# 🧠 Authority Article Strategy | Element | Summary | |---|---| | **Topic** | B2B SaaS customer retention | | **Audience** | SaaS founders and customer success (CS) leaders | | **Goal** | Demonstrate expertise; argue that retention strategy should start with diagnosis | | **Core point of view (POV)** | Teams jump to tactics (health scores, QBRs, discounts, win-back campaigns) before establishing *why* customers leave. Churn has distinct root causes, and each needs a different response. | | **Strongest authority opportunity** | A **cause-first diagnostic framework**. Most retention content is tactic lists, so a clear model for classifying churn before acting gives readers something to reuse. | | **Format** | ~1,800 words, analytical and practical tone, includes a diagnostic framework | **Verified vs. assumed** | Category | Status | |---|---| | Your POV (tactics-before-diagnosis pattern) | ✅ Provided by you | | Five candidate churn causes (fit, onboarding, product value, pricing, circumstances) | ✅ Provided by you | | Firsthand experience, client examples, or metrics | ❌ **Not provided.** Add only what you can substantiate. | | Diagnostic signals and process in this plan | ⚠️ Proposed reasoning, presented as hypotheses to validate, not established facts | --- # 💡 5 Article Angles | # | Approach | Working Title | Core Move | Risk | |---|---|---|---|---| | 1 | **Framework-based** | *The Five Causes of SaaS Churn: A Diagnostic Framework Before You Touch a Tactic* | Introduce a five-cause model with signals and matched responses | Can read as a listicle without a strong POV | | 2 | **Contrarian** | *Stop Building Retention Playbooks* | Argue that playbooks are premature until churn is diagnosed | Overstates the case; playbooks do have value | | 3 | **Analytical** | *"Why Did They Churn?" Is the Wrong Question (Until You Segment)* | Show that aggregate churn hides multiple distinct problems | Needs data or a credible illustration | | 4 | **Myth-busting** | *3 Retention Myths That Send Teams to the Wrong Fix* | Debunk "more touchpoints," "discounts save accounts," and "exit surveys tell the truth" | Each myth needs evidence or careful reasoning | | 5 | **Practical** | *A 4-Step Churn Autopsy Any CS Team Can Run This Quarter* | Step-by-step diagnostic process | Less distinctive POV; may feel like a how-to | --- # 🎯 Recommended Angle **Angle 1 (framework-based), with Angle 5's process as the delivery mechanism and a light contrarian frame in the intro.** | Why it fits | Explanation | |---|---| | **Audience** | Founders want to know where to spend limited resources. CS leaders want a shared language to explain churn to leadership. A framework serves both. | | **Goal** | Diagnosis-first is your thesis. A named model makes it concrete and memorable. | | **Authority** | Frameworks signal structured thinking without unsupported credentials. They are also easy to cite and share. | | **Feasibility** | It works from reasoning and clearly labeled examples, which suits a brief with no supplied data. | **Recommended working title:** *Diagnose Before You Retain: A Five-Cause Framework for Understanding B2B SaaS Churn* **Framework name (original):** **The Churn Diagnosis Lens**. Five causes, four steps, one rule: *classify before you treat.* --- # 🧭 Detailed Article Structure | # | Section | Words | Purpose | |---|---|---|---| | 0 | **Headline + subhead** | n/a | Promise diagnosis over tactics | | 1 | **Introduction: The Prescription Problem** | ~200 | Hook, thesis, roadmap | | 2 | **Why Tactics-First Retention Misfires** | ~250 | Build the case for diagnosis | | 3 | **The Five Causes of Churn** (framework) | ~400 | Teach the model | | 4 | **How to Diagnose: A Four-Step Churn Autopsy** | ~400 | Make it operational | | 5 | **Matching Cause to Response** | ~250 | Connect diagnosis to action | | 6 | **A Worked Example (Illustrative)** | ~150 | Show the framework in use | | 7 | **Conclusion: Start With the Question** | ~150 | Restate POV, give next step | | | **Total** | **~1,800** | | **Subheadings** - **1.** *(no subhead; the hook leads)* - **2.** "Same Symptom, Different Disease" · "The Cost of Guessing" - **3.** "Fit · Onboarding · Value · Price · Circumstance" (one short subhead per cause) - **4.** "Step 1: Segment · Step 2: Time-map · Step 3: Triangulate · Step 4: Assign and Grade Confidence" - **5.** "Diagnosis → Intervention Table" · "When It's Not Your Problem to Fix" - **6.** "Illustrative Scenario" - **7.** "Where to Start Monday" --- # 🧩 Section Arguments & Key Points ## 1. Introduction: The Prescription Problem (~200 words) | Element | Detail | |---|---| | **Argument** | Teams treat churn like a headache and reach for the same pill (more check-ins, discounts, a new health score) without knowing the cause. | | **Support** | Medical analogy (prescribing before diagnosing); the observation that different churn reasons call for different owners (sales, product, CS, finance). | | **Example** | A generic before/after: "A team adds quarterly business reviews (QBRs), but accounts keep leaving. The QBRs were never the problem." Label it as a composite or hypothetical unless real. | | **Takeaway** | Retention improves faster when the team first asks what kind of churn it has. | ## 2. Why Tactics-First Retention Misfires (~250 words) | Element | Detail | |---|---| | **Argument** | Tactics are tied to one cause, but teams apply them to all churn. | | **Support points** | (a) A tactic works only on the cause it addresses. (b) Aggregate churn rate hides mixed causes. (c) Tactics are visible and easy to approve; diagnosis is less exciting but cheaper. (d) Misdiagnosis wastes effort and can mask the true problem. | | **Teaching device** | **"Same symptom, different disease"** table with the symptom "customer cancels" and five possible causes. | | **Takeaway** | A tactic that helps one segment can be wasted, or counterproductive, on another. | ## 3. The Five Causes of Churn (~400 words) **Core framework: The Churn Diagnosis Lens** | Cause | Diagnostic Question | Illustrative Signals *(hypotheses to validate)* | Common Misdiagnosis | Likely Owner | |---|---|---|---|---| | **1. Fit** | Should this customer have been sold in the first place? | Early churn; use case outside product scope; mismatch with ideal customer profile (ICP); one channel or segment over-represented in churn | "They didn't engage enough" | Sales, Marketing, ICP definition | | **2. Onboarding** | Did they reach first value, and how fast? | Never completed setup; key feature never used; stalls at a specific onboarding step | "Product doesn't deliver value" | CS, Implementation, Product-led onboarding | | **3. Product Value** | Did they get value and then stop, or never see it? | Engaged early then declined; workarounds; competitor features cited; value not tied to their goals | "Pricing objection" | Product, CS | | **4. Price** | Is the problem cost itself, or cost relative to perceived value? | Objection at renewal; downgrade requests; heavy users still price-sensitive vs. light users citing cost | "Discount will save it" | Pricing/Finance, Sales, Packaging | | **5. Circumstance** | Did something change on their side that we could not have prevented? | Acquisition, layoff, budget freeze, champion left, company shutdown | "We failed them" | Mostly outside control; focus on win-back and champion mapping | **Distinctions to introduce** - **Reason given vs. reason true.** What customers say in cancellation surveys may differ from what usage data shows. Present this as a caution to test, not an established finding. - **Preventable vs. non-preventable churn.** Circumstance-driven churn should be tracked separately so it does not distort the team's sense of performance. - **Primary vs. contributing cause.** Many accounts have more than one; assign one primary cause for action. **Takeaway:** Five causes, five different questions, five different owners. ## 4. How to Diagnose: A Four-Step Churn Autopsy (~400 words) | Step | Action | Question It Answers | Output | |---|---|---|---| | **1. Segment** | Group churned accounts by plan, size, industry, acquisition channel, tenure | Is churn concentrated or evenly spread? | Segment map | | **2. Time-map** | Place churn on the lifecycle (pre-activation, first 90 days, mid-term, renewal) | *When* does it happen? | Timing pattern. Hypothesis: early churn points toward fit or onboarding; later churn toward value, price, or circumstance | | **3. Triangulate** | Compare usage data, support history, sales notes, and customer conversation | Does stated reason match behavior? | Evidence table per account or cluster | | **4. Assign and grade** | Choose a primary cause and rate confidence (high / medium / low) | Which cause, and how sure are we? | Diagnosis log | **Key points** - Confidence grading keeps the team honest and shows where more evidence is needed. - Start small: 15 to 20 churned accounts (a suggested starting sample, not a validated threshold) is enough to see patterns. - Diagnosis is not a one-time project; keep a rolling log. **Takeaway:** A lightweight, repeatable process beats a big analysis nobody finishes. ## 5. Matching Cause to Response (~250 words) | Cause | Typical Response Category | Tactics That May **Not** Help | |---|---|---| | Fit | Tighten ICP, qualification, and sales handoff | More CS touchpoints | | Onboarding | Redesign time-to-first-value path, implementation support | Renewal discounts | | Value | Product roadmap, adoption programs tied to outcomes | Generic check-in emails | | Price | Packaging, tiering, value communication; discount only where diagnosis supports it | Broad discounting | | Circumstance | Champion mapping, flexible pause options, win-back | Escalating outreach | **Argument:** Retention is often cross-functional. Sales, product, CS, and finance each own part of it, and diagnosis shows which. **Takeaway:** Diagnosis tells you what to do *and who has to do it*. ## 6. A Worked Example (~150 words) - Use a **clearly labeled hypothetical**, or a real anonymized case if you have one. - Show: symptom (rising churn) → segmentation → timing → triangulation → primary cause → matched response. - Include one moment where the tempting tactic would have been wrong. ## 7. Conclusion: Start With the Question (~150 words) | Element | Detail | |---|---| | **Restate** | Retention improves when teams ask *what kind of churn is this?* before *what should we do?* | | **Recap** | Five causes, four steps, one rule: classify before you treat. | | **Next step** | "Pull your last 15 to 20 churned accounts and assign a primary cause to each." | | **Close** | An open question or invitation to compare notes, not a hard sell. | --- # ✍️ Hooks, Openers & Transitions ## Hook options | Style | Example | |---|---| | **Analogy** | "No doctor prescribes before diagnosing. Most retention programs do." | | **Contrast** | "Two customers cancel on the same day. One never finished setup. The other outgrew your pricing. The same win-back email won't work on both." | | **Question** | "When a customer churns, do you know what *kind* of churn it was?" | Avoid: "Churn is killing your business" and other fear-based openers. ## Section openers | Section | Opener | |---|---| | 2 | "Every retention tactic is quietly a bet on a cause." | | 3 | "Churn is one outcome with at least five different origins." | | 4 | "You do not need a data science team to start diagnosing." | | 5 | "Diagnosis only matters if it changes what you do next." | | 7 | "The next time an account churns, ask what kind of churn it was before asking what to do." | ## Transitions | From → To | Transition Line | |---|---| | 1 → 2 | "To see why this keeps happening, look at what tactics actually assume." | | 2 → 3 | "If tactics assume a cause, the first job is naming the possible causes." | | 3 → 4 | "Knowing the five causes helps only if you can tell which one you're facing." | | 4 → 5 | "Once you have a diagnosis, the response nearly chooses itself." | | 5 → 6 | "Here is how that plays out in a scenario." | | 6 → 7 | "The pattern is simple, even when the data is messy." | --- # ✅ Credibility & Evidence Opportunities | Evidence Type | Where to Use | What You Would Need to Supply | Status | |---|---|---|---| | **Personal experience** | Intro, worked example | A real pattern you've seen of teams skipping diagnosis. Include only what you have directly observed. | ❓ Needs your input | | **Client or team anecdote** | Sections 2, 6 | An anonymized story with permission | ❓ Needs your input | | **Own data** | Section 4 | Churn segmentation or timing patterns from your own work | ❓ Optional; strong if available | | **Published research** | Sections 2, 3 | Credible studies on churn causes, cohort-based analysis, or reason-given vs. behavior gaps. Cite specifically. | 🔎 To be sourced; none included in this plan | | **Expert perspective** | Section 3 or 5 | Quote from a CS leader, product leader, or pricing specialist | 🔎 Optional | | **Composite/hypothetical** | Sections 1, 6 | Clearly labeled scenario | ✅ Acceptable if labeled | **Rule:** If you cannot verify it, present it as reasoning, a hypothesis, or an illustration. --- # ⚠️ Weak Claims to Avoid | Weak or Risky Claim | Problem | Better Approach | |---|---|---| | "Most churn is caused by poor onboarding" | Unsupported statistic | "Onboarding is one of several causes worth ruling in or out" | | "Acquiring customers costs 5x more than retaining them" | Widely repeated, often unsourced | Use only with a verifiable source, or skip | | "Exit surveys are useless" | Overgeneralized | "Stated reasons should be checked against behavior" | | "Discounts never work" | Absolute | "Discounts help only when price is the diagnosed cause" | | "Our framework is proven / definitive" | Unsupported authority | "A framework I use to structure the diagnosis" (only if true) | | "Early churn always means poor fit" | Overclaims | "Early churn often points to fit or onboarding, so test both" | | "Every SaaS company should…" | Ignores stage and model differences | Qualify by segment, ACV, or motion | | Generic tips ("communicate more," "focus on customer success") | Adds nothing | Tie every tip to a specific cause | | Invented case studies or numbers | Credibility risk | Label as hypothetical or omit | | Buzzword stacking ("holistic, synergistic retention ecosystem") | Dilutes clarity | Plain language | --- # 📌 Writing Brief | Field | Direction | |---|---| | **Working title** | Diagnose Before You Retain: A Five-Cause Framework for Understanding B2B SaaS Churn | | **Audience** | SaaS founders and CS leaders | | **Reader's problem** | Churn is high or unclear, and they are being asked for retention initiatives | | **Thesis** | Effective retention starts with classifying churn causes, because fit, onboarding, value, price, and circumstance each need different responses | | **Tone** | Analytical, practical, direct, confident but calibrated | | **Length** | ~1,800 words | | **Must include** | Five-cause framework table, four-step autopsy, cause-to-response mapping, one labeled example | | **Must avoid** | Invented stats, absolute claims, unlabeled hypotheticals, generic tactic lists | | **Voice** | First person only where it reflects your real experience | | **CTA** | Audit your last 15 to 20 churned accounts using the lens | | **Formatting** | Short paragraphs, 2 to 3 tables (framework, process, response mapping), clear subheads | --- # 🚀 Article Development Checklist **Phase 1: Foundation** - [ ] Confirm the thesis in one sentence - [ ] Decide which real experience you can honestly draw on - [ ] Choose the worked example (real anonymized or labeled hypothetical) **Phase 2: Evidence** - [ ] Decide whether to cite external research; if so, find and verify sources - [ ] Note any statements to soften into "often" or "may" - [ ] Gather any personal data or anecdotes you can legitimately use **Phase 3: Drafting** - [ ] Draft hook and intro (~200 words) - [ ] Draft framework section and confirm each cause has question, signals, and response - [ ] Draft four-step process with concrete outputs - [ ] Draft response mapping table - [ ] Draft example and conclusion **Phase 4: Quality Control** - [ ] Check every claim against the "Weak Claims" table - [ ] Confirm hypothetical scenarios are labeled - [ ] Remove unsupported superlatives and stats - [ ] Verify word count (~1,800) and section balance - [ ] Check that each section has a takeaway and transition **Phase 5: Final Review** - [ ] Read as a skeptical CS leader: is anything obvious, generic, or unproven? - [ ] Confirm the reader can act on the framework this week - [ ] Finalize headline and subhead If you'd like, I can draft the full 1,800-word article from this outline, or write alternative headlines and intro variations.