More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    SaaS Cancellation Insight Kit

    1

    Analyze SaaS cancellation feedback into source-linked reason codes, verified tallies, and testable product hypotheses.

    $7

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    SaaS Cancellation Insight Kit

    SaaS Cancellation Insight Kit

    Example session with this skill installed

    Analyze the attached feedback.csv containing 10 cancellation events from our Pro plan. Use the standard taxonomy to identify why they left and provide a tally report.

    • Read your context and instructions
    • Compiled the saas cancellation insight
    • Generated the data export

    Processed 10 events from feedback.csv. Identified 8 eligible events for coding; 2 records were flagged for review (1 blank, 1 duplicate ID).

    Cancellation Insight Brief

    Scope and Coverage

    This analysis covers 10 total records provided for the "Pro plan" segment. 8 events contain unique, nonblank feedback eligible for semantic coding. The sample period is unspecified.

    Data Quality

    • Eligible Events: 8
    • Duplicate IDs: 1 (Event_ID: 9022)
    • Blank Feedback: 1 (Event_ID: 9045)
    • Coding Status: 100% of eligible events coded.

    Stated-Reason Themes

    • Price/Value: 5 of 8 events (62.5% of mentions). Primary reason for 3 events.
    • Missing Feature: 3 of 8 events (37.5% of mentions). Primary reason for 2 events.
    • Technical Issues: 1 of 8 events (12.5% of mentions).
    • Unclear: 2 events could not be assigned a primary reason due to conflicting statements.

    Evidence and Exceptions

    • Price/Value: "Too expensive for a solo founder" (ID: 9012); "Switching to a cheaper alternative" (ID: 9015).

    Contradiction: Event 9018 mentions "love the UI" but "too many bugs to stay," making the primary reason "Technical Issues" despite positive UI sentiment.

    Hypotheses to Test

    Hypothesis: A mid-tier pricing plan may retain solo founders currently citing high costs. (Linked to IDs: 9012, 9015, 9028).
    2. Hypothesis: The recent API downtime is the driver for technical complaints. (Linked to ID: 9018).

    Next steps

    1. Review data-review.json to resolve the duplicate ID for record 9022.
    2. Cross-reference the "Missing Feature" IDs (9014, 9021) with the Q3 roadmap.
    3. Run tally_cancellations.py to regenerate the ledger if more feedback is added.
    {
      "coverage": 0.8,
      "duplicate_conflict_counts": 1,
      "segment_tallies": {
        "Pro": 8
      },
      "source_hash": "a1b2c3d4"
    }
    

    feedback.csv

    CSV · data export

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    SaaS cancellation surveys often provide mountains of messy text but zero actionable direction. This skill transforms raw feedback into a structured decision brief by mapping specific event IDs to defensible stated reasons, verified sample counts, and testable product hypotheses. It stops the guessing game of why users leave by enforcing a rigorous coding process that links every insight back to a verbatim source quote.

    What it does

    • Evidence coding maps authorized feedback to a specific taxonomy while preserving exact, nonblank excerpts as evidence.
    • Reproducible tallying generates systematic counts of stated reasons and primary themes using a standalone Python validation script.
    • Data quality auditing identifies duplicate records, conflicting IDs, and blank responses to ensure the sample denominator is accurate.
    • Hypothesis generation builds testable product and business theories linked directly to specific event IDs and evidence snippets.
    • Segment analysis breaks down cancellation reasons by user-defined segments to highlight patterns across different customer types.

    How it works

    1. Source ingestion involves processing a UTF-8 CSV containing event IDs and feedback text, mapping headers to a canonical format.
    2. Semantic coding assigns reason codes and primary states to each eligible event, outputting a structured JSON ledger of evidence.
    3. Script execution runs a Python 3.10+ tally tool to validate ID integrity and generate five distinct analysis artifacts.
    4. Brief delivery produces a final markdown report covering scope, data quality, theme counts, and next-step evidence requests.

    Frameworks & tools

    The skill utilizes Python 3.10+ for local data processing and validation. It operates on standard CSV and JSON formats and does not require external API keys, database connections, or specialized SaaS billing integrations.

    Why this beats prompting it yourself

    Generic prompts often hallucinate churn rates or merge distinct events, leading to false precision. This skill uses a fixed taxonomy and script-verified arithmetic to ensure that every count is tied to a verifiable source-record index, preventing the AI from "summarizing away" contradictory evidence.

    Use cases

    • Product roadmap planning by identifying which missing features are cited as primary exit reasons.
    • Customer success audits to quantify the impact of price sensitivity versus technical friction in a recent cohort.
    • Investor reporting by providing a defensible, source-linked breakdown of cancellation themes for due diligence.
    • Feedback cleaning to deduplicate messy exports from survey tools before performing manual review.

    Known limitations

    Does not connect to live billing systems or perform automated win-back outreach. Requires Python 3.10+ for verified tallies. Cannot infer true churn causes or financial impact (ARR) without external data.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Product Management.

    Requires Python 3.10+ for tally verification. Does not support live billing API connections. Limited to analyzing supplied text feedback; cannot verify actual subscription status.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 19 days ago

    Software
    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Listed19 days ago

    What's inside

    Frequently Asked Questions

    The kit provides a structured analysis of cancellation surveys by mapping raw feedback to specific reason codes and evidence snippets, backed by a Python script that ensures your churn tallies are mathematically accurate rather than just AI-guessed.

    You will receive the core AI analysis logic, a custom Python 3.10+ validation script for data integrity, and a structured markdown template for your final decision briefs.

    This skill requires a standard UTF-8 CSV export containing your survey data; it does not need direct access to your billing system, Stripe, or CRM, ensuring your sensitive financial data remains private.

    Yes, the skill is designed to process user-defined segments, allowing you to compare cancellation reasons between different tiers (e.g., Free vs. Enterprise) or specific customer cohorts.

    While the AI performs the semantic coding, a Python 3.10+ environment is required to run the included validation tool which generates the final reproducible counts and audit artifacts.

    The kit identifies and flags duplicate records, conflicting event IDs, and blank entries to ensure your analysis denominator is clean and your insights are based on high-quality data.

    New in Product Management