Downsell Annual Recurring Revenue
The annualized recurring revenue lost when existing customers downgrade to lower tiers or remove add-ons.
◆ Currency
Formula
Built from
What it measures
The sum of the annualized value of all recurring revenue lost in a period because existing customers downgraded subscription tiers, removed add-ons, or moved to cheaper plan variants. One-time credits, discounts applied at renewal (not reductions in plan), and paused subscriptions are excluded — only active revenue reductions from plan changes count.
Why it matters
Downsell ARR quantifies revenue leakage from your installed base in absolute dollars, revealing whether a retention problem is really customer flight or a hidden expansion failure. Leadership uses it to read the true net picture of retention — a company that keeps customers but watches them trade down is healthier than one with dramatic churn but no downgrades. It forces the conversation from "our net-new looks OK" to "we're losing $1M ARR a year to downgrades — are our plans priced right, or is the product failing at retention?"
How to read it
High downsell ARR doesn't automatically mean the business is broken — it can signal customers right-sizing spend or a healthy price-discovery process. But read it as a trend: compare it month-over-month and cohort-by-cohort, and ask whether it's growing, flat, or shrinking. Normalize against starting ARR to get your gross contraction rate, then compare downsell ARR to new and upsell ARR to gauge the expansion engine. If downsells are rising while new ARR is flat, your installed base is contracting; if downsells stay constant as revenue scales, you may be acquiring lower-LTV logos.
What good looks like
Good
Downsell ARR is under ~5% of starting ARR annually, indicating strong product adoption and customers comfortable on their tier.
Watch
Downsell ARR runs roughly 5–10% of starting ARR; investigate which segments and cohorts are downgrading and why.
Bad
Downsell ARR exceeds ~10% of starting ARR; customers are systematically trading down, signaling pricing misalignment, product dissatisfaction, or economic headwinds.
Watch-outs
- Conflating downsell with churn. A downgrade is a retention win (the customer stayed); churn is a loss. A $100K customer moving to $60K is Downsell ARR of $40K, not churned ARR.
- Counting same-price plan migrations as downsells. If a customer switches from monthly to annual billing at the same effective monthly rate, that's zero downsell — only actual revenue reductions count.
- Not prorating mid-month changes. If a customer downgrades on the 15th, count the old price for the 1st–14th and the new price from the 15th on; an unadjusted full-month swap inflates downsell ARR.
- Forgetting the time-lag. A downgrade requested on April 15 that takes effect May 1 belongs to May's cohort, not April's. Don't prorate backward.
Worked example
Hypothetical
Three existing customers downgrade this month: Company A by $2K/month, Company B by $1K/month, and Company C by $3K/month — $6K/month of contraction in total. That is $6K of Downsell MRR; annualized, Downsell ARR is $6K × 12 = $72K.