or car-connection analysis because those measurements are not supplied by the Godcaster Statistics export.

Final Validation

Before finalizing, verify that:

• Every number matches the uploaded Godcaster export or exports.
• Every percentage uses the correct denominator.
• Period comparisons use consistent definitions and methodology.
• Listening and system or distribution activity are clearly distinguished.
• Technical identifiers are not described as verified people or listeners.
• The strongest positive performance story receives the greatest emphasis.
• Material concerns are reported honestly and proportionately.
• Small, zero-value, unavailable, or technical metrics have not been overinterpreted.
• No section exists solely because data was available.
• Each major finding is discussed once rather than repeatedly.
• Every recommendation is specific, practical, and supported by evidence in the report.
• No Aptabase, previous-report, outside, or unsupported information has been introduced.
• The final report reads as a useful customer performance story, not an analytics inventory.

END COPY