{"site":{"name":"Koji","description":"AI-native customer research platform that helps teams conduct, analyze, and synthesize customer interviews at scale.","url":"https://www.koji.so","contentTypes":["blog","documentation"],"lastUpdated":"2026-05-30T08:02:16.114Z"},"content":[{"type":"documentation","id":"c7e97222-1fad-4b07-a7a5-d8d3ca4958d4","slug":"dormant-user-reactivation-research","title":"Dormant User Reactivation Research: How to Interview Inactive (Still-Subscribed) Users Before They Churn","url":"https://www.koji.so/docs/dormant-user-reactivation-research","summary":"How to interview still-subscribed but inactive SaaS users to prevent churn before it happens. Distinct from churned/win-back research. Covers the 8 behavioral leading indicators of dormancy, why dormant users are harder to recruit than churned ones, 20 dormant-user-specific interview questions, recruitment tactics that bypass standard panels, a recoverability × root-cause × habit-state analysis framework, and how AI-moderated async interviews like Koji unlock 200+ interviews per week vs. the 5 a traditional researcher can manage.","content":"# Dormant User Reactivation Research: How to Interview Inactive (Still-Subscribed) Users Before They Churn\n\n**The answer up front:** Dormant users are customers who are still paying but have stopped using your product. They are *not* churned users — they're churn-in-waiting. Research shows **70–80% of customers who eventually cancel show clear behavioral warning signs at least 30 days before they churn** ([US Tech Automations, 2026](https://ustechautomations.com/resources/blog/saas-usage-analytics-automation-detect-churn-early)), with sustained 30%+ engagement drops being the most reliable leading indicator. Interviewing dormant users — *before* they cancel — is one of the highest-ROI research programs a SaaS team can run: ProfitWell found **reactivated customers have 23% higher ARPU than newly acquired ones**, and Bain & Company's classic data shows a **5% retention lift can grow profits 25–95%** ([Reichheld, HBR](https://hbr.org/2014/10/the-value-of-keeping-the-right-customers)). The catch: dormant users are *harder* to interview than churned ones, and traditional research panels can't reach them. AI-moderated async interview platforms like Koji are built for exactly this problem — async completion rates of ~89% versus ~60% for scheduled video calls turn an impossible recruit into a routine cohort study.\n\n---\n\n## Dormant ≠ churned: the most important distinction in retention research\n\nMost \"reactivation\" content on the internet conflates two distinct populations:\n\n| | **Churned users** | **Dormant users** |\n|---|---|---|\n| Billing | Cancelled, no longer paying | Still paying, still have account access |\n| Self-identity | \"Former customer\" | They've forgotten they're a customer |\n| Motivation to talk | Active grievance to share | None — no current pain |\n| Recruitability | Reachable via email; sales/CS teams can call | Reachable only via your CRM; ignore most outreach |\n| Research goal | Understand *why* they left | Understand *what stopped them* before they leave |\n| Existing Koji guide | [Churned Customer Interviews](/docs/churned-customer-interviews) and [Win-Back Customer Interviews](/docs/win-back-customer-interviews) | *This guide* |\n\nDormant users are the iceberg under churn. By the time you see the cancellation, the engagement decay began 30–90 days earlier. Research on dormant users is research on **the leading indicator of revenue loss** — the only kind of retention research that lets you intervene before the loss happens.\n\n> \"We have a couple of customers that have been customers for 6 months or more that haven't done anything in our software except pay us every month … they're a massive churn threat. They WILL realize they're paying you for something they're not using some day … and they will leave.\"\n> — Lincoln Murphy, [*When Customers Go Dark: Customer Success to Fight the Zombies*](https://sixteenventures.com/saas-customer-success-zombie-customers)\n\n---\n\n## The economics: why dormant users are worth interviewing\n\n- **Acquiring a new customer can cost 5–25× more than retaining an existing one**, and a **5% retention lift can boost profits 25–95%** — Frederick Reichheld / Bain & Company, [HBR](https://hbr.org/2014/10/the-value-of-keeping-the-right-customers).\n- **Reactivated customers carry 23% higher ARPU than newly acquired ones** — ProfitWell, cited in [Monetizely](https://www.getmonetizely.com/articles/understanding-reactivation-rate-a-critical-metric-for-saas-growth).\n- **Personalized reactivation campaigns succeed ~45% of the time vs. ~12% for generic outreach** — Salesforce, cited in [Monetizely](https://www.getmonetizely.com/articles/understanding-reactivation-rate-a-critical-metric-for-saas-growth). Personalization that works comes from *the language dormant users actually use* — which only research surfaces.\n- **Median Net Revenue Retention sits at 110% across 800+ SaaS companies; best-in-class above 115%** ([OpenView/High Alpha 2024 SaaS Benchmarks](https://www.highalpha.com/saas-benchmarks/2024)). The gap between median and best-in-class is largely a dormant-user intervention story.\n\nThe math: a team running 200 dormant-user interviews uncovers verbatim language for a reactivation email sequence that lifts open rates from a typical 18% to 31% ([Aampe](https://aampe.com/blog/crm-increasing-ctr-and-solving-the-inactive-user-problem)). One research sprint funds itself many times over.\n\n---\n\n## How dormancy precedes churn: the 8 leading indicators\n\nYou don't need ML to predict churn — you need to watch for behavioral patterns. From the SaaS retention literature ([US Tech Automations](https://ustechautomations.com/resources/blog/saas-usage-analytics-automation-detect-churn-early), [Automaiva](https://automaiva.com/saas-churn-prevention-automation-early-warning-2026/), [ChurnZero](https://churnzero.com/blog/customer-champion-playbook/), [Sixteen Ventures](https://sixteenventures.com/saas-customer-success-zombie-customers)):\n\n1. **Login frequency drop (50%+ from baseline)** — the single most reliable signal; appears 30–45 days before cancellation.\n2. **Feature-usage decay (30%+ MoM in core features)** — correlates with churn within 60 days.\n3. **Seat-utilization decline (B2B)** — customers at 25% capacity churn **5× more** than those at 80%. Seat removal is almost always pre-cancellation behavior.\n4. **Champion / primary-contact role change** — when a customer champion leaves, **there's a 51% chance the account churns within 12 months. For executive changes the figure is 65%** ([Sturdy AI via ChurnZero](https://churnzero.com/blog/customer-champion-playbook/)).\n5. **Support-ticket pattern shift** — both spikes (frustration) and sudden silence (disengagement). QBR cancellations 6–8 weeks before non-renewal are a red flag.\n6. **NPS / CSAT drops** of 20+ points month-over-month.\n7. **Data export or billing tier-gaming** — extracting their data or downgrading is mid-stage exit behavior.\n8. **Annual-to-monthly downgrade** — ProfitWell flags this as the highest-signal billing event.\n\n> **Rule of thumb:** 80% of SaaS churn is predictable from 3–5 of these signals ([Automaiva, 2026](https://automaiva.com/saas-churn-prevention-automation-early-warning-2026/)). Use them to define your dormant-user cohort *before* you recruit for interviews.\n\n---\n\n## Why dormant users are *harder* to interview than churned ones\n\nCounterintuitive but true. Four reasons:\n\n### 1. Embarrassment protection\nDormant users are paying for something they don't use. Admitting that out loud creates cognitive dissonance. Lincoln Murphy observes that dormant customers *\"will tell others not to use you, but it'll be for made-up negative reasons so they don't have to admit they paid for something they didn't use.\"*\n\n### 2. No identity as \"customer\"\nUnlike a churned user who actively rejected the product, dormant users have *forgotten* they're customers. Your \"we miss you\" email lands as cold outreach, not a re-engagement.\n\n### 3. No active pain\nChurned users carry a complaint they want to share. Dormant users have no current need pushing them to reply. As Andrew Chen frames it in [*The Law of Shitty Clickthroughs*](https://andrewchen.com/the-law-of-shitty-clickthroughs/), engagement decays for predictable reasons:\n\n> \"Humans seek novelty yet are pattern-recognition machines. Your initial marketing strategy will work quite well as your users try it for the first time, but afterwards, they learn to filter your marketing efforts out.\"\n\n### 4. Panels can't reach them\nUser Interviews, Respondent, and dscout recruit from panels *outside* your CRM. Your dormant users are *inside* it. You have to go through your own data — and the standard 30-minute Zoom invite is dead on arrival for someone who hasn't logged in in 90 days.\n\nThe solution is to lower the friction. Async AI-moderated interviews see **~89% completion vs ~60% for scheduled video** ([Hirevire benchmark](https://hirevire.com/articles/best-rated-asynchronous-video-interview-platforms)). For dormant users — already low-motivation — that gap widens further.\n\n---\n\n## How to recruit dormant users (when standard tactics don't work)\n\n### Define the cohort behaviorally\nPull users with:\n- **60+ days since last meaningful login** AND\n- **active billing status** AND (for B2B) **seat utilization < 40%**\n\nAdd product-specific signals — feature usage decay, downgraded plan, support-ticket silence after prior activity.\n\n### Lower the friction\n- Send personalized async interview links — text-mode AI moderation lets a dormant user respond in 10 minutes from their phone instead of carving out a half-hour calendar block.\n- Pop a one-question in-app prompt the next time they *do* log in: \"Quick question — we noticed you haven't been around. Got 10 minutes to tell us why? We'll credit your next bill.\"\n- Skip the 30-minute Zoom invite entirely.\n\n### Use subject lines that perform on dormant audiences\nFrom [User Interviews' recruitment research](https://www.userinterviews.com/blog/recruiting-user-research-participants-by-email):\n- **\"Help improve [product]\"** — clear ask, clear benefit\n- **Personalize the title or use-case** (\"We want to talk to [job title]\")\n- **Lead with the dollar amount of the incentive** when offered\n- **41 characters / 7 words ideal** (70 char Gmail cutoff)\n\nPersonalized subject lines move opens from a baseline ~18% to ~31% ([Aampe](https://aampe.com/blog/crm-increasing-ctr-and-solving-the-inactive-user-problem)).\n\n### Match incentives to the user\nFor dormant *paying* users, a **credit on the current subscription** often outperforms cash — it's higher-relevance and signals the company values their feedback over a transaction.\n\n---\n\n## The 20 interview questions that actually surface what dormancy is hiding\n\nGroup them into four arcs: origin, drift, current reality, and re-engagement.\n\n**Origin (the promise they bought into)**\n1. Take me back to when you signed up — what were you hoping [product] would change for you?\n2. When was the last time you remember opening [product]? What were you trying to do?\n3. Did you ever recommend [product] to anyone? When and why (or why not)?\n\n**Drift (what broke the habit)**\n4. Walk me through a typical week — when this kind of problem comes up, what do you do now?\n5. Has anything changed in your role, team, or workflow in the last 60–90 days?\n6. (B2B) Is the person who championed [product] internally still in the same role?\n7. Tell me about the last time [product] surprised or disappointed you.\n8. Have you started using a different tool — or just stopped doing this kind of task entirely?\n\n**Current reality (where they are now)**\n9. What does the work that [product] *was* helping you with look like today?\n10. On a scale of 1–10, how important is solving [job-to-be-done] to you right now?\n11. If [product] disappeared tomorrow, what would you actually miss?\n12. What's stopping you from cancelling today? (Reveals latent loyalty vs. pure inertia.)\n13. Honestly, if your card got declined tomorrow — would you re-enter the details?\n\n**Re-engagement (what could bring them back)**\n14. What would have to be true for you to log back in this week?\n15. If you got an email tomorrow saying we'd built [X], would that pull you back in?\n16. Is there a feature you forgot existed that you've just remembered we have?\n17. What's the smallest thing we could do that would make you say \"OK, I'll give it another shot\"?\n18. Who else on your team should be using [product] but isn't? Why?\n19. When you opened this interview, what did you assume we wanted to hear?\n20. Anything you wish we'd asked but didn't?\n\nTwo underpinning techniques worth naming:\n\n- **Latent jobs probing** ([JTBD framework](https://www.userinterviews.com/ux-research-field-guide-chapter/jobs-to-be-done-jtbd-framework)) — what are they doing *instead* of using your product? That alternative is your real competitor.\n- **Four Forces mapping** (from JTBD switch interviews) — the push of their current alternative, the pull of your original promise, the anxiety of restarting, the habit of *not* using. See [Switch Interviews: The JTBD Method](/docs/switch-interviews-jtbd-method) for the parent technique.\n\n> \"You don't wake up in the morning and say: 'Hmm, how should I make toast today?' The fact that you have your toaster right in front of you means you're going to use it with little or no conscious thought.\"\n> — Nir Eyal, [*Hooked*](https://clevertap.com/blog/the-core-tenets-of-user-retention/)\n\nDormancy is a broken habit loop. Your job in the interview is to find which link broke — trigger, action, reward, or investment.\n\n---\n\n## An analysis framework that turns interviews into reactivation playbooks\n\nThree orthogonal axes for segmenting dormant interview responses:\n\n**Axis A — Recoverability**\n- **Recoverable** — life or workflow change, forgotten value, broken habit\n- **Lost cause** — solved by a competitor, problem no longer exists, organizational pivot\n\n**Axis B — Root cause type**\n- **Latent value seekers** — still have the JTBD, lost the trigger\n- **Solved-by-other-tool** — switched (often unconsciously) to an alternative\n- **Life-change** — role change, project ended, champion left, budget freeze\n\n**Axis C — Habit state**\n- **Habit-broken** — formed habit, external disruption killed it (recoverable with trigger restoration)\n- **Never-formed-habit** — onboarded but never built a routine (requires re-onboarding, not reactivation)\n\nCross-tab these and you get distinct playbooks:\n\n- **Recoverable × Habit-broken × Life-change** → personalized re-onboard sequence with new-context examples\n- **Recoverable × Never-formed-habit × Latent value seekers** → guided onboarding redo, not a \"we miss you\" email\n- **Lost cause × Solved-by-other-tool** → graceful pause-or-cancel flow that captures honest exit reasons\n\nFor more on segmentation logic, see [Behavioral Segmentation Guide](/docs/behavioral-segmentation-guide).\n\n---\n\n## Why AI-moderated interviews are uniquely good for dormant-user research\n\n> \"Acquisition is the weakest growth lever.\"\n> — Patrick Campbell, founder of ProfitWell (now Paddle), [Business of Software](https://businessofsoftware.org/talks/pricing-retention-and-growth-strategies/)\n\nPatrick Campbell's analysis of 23,400+ SaaS companies showed monetization and retention deliver substantially higher revenue impact than acquisition. Dormant-user research is the engine that turns that thesis into product changes — and AI-moderated interviewing is what makes it operationally possible:\n\n- **Response-rate uplift.** Async AI interviews see ~89% completion vs. ~60% for scheduled video. For dormant users — already low-motivation — the gap widens.\n- **Scale.** Run 200+ dormant-user interviews in a week instead of 5 with a researcher. Dormancy research moves from sample to near-census.\n- **Hybrid question types.** Koji's [six structured types](/docs/structured-questions-guide) (open_ended, scale, single_choice, multiple_choice, ranking, yes_no) let you quantify forgotten value (`scale`), surface dormancy-reason distribution (`multiple_choice`), and capture verbatim narrative (`open_ended`) in the *same* session.\n- **Lower disclosure friction.** Dormant users feel embarrassed talking to a person. They open up to a non-judgmental AI text interview at materially higher rates — the same dynamic documented in [HCI research on chatbot disclosure](https://dl.acm.org/doi/10.1145/3544548.3581384).\n- **Days-not-weeks pipeline.** Findings feed directly into reactivation email sequences using the verbatim language dormant users used, plus product priorities (the \"forgotten features\" they wished they'd known about).\n\nA complete loop:\n\n1. Behavioral signal triggers a dormant cohort.\n2. Personalized interview link goes out in-app and via email.\n3. Koji's AI moderator runs 200 interviews in a week.\n4. Automatic theming clusters reasons into your recoverability × root-cause × habit-state matrix.\n5. Reactivation email sequences and product priorities ship within the same sprint.\n\nSee also: [Continuous Discovery: How to Run Weekly Customer Interviews Without Burning Out](/docs/continuous-discovery-user-research).\n\n---\n\n## Frequently asked questions\n\n### What's the difference between a dormant user and a churned user — and why does it matter for research?\nDormant users are still subscribed but have stopped using the product. Churned users have cancelled. The research goals differ: dormant-user research aims to *prevent* the loss; churn research aims to *understand and prevent recurrence*. The populations also recruit differently — dormant users are harder to reach because they have no active grievance and have forgotten they're customers.\n\n### How long without activity should we wait before classifying a user as dormant?\nMost SaaS teams use **60+ days without meaningful login** combined with active billing. Adjust by product cadence: a tax-prep tool might define dormancy as \"missed a quarter,\" while a daily-use analytics tool might use 14 days. The right threshold is where your data shows engagement decay starts predicting churn 30–45 days out.\n\n### Won't interviewing dormant users just remind them to cancel?\nThis is the most common objection — and the data doesn't support it. Personalized reactivation outreach **succeeds ~45% of the time vs. ~12% for generic** ([Salesforce, via Monetizely](https://www.getmonetizely.com/articles/understanding-reactivation-rate-a-critical-metric-for-saas-growth)). Dormant users are going to leave anyway; the interview is your only chance to surface why and intervene. Lincoln Murphy's \"zombie customer\" framing is explicit: silence is *worse* than the interview.\n\n### What incentive works best for dormant users when money isn't the issue?\nFor dormant *paying* users, a **credit on their current subscription** often outperforms cash — it's higher-relevance and reframes the relationship. For B2B power users, **early access to a roadmap item they care about** can outperform monetary incentives. Standard cash incentives work but punch below their weight for this cohort.\n\n### How do I recruit dormant users when traditional UX research panels can't reach them?\nYou can't — panels recruit *outside* your CRM. You must use your own data: behavioral cohort + personalized email + in-app prompt on next login + async AI-moderated interview links. See [Research Participant Incentives](/docs/research-participant-incentives) and [Research Screener Questions](/docs/research-screener-questions).\n\n### Should I run reactivation campaigns first or research first?\nResearch first. Generic reactivation campaigns succeed at ~12%; personalized ones at ~45%. Personalization that works requires the verbatim language and JTBD framing only interviews surface. Running campaigns first burns your one chance to re-engage on generic copy.\n\n### What's the right sample size for dormant user research?\nAim for **40–80 dormant-user interviews across 2–3 behavioral cohorts** (e.g., 30-day dormant, 90-day dormant, 180-day dormant). Theme saturation typically hits at 20–25 per cohort. With async AI interviews this is a 1–2 week sprint; with traditional moderated research it's a quarter. See [Data Saturation in Qualitative Research](/docs/data-saturation-qualitative-research).\n\n---\n\n## Related Resources\n\n- [Structured Questions Guide](/docs/structured-questions-guide) — Koji's six response types and how to mix scale + multiple-choice + open-ended for dormant-user studies\n- [Churned Customer Interviews: How to Talk to Users Who Left (and Win Them Back)](/docs/churned-customer-interviews)\n- [Win-Back Customer Interviews: How to Reactivate Lapsed and Cancelled Customers](/docs/win-back-customer-interviews)\n- [Customer Retention Research: The Complete 2026 Playbook](/docs/customer-retention-research)\n- [Switch Interviews: The JTBD Method for Understanding Why Customers Buy (and Leave)](/docs/switch-interviews-jtbd-method)\n- [Continuous Discovery: How to Run Weekly Customer Interviews Without Burning Out](/docs/continuous-discovery-user-research)\n- [Customer Health Score: How to Build a CHS Model for SaaS](/docs/customer-health-score-saas-guide)\n- [The B2B SaaS Retention Interview Template](/docs/b2b-saas-retention-interview-template)\n\n---\n\n*Dormant users are the cohort retention dashboards miss until it's too late. The teams turning them into a renewable insight pipeline aren't running more research — they're running async, AI-moderated interviews that meet dormant users where they are. Koji is the platform built for that loop.*","category":"Use Cases","lastModified":"2026-05-30T03:29:54.834308+00:00","metaTitle":"Dormant User Reactivation Research: How to Interview Still-Subscribed Inactive Users","metaDescription":"Interview still-subscribed inactive SaaS users before they churn. 8 leading dormancy indicators, why dormant users are harder than churned ones, 20 interview questions, recruitment tactics, and an analysis framework. With data from ProfitWell, Bain/HBR, ChurnZero, OpenView.","keywords":["dormant user reactivation","inactive user research","user reactivation campaign","saas reactivation playbook","how to win back inactive customers","dormant users vs churned users","zombie customers","silent churn","at-risk customer interviews"],"aiSummary":"How to interview still-subscribed but inactive SaaS users to prevent churn before it happens. Distinct from churned/win-back research. Covers the 8 behavioral leading indicators of dormancy, why dormant users are harder to recruit than churned ones, 20 dormant-user-specific interview questions, recruitment tactics that bypass standard panels, a recoverability × root-cause × habit-state analysis framework, and how AI-moderated async interviews like Koji unlock 200+ interviews per week vs. the 5 a traditional researcher can manage.","aiDifficulty":"intermediate","aiEstimatedTime":"16 min read"}],"pagination":{"total":1,"returned":1,"offset":0}}