Lenss to plan deprecations

Sunset a feature without sunsetting the accounts that pay for it.

Deprecation and sunsetting plans, checked against who actually depends on the thing you are retiring and what they pay.

Works on: Deprecation and sunset plans · API version retirements · Legacy tier and plan removals

How the audit runs

  1. 01

    Pin what the sunset must achieve

    The saving or simplification the plan promises, as a number.

  2. 02

    Quote the usage argument

    The sentence that says almost nobody uses it, and the measure it uses: users, calls or accounts.

  3. 03

    Weigh it by revenue

    The accounts behind that small share and the revenue they carry, from the figures in the plan.

  4. 04

    Compute the churn the saving can absorb

    How much of the affected revenue can leave before the sunset costs more than it saves.

Typical findings

Users are not revenue

A small share of users that includes the largest contracts.

No migration path

A sunset date with nothing to move dependent customers onto.

Notice shorter than contracts

Deprecation notice periods that conflict with enterprise terms.

Worked example

The same draft, read by a standard AI assistant and by Lenss. Every figure is computed from the draft.

Source draft

Sunset the v1 Reports API

Proposal: retire the v1 Reports API on March 31. Usage. v1 is used by only 4% of monthly active users (1,240 of 31,000). Everyone else is on v2. Revenue context. Company ARR is $12M. Our three largest enterprise accounts, together $3.6M ARR, still call v1 for their nightly exports. Plan. Announce 60 days before shutdown. No migration tooling: v2 docs are good enough. Benefit. Retiring v1 saves $180,000 a year in maintenance and on-call.

Pinned objective

Save $180,000 a year in maintenance without losing revenue.

Hidden assumption

That 4% of users means 4% of the impact. The same 4% includes the three accounts that carry 30% of ARR.

“Our three largest enterprise accounts, together $3.6M ARR, still call v1 for their nightly exports.”

If it’s wrong: One cancellation at renewal costs about $1.2M of ARR, 6.7 times the annual saving.

Flip threshold

Enterprise ARR on v1 the saving can absorb losing

≤ $180,000 (5.0%) $3.6M depends on v1

Share of users on v1
4%
Share of ARR on v1
30% ($3.6M of $12M)
Annual saving
$180,000
Average ARR per affected account
$1.2M
One lost account vs saving
6.7×

At today's figures, losing any one of the three accounts wipes out the saving for almost seven years. Ship a migration path first.

Cheapest test: Ask the three account owners this week what their nightly export would need from v2, before announcing a date.

Used most by Product managers

Audit your own draft now.

Paste up to 24,000 characters. You get the objective, every quoted assumption, the tensions, the flip threshold and a hardened rewrite.

Audit your draft