# Concentration risk thresholds

How much of your revenue one client can carry before it's a problem, not a compliment.

## The first warning line

A single client at 10% or more of your trailing 12 months of revenue, or your top 5 clients combined at 25% or more, is the point where accountants, business brokers, and lenders start asking questions. Not because 10% is dangerous on its own. Because it's the line where "a great client" starts turning into "a dependency," and dependencies grow quietly if nobody's watching the number.

## Where it turns into a bind

A single client at 25% or more, or your top 5 combined at 50% or more, is what a business broker or a buyer's diligence team will flag as high concentration, full stop. At this level, one client's decision (to leave, to renegotiate, to get acquired and switch vendors) can knock a real hole in your revenue in a single quarter. If you're at 25% on one client, you don't have a business with a strong client. You have a business with one boss who doesn't sign your paycheck but might as well.

These aren't hard legal thresholds. They're the range where the people who look at businesses for a living, brokers pricing a sale, lenders underwriting a loan, accountants doing your books, start treating your revenue as less durable than the total on the page suggests. Different advisors draw the line a few points apart. The direction is the same across all of them: 10 to 25% is worth a written plan, 25% and up is worth an active one.

## Setting your own ceiling

The benchmark range tells you where the general danger zone starts. It doesn't tell you your number. That's yours to set, in writing, with a deadline.

A few things that help pick it:

- If you're nowhere near the warning line, your ceiling can be generous. Set it a few points above where you are today, so you have room to grow with a client without tripping your own alarm on a good quarter.
- If you're already past the warning line, your ceiling should sit below where the "high concentration" range starts, and your deadline should be short enough that you're not just hoping it improves.
- If you're already in high-concentration territory, your ceiling is a target to work down to, not a line to hold. Say so plainly when you set it. "Get single-client share under 30% within three years" is a real ceiling. "Try to diversify more" is not.

Write the number and the date down somewhere you'll actually see again next quarter. A ceiling nobody looks at isn't a ceiling.

## Reference point: one owner's path from 48% to under 30%

An owner running a service business found, the first time she actually pulled the numbers, that her single largest client was 48% of her trailing 12 months. It had crept there over three years of a good relationship: more projects, bigger scope, and she'd never once added it up.

She didn't drop the client. She set a written ceiling instead: no single client above 30% of revenue, within three years. Then she worked both tactics in the diversification playbook, at the same time, for two years straight rather than in a single push after a scare. Repricing and scope conversations with a handful of smaller clients who hadn't had one in over a year grew that side of the business. New prospecting was aimed at clients who looked like her healthy mid-size accounts, not at the next big logo. Eighteen months in, the big client was down to 34%. By the three-year mark, it was at 27%, and no single conversation had done it. The quarterly number moving a few points at a time had.

The number that mattered wasn't the ceiling itself. It was having one written down to measure against every quarter, instead of only noticing the share had grown the day the client's business got sold and the new owner brought in their own vendor.

# Measurement method: the trailing-12-month calculation

The two numbers this whole register runs on: one client's share of revenue, and the top 5 clients' combined share. Both come from the same twelve months of billing, recalculated every quarter.

## Why trailing 12 months, not a calendar year or a single quarter

A calendar year or a single quarter can be skewed by one big project landing in December, or by a slow month that happens to fall inside the window. A trailing 12 month window (the twelve months ending on the date you're checking, not the twelve months of the last fiscal year) moves with you every quarter and smooths out that kind of noise. It also means the register never waits for a fiscal year-end to catch a trend.

## Gathering the input numbers

Pull actual revenue collected or invoiced (pick one method and stay consistent every quarter) for each client, for the trailing 12 months ending on the check date. Two totals per client:

- client_ttm_revenue: that one client's trailing-12-month revenue
- total_ttm_revenue: every client's trailing-12-month revenue added together

Never estimate a client's revenue from memory or a rough sense of "they're big." If the number isn't in the books yet, ask for it or leave the row blank until it is.

## The calculation

For each client:

`client_pct_of_revenue = client_ttm_revenue / total_ttm_revenue * 100`

Rank every client by client_ttm_revenue, highest to lowest. The top 5 get `in_top5 = yes`; everyone else gets `no`. Add those five clients' client_ttm_revenue together and divide by total_ttm_revenue to get `top5_combined_pct`, written the same on each of the five top-5 rows so the number travels with the data instead of living in a separate place.

### Worked example

A business with $900,000 in trailing-12-month revenue across all clients, and a single largest client billing $432,000 over that same window:

`432,000 / 900,000 * 100 = 48%`

That 48% is what goes in client_pct_of_revenue for that client's row. If the next four largest clients add up to $130,000 combined, the top 5 total is $562,000:

`562,000 / 900,000 * 100 = 62%`

That 62% goes in top5_combined_pct on all five of the top-5 rows for that quarter.

## What throws this off

- Mixing invoiced revenue for one client with collected revenue for another. Pick one definition, use it for every client, every quarter.
- Counting a one-time project or an early-termination payout as ordinary revenue. Flag it in the notes column instead of letting it inflate a client's normal run rate.
- Recomputing from a calendar year instead of a rolling trailing 12 months. The window has to end on the check date, not on December 31, or the quarter-over-quarter comparison in Phase 2 stops meaning anything.
- Forgetting to re-rank the top 5 each quarter. A client that was 6th last quarter can move into the top 5 this quarter, and the combined percent has to reflect whoever is actually in the top 5 now, not who was there last time.

# Valuation consequence: what a concentrated client does to a sale price

A buyer, or a business broker running diligence on a buyer's behalf, does not price revenue on the total alone. They price it on how durable that revenue is without the current owner and without any one relationship holding it together. A large share sitting on one client changes the number, whether or not the deal ever explicitly says so.

## The three ways it shows up in an offer

**A straight multiple discount.** The buyer applies a lower earnings multiple to the whole business, reasoning that a chunk of the revenue could walk with one phone call. A business that would otherwise price at 4x EBITDA might get offered 3x or 3.5x instead, with the difference explained as "concentration risk" in the diligence notes.

**An earnout tied to the client staying.** Instead of a flat discount, part of the price gets deferred and made conditional: the seller gets the rest only if the concentrated client is still on the books, at a similar spend, one or two years after close. This shifts the risk from buyer to seller without lowering the headline price, which can look better on paper and cost more in practice if the client leaves in year one.

**A stay-on requirement.** The buyer asks the owner to remain involved, sometimes for years, specifically to keep the relationship with the concentrated client intact through the transition. For an owner trying to exit, this is usually the least welcome version: it delays the actual exit past the closing date.

## Worked example

A business does $2,000,000 in trailing-12-month revenue with $500,000 in EBITDA. One client accounts for $800,000 of that revenue, 40%. At a clean 4x EBITDA multiple with no concentration issue, the business would price around $2,000,000.

A buyer's diligence team flags the 40% figure and does one of three things: drops the multiple to roughly 3x ($1,500,000), structures $400,000 of the price as an earnout paid out over two years only if the client's spend holds, or asks the owner to stay on in a client-facing role for 18 months post-close. Any of the three costs the seller real money or real time, and all three trace back to the same 40% number this register tracks every quarter.

## When to raise this with the owner

Raise it the first time concentration comes up in Phase 1, even if a sale isn't imminent: knowing the consequence in advance is what makes a written ceiling feel worth setting rather than optional. Raise it again, specifically, the moment the owner mentions preparing for a sale, being approached by a buyer, or starting to think about an exit timeline. A concentration number that's been under a written ceiling for two or three years by the time a sale process starts is a materially different conversation with a buyer than one discovered for the first time during diligence.

# Pipeline infrastructure: running the weekly block

The register measures the problem. This is the standing mechanism that keeps a new one from building back up, whether or not any client is currently near the ceiling.

## Setting up the block

Pick one fixed day and time with the user, weekly, and put it on the actual calendar, not a mental note to "get to outreach when things are quiet." The block needs a real duration (60 to 90 minutes works for most solo and small-team owners) and it needs to survive a busy week. The test for whether it's really protected: would the user cancel a client deliverable's deadline to free up this slot? If the honest answer is no, the block isn't protected the same way yet, and step 20 in Phase 4 exists to catch that before moving on.

## What happens inside the block

Three things happen every week inside this slot, not just when a client feels shaky:

**Stay visibly involved in the big client's own planning.** This doesn't mean extra meetings. It means the owner keeps a seat, even a passive one, in whatever planning or budget conversations that client runs on their end: a quarterly business review, a renewal check-in, a budget cycle update. A pullback shows up here first, weeks or months before it shows up as a cancelled invoice. Use part of the block to review whatever visibility already exists (recent emails, meeting notes, a shared roadmap) and note anything that reads differently than it did last month.

**Check delivery quality and documentation.** A concentration problem becomes a service problem fast if the relationship starts fraying at the same time work quality slips. Use part of the block to spot-check recent deliverables against what was promised, and to confirm documentation (contracts, scope changes, correspondence on anything sensitive) is current. This isn't extra work invented to fill the slot; it's the insurance that keeps one risk from compounding into two.

**Run outreach to the prospect profile from Phase 3, step 15.** Not the biggest logo available, the profile that matches the healthy mid-size clients already in the register. A short list of names, a message drafted from examples/prospect-outreach-draft-example.md, presented for approval before anything goes out.

## Multi-year terms, when realistic

If the concentrated client relationship allows for it, step 19 in the skill asks whether a multi-year agreement with a real early-termination fee is worth proposing. Raise it inside this block, not as a separate ambush conversation with the client. It doesn't replace the diversification work in Phase 3; it buys time on the clock while that work brings the percentage down.

## Checklist for keeping the block alive

- [ ] Fixed day and time exists on the calendar, confirmed, not just agreed to in conversation
- [ ] At least one prospecting or outreach action taken this week, logged with date and outcome
- [ ] Big client's planning activity reviewed for anything that reads differently than last month
- [ ] A recent deliverable and its documentation spot-checked
- [ ] Multi-year contract terms raised where realistic, logged whether or not the client said yes

A skipped week gets logged as skipped, the same way a skipped quarterly check does. A block that quietly stops happening is exactly the pattern this phase exists to prevent.

INSTRUCTIONS: delete this row before use. One row per client per quarter. Fill client_ttm_revenue and total_ttm_revenue with real trailing-12-month numbers, never estimates. Leave a cell blank rather than guess.,,,,,,,,,
quarter,client_name,client_ttm_revenue,total_ttm_revenue,client_pct_of_revenue,in_top5,top5_combined_pct,written_ceiling_single_client_pct,written_ceiling_deadline,notes
Q1 2026,Client A,,,,yes,,,,
Q1 2026,Client B,,,,yes,,,,
Q1 2026,Client C,,,,yes,,,,
Q1 2026,Client D,,,,yes,,,,
Q1 2026,Client E,,,,yes,,,,
Q1 2026,All other clients combined,,,,no,,,,

# Filled risk register example

A worked year of quarterly entries, so you can see what the register looks like once it's running instead of just what the blank columns mean. This is the owner from the reference point in references/thresholds.md, four quarters into her three-year ceiling.

Written ceiling: single client under 30% of trailing-12-month revenue, by Q1 2029. Set Q1 2026.

## Q1 2026 (baseline, the quarter the ceiling was set)

| Client | Trailing-12mo revenue | % of total | Top 5 combined |
|---|---|---|---|
| Riverside Logistics (largest) | $432,000 | 48% | |
| Dune Creative | $61,000 | 7% | |
| Marlowe & Pratt | $54,000 | 6% | |
| Sable Tax Group | $47,000 | 5% | |
| Harker Family Office | $39,000 | 4% | |
| All other clients combined | $267,000 | 30% | |
| **Total** | **$900,000** | | **70%** |

Both numbers over the ceiling range from references/thresholds.md before a written ceiling even existed. Action: Phase 3 opened same day. Tactic one (repricing tail clients) and tactic two (prospect profile shift) both started this quarter.

## Q3 2026

| Client | Trailing-12mo revenue | % of total | Top 5 combined |
|---|---|---|---|
| Riverside Logistics (largest) | $441,000 | 45% | |
| Dune Creative | $79,000 | 8% | |
| Marlowe & Pratt | $68,000 | 7% | |
| Sable Tax Group | $61,000 | 6% | |
| Harker Family Office | $55,000 | 6% | |
| All other clients combined | $286,000 | 29% | |
| **Total** | **$990,000** | | **72%** |

Single-client share down 3 points, not from Riverside shrinking, from the rest of the client base growing around it. Repricing conversations with Dune, Marlowe & Pratt, and Sable Tax landed in Q2. Still over ceiling. Playbook stays open.

## Q1 2027

| Client | Trailing-12mo revenue | % of total | Top 5 combined |
|---|---|---|---|
| Riverside Logistics (largest) | $448,000 | 39% | |
| Dune Creative | $101,000 | 9% | |
| Marlowe & Pratt | $89,000 | 8% | |
| Sable Tax Group | $77,000 | 7% | |
| New: Ferro Manufacturing | $64,000 | 6% | |
| All other clients combined | $362,000 | 32% | |
| **Total** | **$1,141,000** | | **69%** |

First new top-5 client from the prospecting shift (profile matched to Dune and Marlowe & Pratt, not to Riverside's size). Single-client share down another 6 points in two quarters. Still above the 30% ceiling; deadline is Q1 2029, two years out.

## Q3 2027

| Client | Trailing-12mo revenue | % of total | Top 5 combined |
|---|---|---|---|
| Riverside Logistics (largest) | $452,000 | 34% | |
| Dune Creative | $118,000 | 9% | |
| Ferro Manufacturing | $96,000 | 7% | |
| Marlowe & Pratt | $94,000 | 7% | |
| Sable Tax Group | $84,000 | 6% | |
| All other clients combined | $477,000 | 36% | |
| **Total** | **$1,321,000** | | **63%** |

34% and closing on the ceiling. Riverside's dollar revenue kept growing the whole time; the share dropped because everything else grew faster. That is the point of tactic one and tactic two run together rather than a single push: the biggest client doesn't have to shrink for the risk to shrink.

## What the quarterly comparisons actually track

Read down the "% of total" column for Riverside across the four rows: 48, 45, 39, 34. That is the number Phase 2 step 9 asks you to state and compare each quarter, against the written ceiling from Phase 1, not against the general benchmark range. The benchmark range only mattered once, the quarter the ceiling got set.

# Diversification playbook

Two tactics for bringing a client's share of revenue down. Neither one is "go find more clients." That advice is true and useless. This is the specific version of it.

## Tactic one: reprice and rescope the tail

Look at your risk register (examples/risk-register-template.csv or examples/filled-risk-register-example.md for the worked version). Below the top 5, you have a tail of smaller clients. Some of them haven't had a pricing conversation, or a scope conversation, in over a year.

That's the group this tactic works on. Not the biggest client, the smallest active ones.

1. Pull every client who hasn't had a rate or scope change in 12+ months.
2. For each one, check: are they getting more from you today than they were a year ago, at the same price? If yes, that's the case for a conversation.
3. Draft a repricing or scope-adjustment message using examples/repricing-email-draft-example.md as the starting point. One draft per client, not a form letter blasted to all of them.
4. Present every draft before it goes anywhere. Wait for an explicit yes on each one. A tail client is still a client; a clumsy repricing email can lose one faster than the concentration problem it was meant to fix.
5. Log the date and the outcome next to the register entry it relates to.

This works by growing the denominator, not shrinking the numerator. Your biggest client's dollar revenue doesn't have to go down for their percentage share to go down. Everyone else's revenue just has to go up faster.

## Tactic two: prospect the profile that got you here, not the biggest logo available

Look at where your healthiest existing accounts sit, size, industry, how you found them, what they pay. That's your working profile. Chase more of that.

The instinct when concentration feels uncomfortable is to go land one huge new client to "balance it out." That is how you got a concentration problem in the first place: a big account grows because it's the path of least resistance, and it keeps growing because nobody's tracking the share it's taking of the total.

1. Write down the profile of your 3 to 5 best mid-size clients: what they buy, what they pay, where they came from.
2. Point outbound and referral effort at that profile specifically, not at "growth" in general.
3. If you want draft outbound messages, use examples/prospect-outreach-draft-example.md, with the same approval discipline as examples/repricing-email-draft-example.md: draft, present, explicit yes, then send.
4. Treat any inbound lead that looks like it would become another Riverside-sized account (see references/thresholds.md's reference point) as a flag, not a win. Take the meeting. Just don't let it quietly become 40% of the business three years from now with nobody watching.

## Running the log

Every time either tactic touches a client or a prospect, record it: date, client or prospect name, which tactic, what happened. This log sits next to the register, not inside it. The register tracks the percentages. The log tracks the actions that are supposed to move them, so at the next quarterly check you can tell whether the plan is working or just being logged.

| Date | Tactic | Client / prospect | Action | Outcome |
|---|---|---|---|---|
| | 1 (reprice) or 2 (prospect) | | | |

# Repricing / rescope draft example

Template plus one filled example, for tactic one in references/diversification-playbook.md. Draft one of these per client. Present it and wait for an explicit yes before anything gets sent. Nothing here is pre-approved for sending.

## Template

Subject: [Client name] - checking in on scope and rate

Hi [name],

It's been [X months] since we last looked at pricing and scope on our end, and looking back, [specific change: "you've added Y" / "we've been doing Z on top of the original scope" / "usage has grown from A to B"].

I want to make sure our rate still matches what we're actually delivering. [Proposed change: new rate, new scope boundary, or both.]

Happy to talk it through if you want to hop on a call, or I can just walk you through the reasoning here if that's easier. Let me know what works.

[Your name]

## Filled example

The client: Dune Creative, a design retainer that started at a fixed monthly scope 14 months ago and has quietly grown without a rate change, pulled from the risk register's tail because no pricing conversation had happened in over a year.

Subject: Dune Creative - checking in on scope and rate

Hi Priya,

It's been about 14 months since we last looked at pricing and scope on our end, and looking back, the retainer's grown a fair bit past what we scoped originally. We started at 2 campaigns a month; the last two quarters it's been closer to 4, plus the packaging refresh in June that wasn't part of the original agreement.

I want to make sure our rate matches what we're actually delivering now. I'd like to move the retainer from $4,200 to $5,800 a month starting next billing cycle, which reflects the campaign volume increase and folds the packaging work in as a standing line item instead of a one-off.

Happy to talk it through if you want to hop on a call, or I can just walk you through the numbers here if that's easier. Let me know what works.

Marcus

## Before this goes out

- [ ] The specific change named (added scope, added volume, added deliverables) is real and something the client would recognize, not a generic "costs have gone up"
- [ ] The proposed rate or scope change is stated as a number, not "let's discuss a bump"
- [ ] The client has been told plainly, not softened into vagueness
- [ ] Owner has said yes to this exact draft, in this exact conversation, before it is sent
- [ ] Logged in the diversification playbook's running log once sent

# Prospect outreach draft example

Template plus one filled example, for tactic two in references/diversification-playbook.md. Draft one of these per prospect, matched to the profile from the owner's healthiest existing mid-size clients, not the biggest logo available. Present it and wait for an explicit yes before anything gets sent. Nothing here is pre-approved for sending.

## Template

Subject: [A specific reason for reaching out, not "quick question"]

Hi [name],

[How you found them or why you're reaching out: a referral, something specific about their business that matches your best clients, a shared connection.]

[One or two sentences on what you do, kept specific to the kind of work that fits their situation, not a general pitch.]

[A low-pressure next step: a short call, a few questions by email, whatever fits.]

[Your name]

## Filled example

The prospect: a mid-size logistics coordination firm found through a referral from Marlowe & Pratt, one of the register's healthy mid-size clients, matching the profile from Phase 3 step 15 rather than a large enterprise account.

Subject: A referral from Marlowe & Pratt

Hi Danielle,

Priya at Marlowe & Pratt mentioned you're looking at ways to tighten up scheduling across your regional routes, and thought our work together might be useful for you to see, since we've been doing something similar for her team.

We handle route and scheduling coordination for a handful of logistics and field-service businesses around your size, usually starting with a short audit of where the current process loses time before proposing anything.

Would a 20 minute call make sense sometime in the next couple weeks? Happy to just send over a short writeup instead if that's easier on your schedule.

Marcus

## Before this goes out

- [ ] The prospect matches the profile from Phase 3 step 15 (size, industry, or type similar to a healthy existing client), not the largest logo available
- [ ] The reason for reaching out is specific and true (a real referral, a real shared trait), not a generic opener
- [ ] The next step asked for is small and low-pressure
- [ ] Owner has said yes to this exact draft, in this exact conversation, before it is sent
- [ ] Logged in the diversification playbook's running log once sent, same as a repricing draft

Add up that client's trailing 12 month billings and divide by your total trailing 12 month revenue. If it comes out to 10 percent or more, that is the first line worth watching; 25 percent or more is the range where losing that client would genuinely hurt the business.

10 percent from a single client, or 25 percent combined across your top 5 clients, is the point accountants, business brokers, and lenders treat as a first warning; 25 percent from one client, or 50 percent combined across the top 5, is where it turns into a real dependency.

Set a written target for what that percent should be and by when, then work two tracks at once: reprice or restructure the relationship you already have, and keep a standing weekly pipeline block running so new client revenue is coming in to dilute the share, not just hoping the number improves on its own.

Yes. A buyer does not treat revenue tied to one flight-risk client as durable earnings. It gets discounted, turned into an earnout tied to that client staying, or made conditional on you personally staying on to keep the relationship, which is usually the opposite of what an owner wants from a sale.

Every quarter, on a fixed date, using a fresh trailing 12 month pull. Checking less often means you find out about the drift only when the client is already gone; checking on a single month or a calendar year instead of a trailing 12 month window gets you a number skewed by seasonality.

There is no universal number, but owners who have worked through this commonly land on a written target in the 25 to 30 percent range for their single largest client, set with a real deadline, usually two to three years out, and backed by an actual diversification plan rather than a hope.