# Customer Concentration Risk: How to Measure It in Pipedrive and What a Small Sales Team Should Do About It

https://salesbond.app/blog/customer-concentration-risk-pipedrive/
Published: 2026-10-07 · Updated: 2026-10-07

Customer concentration risk is the share of your revenue that depends on one customer, or on a handful. To measure it, add up won deals for the last 12 months by organization and divide the biggest organization’s total by the total of everything. If one customer is above 10% of ARR, investors start raising eyebrows; above 20%, many walk away, and a top-5 share above 40% counts as high risk ( [Monetizely](https://www.getmonetizely.com/articles/customer-concentration-risk-what-it-is-why-it-matters-and-how-to-measure-it)). Those are finance thresholds. For a sales team of 3 to 15 people, the same number means something more practical: you need a second line of contacts inside that account, a replacement pipeline sized to cover it, and a discount rule before the renewal call, not after.

## “He pays for a quarter of our year”

Wren runs an eight-person company. Forty customers bought something in the last twelve months. On a Tuesday in September a buyer at her largest customer forwards an email: their procurement team is “reviewing vendors”, would she mind filling in a questionnaire. Nothing dramatic. No complaint, no competitor named.

She asks her two reps how much that customer is worth. Soren says “a lot, maybe fifteen percent”. Quill says “twenty, with the reorders”. Neither of them has looked. Wren opens Pipedrive, scrolls the won deals for that organization, adds them in her head, and stops. It is closer to a quarter. The next two accounts together make it 46% of the year sitting with three logos (example, numbers illustrative).

The questionnaire she can handle. What bothers her is that the number lived in nobody’s head. The team had been treating that customer as “the easy one”: reorders come in by email, nobody books a meeting, the only contact is one buyer who has just told them, politely, that he is being asked to shop around.

Every small company we talk to has a version of this. The share is rarely a secret in the finance spreadsheet. It is a secret in the sales team, where the decisions that change it get made. So this article covers how to get the number out of the CRM, which thresholds actually matter at your size, and what a seller does differently on Wednesday because of it.

## How to measure customer concentration in Pipedrive

Customer concentration (%) = one customer’s revenue ÷ total revenue.

Finance people call this the customer concentration ratio. You need two inputs: revenue from won deals in the last 12 months, by organization, and the total. We report two shares at every quarterly review: top-1 and top-3. Top-1 tells you about a single phone call. Top-3 tells you about a bad quarter.

Pipedrive gives you the raw material in two places, and neither of them is a ready-made concentration report. There is no button for this. Here is the route we use.

1. **Start in Insights with a deal performance report.** It gives an overview of how many deals were started, won and lost, and lets you change the view by, measure by and segment by options ( [Pipedrive: deal performance report](https://support.pipedrive.com/en/article/insights-reports-deal-performance) ). The default filter is “Deal created > is > this year” ( [same page](https://support.pipedrive.com/en/article/insights-reports-deal-performance) ), which is the wrong cut for this question: a deal created in December and won in March belongs to the year it was won. Change the filter to won deals, and the period to the last 12 months, not the calendar year. A rolling year does not reset the number to zero on January 1st.
2. **Try segmenting by organization, measured by value.** Pipedrive’s help page does not list organization among the segment options, so check what your report actually offers. If it works, you get the per-customer totals right there. If it doesn’t, go to the deals list view, apply a filter for won deals over the same 12 months ( [Pipedrive: bulk editing and filtering](https://support.pipedrive.com/en/article/bulk-editing-and-filtering) ), and bring the organization and value columns into a spreadsheet. Sum by organization, sort descending. It is ten minutes the first time and two minutes every quarter after that.
3. **Write down four numbers**: total, top-1, top-3, and the date you counted. Keep them where the pipeline review happens, not in a finance folder.

Two honesty checks before you trust the result.

First, repeat orders. A customer who buys twelve times a year is twelve won deals, not one, and that is exactly how you want it: the share is about the organization, so the sum is what counts. The trap is the opposite one. If reps log reorders as updates to an old deal, or as activities with no deal at all, the organization looks smaller than it is. Wren’s “easy” customer was under-counted for that reason. If you track repeat customers at all, the [customer health score you build from won deals](https://salesbond.app/blog/customer-health-score-pipedrive/) depends on the same discipline, so fixing it pays twice.

Second, currencies and duplicates. Sum in one currency, the one your reporting runs in. And check the organization list for the same company under two names before you sort, a duplicate splits one big customer into two medium ones and hides the problem you are looking for.

![Single bar of 12 months of won deals by organization: the top customer and the next two are almost half of revenue, 37 others share the rest](https://salesbond.app/wp-content/uploads/2026/10/customer-concentration-risk-pipedrive-2.png)

## Thresholds: what counts as a warning sign

Most of the thresholds you will find online were written for investors looking at a SaaS company. They are still useful, because they tell you how someone pricing your business reads the number. Here is what two published sources say, and what we use ourselves.

| Measure | Threshold | Who says so |
| --- | --- | --- |
| One customer, share of annual recurring revenue | above 10%: “most institutional investors start raising eyebrows”; above 20%: “many will walk away regardless of your growth rate” | [Monetizely](https://www.getmonetizely.com/articles/customer-concentration-risk-what-it-is-why-it-matters-and-how-to-measure-it) |
| One customer, or the top five, share of total revenue | one customer above 10%, or the top five above 25%: “potential red flags” | [Wall Street Prep](https://www.wallstreetprep.com/knowledge/customer-concentration/) |
| Top 5 customers, share of revenue | under 20% low risk; 20 to 40% moderate; above 40% high | [Monetizely](https://www.getmonetizely.com/articles/customer-concentration-risk-what-it-is-why-it-matters-and-how-to-measure-it) |
| Valuation effect | companies with high concentration “often receive valuation multiples 20-30% lower” | [Monetizely](https://www.getmonetizely.com/articles/customer-concentration-risk-what-it-is-why-it-matters-and-how-to-measure-it) (their estimate) |
| Top-1 and top-3, team of 3 to 15 | top-1 above 15% or top-3 above 35%: the account gets a named second owner and a replacement-pipeline target this quarter | our rule of thumb, not a published benchmark |

Our rule is deliberately looser on top-1 than the investor view and stricter in what it demands. A company with eight people and forty customers will almost always have one account at 15%; that is how small companies grow, one anchor at a time. The number alone doesn’t worry us. What worries us is an anchor with one contact, no open deal, and nobody whose job it is to notice.

Two cases where we would move the line. If your average deal is small and your customers are many, a single customer at 10% is already strange; it probably means one buyer is quietly acting as a distributor, and you want to know. If you sell three large projects a year, every customer is above a third and the measure says nothing. Look at teh next-twelve-months pipeline instead, where the question becomes [how much of the forecast hangs on one deal](https://salesbond.app/blog/one-big-deal-forecast-pipedrive/).

## What a sales team changes once it knows

Three things. All of them are sales work, not finance work, and all of them can start the week you count.

**A second and third line of contacts in the big account.** One buyer who likes you is a relationship. One buyer who likes you and is the only person in the building who knows your name is a risk with a face. For every organization above your top-1 line, the account should have at least two more people logged on its deals: the person who signs, and the person who feels the pain when your product stops working. In Pipedrive that means [adding them as participants on the deal rather than leaving them in a note](https://salesbond.app/blog/pipedrive-switch-deal-contact-participants/), and being honest about [which of them is actually your champion](https://salesbond.app/blog/sales-champion-pipedrive-deals/). The rule we give reps: if your contact at the top account went on leave for three months, who would you call? If the answer is “nobody”, that is the next activity.

**A replacement-pipeline target.** Ask one question: if the top customer stopped buying tomorrow, how much new business would we need to win this quarter to end the quarter where we planned? Take the top customer’s share of last year’s revenue, scale it to a quarter, and divide by your win rate for new business. That is the open pipeline you need to be carrying, on top of the normal plan, before you can say the risk is covered. Wren’s top account was a quarter of the year; at a one-in-three win rate on new logos, her team needs open new-business pipeline worth roughly three quarters of a normal quarter’s revenue, just to neutralize that one account (example, numbers illustrative). Written down it changed what Soren prospected on Thursdays. Same arithmetic as a [pipeline coverage ratio](https://salesbond.app/blog/pipeline-coverage-ratio/), aimed at one risk instead of the whole plan.

**A discount rule for the large customer, agreed before the call.** The customer who is 24% of revenue knows it, or will find out at renewal. If the first time your team discusses “how low would we go” is on the call, you will go lower than you meant to. Decide now: what is the floor, who approves below it, and what you ask for in return (a longer term, a second product, an introduction to another site). If you already run a [discount policy](https://salesbond.app/blog/discount-policy-sales-teams/), add one line to it for accounts above the top-1 threshold. Be as generous with your best customer as you like. Just make each concession a decision rather than a reflex.

If the big account has a plan at all, keep it to one page; we have written separately about a [key account plan for a small team](https://salesbond.app/blog/key-account-plan-small-sales-team/), so we will not repeat it here.

A disclosure, since we are on the subject of watching one account closely: we build SalesBond, a sales management layer on top of the CRM a team already uses; we started with Pipedrive. When a large customer has an open deal, Deal Health runs more than 20 rules against it and tells you the state, the reason and a recommended action, so a stalled renewal at your top account does not sit quietly for six weeks. The share of revenue itself is not something the product computes; that is the spreadsheet above, done by hand once a quarter. If you want to see what your CRM says about the deals that matter most: [Check your CRM with SalesBond](https://salesbond.app/).

## What not to do: diversify by neglect

There is a reading of “concentration risk” that goes: our biggest customer is too big, so let us spend less time on them and more on everyone else. We think that is backwards, and we have watched it cost a company it’s anchor.

A customer at 24% of your revenue is, almost by definition, your best fit. They buy repeatedly, they pay, they have expanded. You lower their share by winning more customers who look like them, never by serving them worse. Reducing concentration is a numerator-and-denominator problem, and the only healthy lever is the denominator.

So the rule is: the big account gets more attention, not less, and the new attention goes to breadth inside the account (more contacts, more sites, more products) rather than to defending a single buyer. Meanwhile the prospecting plan from the previous section grows the rest. If a year from now the top account is the same size and the share has dropped from 24% to 18%, that is a good year. If the share dropped because the account shrank, you did not reduce risk. You realized it.

One more thing not to do: do not confuse this number with [revenue at risk in the open pipeline](https://salesbond.app/blog/revenue-at-risk-sales-pipeline/). That one is about deals you have not won yet and might lose. Concentration is about the revenue you already won and which door it came through. Both matter, and the fixes are different.

## Five minutes with the number, once a quarter

The whole thing fits at the top of a quarterly pipeline review. Four numbers on one line: total won in the trailing twelve months, top-1 share, top-3 share, and the delta since last quarter. Then three questions, asked out loud with the the reps in the room.

Who are our second and third contacts at each account above the line, and when did we last talk to them? Is the replacement pipeline for the top account open and moving, or is it a target on a slide? Has anyone offered the big customer a discount since we last met, and was it the rule or a reflex?

Wren’s version of this took four minutes the second time and three the third. By the fourth quarter the top account was still a quarter of the year, there were four names on its deals instead of one, and the questionnaire from September had turned into a two-year renewal with a second site. Nobody at her company would describe that as managing concentration risk. They would say they finally knew what the customer was worth, and acted like it.