--- title: "Upgrade-Moment Offers — Expansion Growth Tactic" description: "Watches the behavioral signals that precede upgrades and makes the offer at the subscriber’s next natural moment of use, not on a calendar." canonical: "https://churnkey.co/growth/library/upgrade-moment-offers" category: "growth" related: - /growth/library/index.html.md - /growth/library/methodology.html.md - /tools/dollar-retention-rate-calculator.html.md last_updated: "2026-06-09" --- # Upgrade-Moment Offers Watches the behavioral signals that precede upgrades and makes the offer at the subscriber’s next natural moment of use, not on a calendar. - **Stage:** Expansion - **Evidence grade:** validated (replicated across many companies) - **Evidence:** 1.7× upgrade conversion when readiness signals are combined into one score - **Trigger type:** Event - **Channels:** in-app ## What is it? Most upgrade prompts run on a schedule: the month-three lifecycle email, the quarterly account review. This tactic inverts that. It maintains a per-subscriber expansion-readiness score that combines four behavioral signals (integration adoption, billing-page visits, settings-page activity, and feature breadth), and when the score crosses its threshold, it makes the upgrade offer at the subscriber’s next natural moment of use. The premise is simple: subscribers tell you they are ready to expand long before they open the pricing page with a credit card in hand. The signals are already in your product’s usage data. The tactic’s job is to read them and pick the moment. ## When it fires Readiness accumulates from behavior, but the firing moment is itself behavioral: the next pricing-page visit, the completion of a meaningful task, or the next active session after the score crosses its threshold. The offer never interrupts work in progress and never arrives as a scheduled email. Timing matters more than most teams assume. In our dataset, 41 percent of upgrades happen in the subscriber’s first month, so the tactic watches from day one rather than waiting for onboarding to settle. ## What the evidence shows In the window before subscribers upgrade, their billing-page visits run 477 percent above their own baseline. Subscribers signal upgrade intent by quietly checking prices. Subscribers who adopt integrations upgrade at 376 percent the rate of those who do not, the strongest single expansion signal in the dataset. No single signal is the story, though. Combining them into one readiness score raises upgrade conversion to 1.7 times that of untargeted prompting. These predictive patterns hold across usage data from roughly 90 subscription companies, consistent and large enough to act on, though not a controlled test. ## How it runs The tactic ingests the four readiness signals from product usage data, recomputes the combined score on each new event, and queues the offer when the threshold is crossed. The render waits for a natural moment: a pricing-page visit, a post-completion screen, or the start of the next session. Frequency caps and collision rules keep the experience respectful: one prompt per readiness crossing, suppression while any other offer or save flow is active, and threshold tuning per engagement pattern from observed conversion. ## Run it with the Churnkey MCP This tactic runs via the Churnkey MCP. Connect it to your favorite AI agent and it reads your own usage and billing data, then recommends and runs the plays most likely to move your LTV — starting with whether this one fits. **Trigger:** A per-subscriber expansion-readiness score, combining integration adoption, billing-page visits, settings-page activity, and feature breadth, crosses the configured threshold. **Surface:** In-app only, at the next natural moment of use: a pricing-page visit, a post-task completion moment, or the start of the next active session. Never a scheduled email and never mid-task. **Guardrails:** - Rate limit: at most one upgrade prompt per subscriber per 14 days, and no more than three per quarter. - Eligibility: active subscription in good standing, a higher tier available, no open payment failure, and no recent downgrade. - Collision avoidance: suppress while a Cancel Flow, Payment Recovery sequence, or any other offer is active for the subscriber. Defer rather than queue-jump. - Threshold discipline: calibrate the readiness threshold on historical upgrades before going live, and re-tune it per engagement pattern from observed conversion. **Evidence:** Billing-page visits run 477 percent above a subscriber’s own baseline in the window before an upgrade. Integration adopters upgrade at 376 percent the rate of non-adopters. A combined readiness score raises upgrade conversion to 1.7 times that of untargeted prompts. And 41 percent of upgrades happen in the subscriber’s first month. Validated at scale across usage data from roughly 90 subscription companies. **How to run it:** 1. Instrument the four readiness signals: integration adoption, billing-page visits, settings-page activity, and feature breadth. 2. Combine them into a per-subscriber expansion-readiness score and recompute it on each new event. 3. Calibrate the threshold on historical upgrade data so the high-readiness group converts well above baseline before any subscriber sees a prompt. 4. When the score crosses the threshold, queue the offer and render it at the next natural moment of use: pricing-page visit, post-completion moment, or next session start. 5. Start scoring from the subscriber’s first day, since a large share of upgrades lands in the first month. 6. Track prompt-to-upgrade conversion by signal mix and re-weight the score on a regular cadence. Install the MCP server: ```bash npm install -g @churnkey/mcp ``` The full reference is in the [Churnkey MCP docs](https://docs.churnkey.co/data-integrations/mcp). ## Related - [Growth Tactics Library](/growth/library/index.html.md) — The full register of graded growth plays. - [How Tactics Are Graded](/growth/library/methodology.html.md) — The four-grade evidence ladder behind this grade. - [Dollar retention rate calculator](/tools/dollar-retention-rate-calculator.html.md) — Related calculator.