# Break-even calculator for a price increase

Use this before anyone names a percentage out loud. It answers one question: if you raise price by X percent, what retention rate do you need just to break even on revenue? Below that retention rate, the increase costs you money even though each remaining client pays more.

## The formula

`retention_needed = 1 / (1 + increase_percent)`

Where `increase_percent` is written as a decimal (10 percent is 0.10, not 10).

`allowable_churn = 1 - retention_needed`

Read it as: raise price 10 percent, and you can afford to lose about 9 percent of clients and still land at the same total revenue as before the increase. Lose more than that, and the increase costs you money overall, even though each remaining client pays more.

This formula measures break-even on revenue. It assumes your cost per remaining client stays flat, so it doesn't capture a profit break-even on its own. If the increase is meant to fund new costs (more support hours, a new tool, a bigger team), size the percentage against that cost first, then run this formula on top of it.

## Why the churn number is not optional

The Federal Reserve Bank of Richmond's Economic Brief 23-16 (Gholami and Trachter, May 2023) found that a 1 percent price change can raise annual customer turnover from 14 percent to 21 percent. A small bump on price moves turnover by 7 full points in that data set, a real cost even on a modest increase. Run the numbers while a percentage can still be changed, before any notice goes out.

## Candidate percentage table

| Price increase | Retention needed to break even | Allowable churn |
|---|---|---|
| 2% | 98.0% | 2.0% |
| 5% | 95.2% | 4.8% |
| 8% | 92.6% | 7.4% |
| 10% | 90.9% | 9.1% |
| 12% | 89.3% | 10.7% |
| 15% | 87.0% | 13.0% |
| 18% | 84.7% | 15.3% |
| 20% | 83.3% | 16.7% |
| 25% | 80.0% | 20.0% |
| 30% | 76.9% | 23.1% |

Use the row closest to the percentage under discussion. If the number you want is not on the table, run `scripts/breakeven.js` with that percentage as the argument.

## How to use this with a client roster

1. Pull the total number of active clients and the current average price.
2. Pick two or three candidate percentages, never just one.
3. For each candidate, read off the allowable churn from the table above.
4. Multiply allowable churn by total active clients. That is the maximum number of clients you can lose at that percentage before the increase becomes a net loss.
5. Compare that number against the clients you already suspect might leave (from the segmentation work in `references/segmentation-rubric.md`). If the suspect count is close to or above the allowable count, the percentage is too aggressive for this roster right now.

## Worked check

Say you have 40 clients at $500/month and you are weighing a 15 percent increase.

- Retention needed: 1 / 1.15 = 0.870 (87.0%)
- Allowable churn: 13.0%
- 13.0% of 40 clients = 5.2, so you can lose up to 5 clients and still come out ahead on revenue.

If your gut says more than 5 of your 40 clients will balk at 15 percent, either lower the percentage, tie it to a named cost driver (see the value-tied template in `references/notice-templates.md`), or plan to lose the margin on purpose, because the relationship is worth it.

# Client segmentation rubric

Tier every active client before drafting a single notice. The tier decides which template a client gets in Phase 2 and the order they are contacted in.

## Score each client on four dimensions (1 to 3 each)

| Dimension | 1 point | 2 points | 3 points |
|---|---|---|---|
| Tenure | Under 6 months | 6 months to 2 years | Over 2 years |
| Revenue value | Below your median client revenue | At your median client revenue | Top quartile of client revenue |
| Relationship strength | Transactional, little contact beyond invoices | Friendly, occasional unprompted positive feedback | Advocate: refers people, responds personally, treats you as a partner |
| Referral likelihood | Has never referred anyone | Has referred once | Has referred more than once, or is a known connector in their circle |

Total score range: 4 to 12.

## Assign the tier from the total

| Total score | Tier | What it means |
|---|---|---|
| 4 to 6 | Test group | Smallest exposure if this goes wrong. Goes first, gets the test-group email (`references/notice-templates.md`, template 1). |
| 7 to 9 | Standard | The bulk of the roster. Gets the value-tied email when a named cost driver applies, the standard email when none does (`references/notice-templates.md`, templates 3 and 2), batched after the test group holds. |
| 10 to 12 | Top tier | Highest cost if lost, highest value if kept. Gets the personal-delivery script, never a cold email. |

## Rules while scoring

- Score every active client. No one skips scoring because they "seem easy" or "seem obviously fine": a client assumed to be fine and left unscored still ends up with the wrong template later.
- Score honestly on relationship strength. A client who pays on time but never responds to anything beyond an invoice is a 1, not a 2, even if they have never complained.
- When a client sits on the boundary between two tiers (a 6 vs. a 7, a 9 vs. a 10), round down in risk: treat them as the lower tier, which gives them more notice time and a softer approach.
- Re-run this scoring every time a price increase is planned. A client's tenure and relationship strength change between cycles; do not reuse last cycle's tier without checking.

## Minimum fields to log per client

These columns are the segmentation slice of the full roster log; see
`references/roster-log-schema.md` for the complete schema, including the
contract fields and each cycle's rollout outcomes. Keep this as a simple table
or spreadsheet, one row per client:

| Client | Tenure score | Revenue score | Relationship score | Referral score | Total | Tier | Contract type | Renewal date | Required notice period (days) |
|---|---|---|---|---|---|---|---|---|---|

The last three columns come from the contract audit in Phase 1, step 2 of the main procedure, not from this rubric. Keep them in the same table so the tier and the contract constraint are visible together: a top-tier client with a 90-day notice requirement needs to enter the cadence in `references/cadence-timeline.md` earlier than a test-group client on a month-to-month agreement.

## What this feeds

- The tier decides the notice order in Phase 2 (test group first, then standard tiers in ascending value, top tier last and by a different channel).
- The tier decides the template in `references/notice-templates.md`.
- The contract-type and notice-period columns decide where each client lands on `references/cadence-timeline.md`.

# Roster log schema

This is the one living file this system starts in Phase 1, extends through
Phase 2, and re-opens every cycle in Phase 3. Keep it as a single spreadsheet
or table, one row per client, with one block of rollout-record columns added
per cycle (add new columns for the next cycle rather than overwriting the
last one, so the trend survives across quarters).

## Identity and contract columns (started in Phase 1, step 2)

| Column | What goes here |
|---|---|
| Client | Client name or account id |
| Contract type | Month-to-month, annual retainer, fixed-term, etc. |
| Renewal date | Next contract renewal or expiry date |
| Required notice period (days) | Pulled from the contract itself |
| Price lock until | Date a price-lock clause expires, blank if none applies |

## Segmentation columns (added in Phase 1, step 3; see `references/segmentation-rubric.md`)

| Column | What goes here |
|---|---|---|
| Tenure score | 1 to 3 |
| Revenue score | 1 to 3 |
| Relationship score | 1 to 3 |
| Referral score | 1 to 3 |
| Total score | Sum of the four, 4 to 12 |
| Tier | Test group, Standard, or Top tier |

## Rollout-record columns, one block per cycle (Phase 2, step 12; Phase 3)

Label each block with the cycle it covers, for example "2026 Q1 rollout":

| Column | What goes here |
|---|---|---|
| Cycle label | e.g. "2026 Q1" |
| Increase percent | The percentage applied this cycle |
| Template used | Test-group email, Standard email, Value-tied email, or Personal-delivery script |
| Effective date | Date the new price took effect for this client |
| Sent date | Date the user confirmed the notice went out |
| Outcome | Accepted, Renegotiated, or Left |
| Notes | Anything from step 9's test-group read or step 14's churn comparison worth carrying into the next cycle |

## How each step uses this file

- Step 2 starts the Identity and contract columns for every active client.
- Step 3 fills in the Segmentation columns and assigns the Tier.
- Step 12 fills in the current cycle's Rollout-record block once every tier has a notice out.
- Step 13 re-opens this file at the start of the next cycle, re-scores the Segmentation columns for any client whose situation changed, and adds a new Rollout-record block instead of overwriting the last one.
- Step 14 reads the Outcome column against that cycle's break-even prediction and logs the gap in the Notes column.

Keep one file across cycles. Starting a fresh spreadsheet each quarter loses the trend line that makes the next break-even estimate better than a guess.

# Notice templates

Four templates. Pick the one that matches the client's tier from `references/segmentation-rubric.md`. Fill every bracket before sending. Do not send with a bracket still in the text.

## 1. Test-group email

Use for the smallest, lowest-risk tier, the first batch out the door.

Subject: A price update for your account, effective [EFFECTIVE_DATE]

Hi [CLIENT_NAME],

I'm writing with advance notice on pricing. Starting [EFFECTIVE_DATE], your rate will move from [CURRENT_PRICE] to [NEW_PRICE]. That's [NOTICE_PERIOD_DAYS] days from today, so nothing changes before then.

This applies across the board as part of a planned update to how [YOUR_BUSINESS_NAME] prices its work. If the new rate is a problem for your budget, reply to this email and let's talk before the change date.

Thanks for being a client.

[YOUR_NAME]

## 2. Standard email

Use for the standard tier once the test group has held, when no specific, named cost driver or scope change applies to this client (use the value-tied email, template 3, instead whenever one does). Slightly more context than the test-group version since this batch is larger and more varied.

Subject: Your rate is changing on [EFFECTIVE_DATE]

Hi [CLIENT_NAME],

Starting [EFFECTIVE_DATE] ([NOTICE_PERIOD_DAYS] days from today), your rate moves from [CURRENT_PRICE] to [NEW_PRICE]. Nothing else about the work changes.

This is part of a pricing update across all [YOUR_BUSINESS_NAME] clients, not something specific to your account. New clients signing on today already pay the new rate.

If you want to talk it through, reply here or grab time on my calendar: [CALENDAR_LINK].

[YOUR_NAME]

## 3. Value-tied email

Use for standard-tier clients where a specific, named cost driver or scope change explains the increase. This is the default pick for the standard tier whenever a real driver exists; use the standard email (template 2) when none does. Never send a bare percentage to a client you have a real relationship with when a real reason exists; name it.

Subject: Why your rate is changing on [EFFECTIVE_DATE]

Hi [CLIENT_NAME],

Starting [EFFECTIVE_DATE], your rate moves from [CURRENT_PRICE] to [NEW_PRICE]. Here's the reason: [NAMED_COST_DRIVER_OR_SCOPE_CHANGE, for example "the scope has grown to include weekly reporting that wasn't part of the original agreement" or "the tools this work depends on raised their own prices this year"].

I wanted you to hear the reason directly rather than just see a new number on an invoice. The new rate takes effect [NOTICE_PERIOD_DAYS] days from today.

Questions or concerns, reply here.

[YOUR_NAME]

## 4. High-value personal-delivery script

Use for the top tier. This is delivered by phone or in person, by the owner or the account lead, never by cold email. Use these as talking points to adapt in the moment; reading them word for word defeats the purpose.

Opening:
"I wanted to talk to you directly about something rather than send it in an email."

The news:
"Starting [EFFECTIVE_DATE], the rate on your account is moving from [CURRENT_PRICE] to [NEW_PRICE]. I wanted you to hear that from me, not find it on an invoice."

The reason (always include one for a top-tier client):
"[NAMED_REASON, tied to cost, scope, or the value delivered over the relationship's tenure]."

The relationship anchor:
"You've been a client for [TENURE], and that matters to me."

The open door:
"If this timing is a problem, or the number is a problem, tell me now and we'll find something that works for both of us. I'd rather have that conversation than lose you over a number."

Close:
"I'll follow up in writing with the details, but I wanted you to hear it from me first."

Confirm before marking this tier as handled: the owner (or account lead) actually has a delivery method booked, a call on the calendar or an in-person meeting set, not just an intention to call "sometime."

## Before sending any of these

- Every bracket is filled. A visible bracket in a sent notice is a signal this was templated and rushed.
- The effective date matches the notice period logged for that specific client's contract, not a flat default for everyone.
- The draft has been shown to the owner and approved in the same conversation before it is sent, scheduled, or queued.

# Cadence timeline: audit to sustain

Two tracks below. Use the small-account track as the floor (30 days) and the large-account track for retainer or annual contracts (60 to 90 days). Build the actual calendar from whichever notice period a client's contract requires (logged in the segmentation table), not from a single flat date for the whole roster.

## Small / month-to-month track (30-day floor)

| Day | Action |
|---|---|
| Day 0 to 3 | Audit contracts for price locks, renewal dates, required notice periods. Run the break-even math on 2 to 3 candidate percentages. |
| Day 3 to 5 | Segment the full roster into tiers. Set the new public, new-customer price. |
| Day 5 | Draft the test-group notice. Show the owner, get approval. |
| Day 5 to 7 | Send the test-group notice. Log the send date. |
| Day 7 to 14 | Collect test-group replies and any cancellations. Compare against the break-even allowable churn. |
| Day 14 | Decision point: hold the percentage and proceed, or pause and revisit. |
| Day 14 to 21 | Draft and send the standard-tier batches, value-tied template where a cost driver exists. |
| Day 21 to 28 | Deliver the top tier by phone or in person, using the personal-delivery script. |
| Day 28 to 30 | New rate takes effect for every client whose individual notice period has elapsed. Log final outcomes (accepted, renegotiated, left). |

## Large / retainer or annual track (60 to 90 days)

| Day | Action |
|---|---|
| Day 0 to 7 | Audit contracts (these often carry longer required notice periods; 60 or 90 days is common in annual agreements). Run the break-even math. |
| Day 7 to 14 | Segment the roster. Set the new public price. |
| Day 14 | Draft and approve the test-group notice. |
| Day 14 to 21 | Send the test-group notice. |
| Day 21 to 35 | Collect test-group signal, a longer window than the small-account track, since larger clients often take longer to respond. |
| Day 35 | Decision point. |
| Day 35 to 55 | Standard-tier batches. |
| Day 55 to 75 | Top-tier personal delivery, scheduled individually since these clients often need a real meeting, not a phone call squeezed in. |
| Day 75 to 90 | New rate takes effect per client's individual notice requirement. Log outcomes. |

## Rule for mixed rosters

Most rosters have both month-to-month and annual clients. Do not force everyone onto one track. Build one calendar with each client's row dated from their own contract's notice period (the column logged in `references/segmentation-rubric.md`), using these two tracks as the floor and ceiling; a client's own contract can land anywhere between them.

## Quarterly reuse (Phase 3)

Keep this file as a living document. On the next price increase:

1. Do not start from a blank calendar. Copy this file and re-date it from the new start date.
2. Compare the actual Day 7-to-decision-point churn from last time against the break-even number predicted for last time. Note the gap at the top of the new copy.
3. Carry forward any client whose notice period or contract changed since the last cycle.

#!/usr/bin/env node
/**
 * breakeven.js
 *
 * Turns a price-increase percentage into the retention rate needed to
 * break even on revenue, and the churn rate you can afford before the
 * increase costs you money.
 *
 * Usage:
 *   node breakeven.js                   Prints the default reference table.
 *   node breakeven.js 12                Prints the result for a single 12% increase.
 *   node breakeven.js 5 10 15           Prints results for each percentage given.
 *   node breakeven.js --clients 40 15   Also shows the client-count math
 *                                       for a 40-client roster at 15%.
 *
 * No dependencies. No network access. No files are read or written.
 */

function breakeven(increasePercent) {
  const x = increasePercent / 100;
  const retentionNeeded = 1 / (1 + x);
  const allowableChurn = 1 - retentionNeeded;
  return { increasePercent, retentionNeeded, allowableChurn };
}

function formatPct(n) {
  return (n * 100).toFixed(1) + '%';
}

function printRow(result, totalClients) {
  const base = `${result.increasePercent}% increase -> retain ${formatPct(result.retentionNeeded)}, can afford to lose ${formatPct(result.allowableChurn)}`;
  if (totalClients) {
    const maxLost = (result.allowableChurn * totalClients).toFixed(1);
    console.log(`${base} (up to ${maxLost} of ${totalClients} clients)`);
  } else {
    console.log(base);
  }
}

function main(argv) {
  let totalClients = null;
  const percents = [];

  for (let i = 0; i < argv.length; i++) {
    const arg = argv[i];
    if (arg === '--clients') {
      totalClients = Number(argv[i + 1]);
      i++;
      continue;
    }
    const n = Number(arg);
    if (!Number.isNaN(n) && n > 0) {
      percents.push(n);
    }
  }

  if (percents.length === 0) {
    // Default reference table, matches references/breakeven-calculator.md
    [2, 5, 8, 10, 12, 15, 18, 20, 25, 30].forEach((p) => {
      printRow(breakeven(p), totalClients);
    });
    return;
  }

  percents.forEach((p) => {
    printRow(breakeven(p), totalClients);
  });
}

main(process.argv.slice(2));

# Worked example: running the system twice

Two worked examples below, not real client records. Both use invented client names and numbers on an invented roster, so this fictional roster is never confused with a real client list. The stats behind them are real and sourced: the churn research cited in step 1 of the main procedure (Federal Reserve Bank of Richmond, Economic Brief 23-16, Gholami and Trachter, May 2023) and the break-even model match `references/breakeven-calculator.md`. The 10 to 20 percent "commonly accepted annual increase" range used below matches the range given in sourced pricing guides, including Nutshell's and HoneyBook's price-increase guides (see this skill's source list); several independent guides converge on this range, though it reflects a repeated pattern across guides rather than a single rigorous study's finding. Treat this file as a template for the arithmetic and the sequence of moves. The arithmetic and sequence are what to reuse; the specific client names and numbers below are invented for illustration only.

## Walkthrough 1: a 12-client design studio, month-to-month contracts

**Starting point:** 12 clients, average $1,200/month, all month-to-month, no price-lock clauses (confirmed by the contract audit in Phase 1, step 2).

**Step 1, break-even math:** Owner is weighing 10% or 15%.
- 10%: retention needed 90.9%, allowable churn 9.1%. On 12 clients, roughly 1 client.
- 15%: retention needed 87.0%, allowable churn 13.0%. On 12 clients, roughly 1.6 clients.
Owner picks 12%, between the two, after also weighing the Richmond Fed finding that even a 1 percent price change moves annual turnover from 14 to 21 percent. At 12% the owner goes in expecting real churn risk, not a free bump.

**Step 3, segmentation:** All 12 clients scored on the four-dimension rubric. Three land in the test group (newer, lower revenue), seven in the standard tier, two in the top tier (longest tenure, most referrals).

**Step 5:** Owner updates the public price on the studio's site to the new, higher number before any existing client sees a notice.

**Phase 2, test group:** The three test-group clients get the test-group email, effective 30 days out (the month-to-month floor from `references/cadence-timeline.md`). All three reply; two say nothing beyond "got it, thanks"; one asks for a 60-day delay, which the owner grants.

**Reading the test signal:** Zero cancellations against an allowable churn of roughly 1 client at 12%. The owner proceeds.

**Standard tier:** Seven clients get the value-tied template; the named driver is a new weekly reporting step added to the service earlier in the year. One client (the smallest of the seven) cancels at the end of their notice period. That is within the allowable-churn range from step 1, so the rollout is not paused.

**Top tier:** Two clients get the personal-delivery script, delivered by phone. Both stay. One negotiates a locked rate for 12 months in exchange for a referral commitment, logged as a renegotiation, not a plain acceptance.

**Outcome logged for Phase 3:** 11 of 12 clients retained, 1 lost, against a modeled allowable loss of about 1. The actual result landed right at the model's edge, which the owner notes in the roster log as a reason to be slightly more conservative on the percentage next time.

## Walkthrough 2: a 6-person marketing agency, mixed month-to-month and annual retainers

**Starting point:** 22 clients. 14 month-to-month, 8 on annual retainers (two of which carry a 90-day notice clause found in the contract audit).

**Step 1:** Owner tests 8%, 10%, and 15%. Chooses 10% (retention needed 90.9%, allowable churn 9.1%, roughly 2 of 22 clients) after checking that this fell inside the commonly cited 10 to 20 percent range for an accepted annual increase.

**Step 2, contract audit:** The two 90-day-notice annual clients are flagged and routed onto the large-account track in `references/cadence-timeline.md`, dated separately from the other 20 clients on the small-account track.

**Step 3, segmentation:** Tiering splits roughly evenly: 8 test-group, 10 standard, 4 top-tier (the two 90-day contracts are both top-tier by revenue, which is why their longer notice period matters).

**Phase 2:** The test group of 8 holds with one cancellation (a client already flagged in the owner's notes as planning to wind down the engagement regardless of price). The standard tier of 10 gets the value-tied email, naming a cost driver (a tools and software cost increase the agency absorbed the prior year without passing it on). One more cancellation here. The top tier of 4, including the two 90-day clients, gets the personal-delivery script; the two 90-day clients each get a scheduled call rather than being squeezed into a standing weekly sync, and both stay.

**Outcome:** 20 of 22 retained, against a modeled allowable loss of about 2. The result matched the model almost exactly. Logged in the roster log as the baseline to compare the next cycle against (Phase 3, step 14 of the main procedure).

## What both walkthroughs show

- The break-even number sets a planning ceiling for how much churn the increase can absorb. Walkthrough 1 landed right at the edge of it; walkthrough 2 landed inside it. Both count as "the system worked," because both outcomes were checked against that model instead of judged by gut feeling after the fact.
- Contract notice periods change the calendar. The 90-day clients in walkthrough 2 used the same break-even formula and the same tier logic, just a longer cadence track.
- A single bad test-group signal, none appeared in either walkthrough, is always the trigger to pause the rollout and revisit the plan with the owner; the rule stands regardless of how smoothly a rollout goes.

Run the break-even math first: retention needed = 1 / (1 + increase percent). A 10 percent increase needs roughly 91 percent of clients to stay to break even; a 20 percent increase needs roughly 83 percent. See what retention rate a number requires before you commit to it.

Stop sending anything else out. Log who has left and who has complained, compare that against what the break-even math says you could afford, and for everyone still active, move to the tiered rollout before another blanket notice goes out.

Tie the number to something real, such as a cost driver, added scope, or a service change. State the new price, the effective date, and what's included, and skip the apologetic tone entirely.

No. The highest-value tier gets a call or an in-person conversation, not an email. Everyone else can get a written notice, staged by tier.

At least 30 days for month-to-month or small accounts, 60 to 90 days for larger retainer or annual contracts, plus whatever the actual contract already requires on top of that.

Quarterly, or whenever the topic comes up again. Re-run the break-even math against the new candidate number, re-tier anyone whose situation changed, and compare the last rollout's actual churn to what was predicted.