Average LARR per User

What it measures

Live ARR divided by the count of active user seats on live, revenue-generating contracts at period end. The numerator is recurring revenue only — one-time fees, professional services, and usage overages are excluded — and counts only contracts that are actually billing today, not signed-but-not-yet-active (contracted) deals. The denominator counts every live seat equally, regardless of which account it sits in; trial, suspended, and churned seats are excluded. It captures the average annual value of a single live seat, including tier pricing and seat-level add-ons folded into the subscription.

Why it matters

You track Average LARR per User to read seat-level monetization across your live, paying base. It answers one question: how much recurring revenue does the average active seat carry today? Sales ops uses it to forecast expansion revenue; product uses it to see whether feature adoption is pushing seats up-tier; finance pairs it with seat-level acquisition cost to model payback per seat. Because it sits on Live ARR rather than contracted ARR, it reflects realized run-rate — what you are actually billing — which is the number a board trusts when judging whether seat economics are improving. Rising LARR/User alongside a stable seat base proves expansion is outpacing the dilution of cheap new seats; falling LARR/User often precedes an ARR decline, so it works as an early-warning gauge.

How to read it

Read LARR/User as a trend, never a single snapshot — compare period to period and against plan. Rising LARR/User means seats are moving up tiers or buying add-ons faster than entry-level seats dilute the mix; flat LARR/User while seat count climbs means new seats are landing at thin pricing. A 20% year-over-year increase while Live ARR grows 30% is healthy — value per seat is expanding faster than you are adding cheap seats. Flat LARR/User on 40% ARR growth means you are winning a lot of low-value seats; decide whether that is a deliberate land-and-expand motion or a pricing and segmentation problem. Always cross-check with net revenue retention and seat-level CAC payback: rising LARR/User without rising NRR can mask contraction in some segments hidden by a churn of cheap seats.

What good looks like

Watch-outs

• Treating LARR/User as a period average. It is the ratio at period end, not the blended rate across the period. If you opened with 150 seats and closed with 200, use 200 — the metric asks what the average live seat is worth today, not what the average was during the period.
• Counting trial, free, or suspended seats. Including non-billing seats inflates the denominator and pushes LARR/User down artificially. Count only seats on live, paying contracts as of the snapshot date.
• Mixing Live ARR with Contracted ARR. Use Live ARR — contracts actually billing today — unless you specifically want a contracted variant. Live is realized run-rate; contracted is committed future value, and the two tell different stories.
• Reading LARR/User without churn context. Two companies both at \$25K LARR/User look identical but diverge entirely if one acquires \$50K seats and keeps them while the other acquires \$60K seats but loses half to churn. Always pair LARR/User with NRR and seat churn.

Worked example