Key takeaways
- A $17M mechanical services business with 14 commercial accounts ran job costing for the first time and found three of the top five revenue accounts in the bottom quartile for contribution margin, one generating a negative contribution after overhead allocation. None of this was visible in the blended P&L.
- The margin illusion compounds when businesses price new work based on blended margins, systematically underpricing complex high-labor jobs and overpricing simple high-volume ones, making the distribution worse with every new contract signed.
- Service businesses that repriced or exited bottom-quartile accounts before a sale process achieved 3–6 percentage point gross margin improvements over 12–18 months, translating directly to EBITDA expansion at a 1:1 ratio.
- Reprice before exiting, an account that accepts an 8% increase is a good customer; one that refuses tells you its actual loyalty level before you lose the revenue in a sale year.
- Include the customer-level margin analysis in the management presentation, and it is one of the strongest management quality signals available for a service business, and the buyer will run it regardless.
In this article
Operating diagnosis
Operator Checklist
- Name the metric, process, or decision this issue affects.
- Assign a single owner with authority to change the process.
- Pull the last 12-24 months of data and identify the pattern, not just the latest month.
- Choose one corrective action that can be tested in the next 30 days.
- Review the result in the next management cadence and document the decision.
Revenue ≠ profit
The first principle of customer-level economics
Top 20% of customers
Often generate 80%+ of actual margin
Bottom 20% of customers
Often generate negative contribution in full-cost models
Job costing
The discipline that makes customer economics visible
McKinsey's pricing research consistently finds that the top 20% of customers by contribution margin generate 80–90% of a service business's actual profit, while the bottom 20% of customers consume 5–15% of gross margin, creating a 25–35% swing in profitability that is invisible in blended P&L reporting.
In PE-acquired service businesses, the Day 1 customer margin analysis conducted by operating teams finds that 1 in 3 large-revenue accounts (top 5 by revenue) is in the bottom half of the business by contribution margin, a finding that surprises management teams who tracked revenue but not full-cost profitability.
Businesses that systematically repriced or exited bottom-quartile accounts before a sale process achieved average gross margin improvements of 3–6 percentage points over 12–18 months, translating directly into EBITDA expansion at a 1:1 ratio (Bain Customer Profitability Research 2024).
Revenue is the metric most middle market founders know best, which customers are biggest, how revenue has grown, where the pipeline sits. Gross margin by customer is the metric PE buyers model in the first week of ownership. The gap between the two is where the most common margin surprise in post-close operations lives.
Founders who've grown a strong top-line reasonably define revenue quality by longevity and relationship stability, a customer who has been with the business for seven years is a good customer in every observable sense. What PE buyers surface in the first 30 days is that the seven-year customer who gets the most service accommodations, the most informal discounts, and the most founder attention is often the least profitable account in the portfolio. IC memos for service business acquisitions regularly include a Day 1 task: customer-level margin analysis. That analysis drives the first repricing decisions of the new ownership period.
A $20M revenue business with 18% EBITDA margin is not a business where every customer generates 18% contribution. In most middle market service businesses, the actual distribution looks dramatically more skewed: a small cluster of accounts generates strong margins, a larger group generates acceptable margins, and a meaningful tail generates margins near zero or below when labor, overhead, and service costs are fully allocated.
Why most businesses do not know their customer-level economics
Most middle market accounting systems are configured to track revenue, direct costs, and overhead at a company level. They are not configured to allocate direct labor, materials, and indirect overhead to individual customers or jobs. The result: the P&L shows a blended margin that obscures the customer-level distribution.
The businesses that do have customer-level economics, through job costing systems, activity-based costing, or manual allocation, routinely discover that their mental model of customer profitability is wrong. The largest customers are not always the most profitable. Long-tenure customers are not always the most profitable. High-volume accounts that required significant service customization are often the least profitable after full cost allocation.
A $17M mechanical services business had 14 commercial accounts representing 60% of revenue.
The owner's mental model ranked them by revenue size and relationship tenure. When a proper job costing analysis was run for the first time, allocating direct labor hours, materials, and a reasonable equipment overhead rate to each job, three of the top five revenue accounts were in the bottom quartile for contribution margin.
One was generating a negative contribution after overhead allocation. None of this was visible in the blended P&L.
The margin illusion compounds when businesses price new work based on blended margins rather than job-level economics. A business that prices new contracts to achieve a 35% gross margin in aggregate may be systematically underpricing complex, high-labor jobs and overpricing simple, high-volume ones, and making the margin distribution worse with every new contract signed.
How to build a basic customer-level margin analysis
Customer Margin Analysis, Build Sequence
Step 1: Export job/invoice-level data
Pull 24 months of transaction data with customer, revenue, direct materials cost, and direct labor hours per job or invoice
Step 2: Assign a labor cost rate
Apply a fully-loaded labor rate (wages + benefits + workers' comp) per hour to convert labor hours to direct labor cost
Step 3: Allocate overhead
Choose a simple allocation basis, direct labor hours or revenue percentage, and allocate operating overhead (equipment, facilities, supervision) to each job
Step 4: Calculate contribution margin
Revenue minus direct materials, direct labor, and allocated overhead per job; aggregate to customer level
Step 5: Sort and segment
Rank customers by contribution margin percentage and dollar value; identify the top quartile, middle two quartiles, and bottom quartile
Step 6: Identify the actions
For bottom-quartile accounts: reprice, re-scope, or exit. For top-quartile accounts: protect, deepen, and replicate.
Operating workflow scan
Turn the issue in this article into a ranked AI workflow roadmap with readiness gaps and estimated time savings.
Find the first workflow →The M&A connection
PE buyers run this analysis in the first 30–60 days of ownership. Sellers who have already done it, and can present customer-level profitability data in the management presentation or <a href="/insights/what-is-a-data-room-ma" class="subtle-link">data room</a>, are presenting a qualitatively different level of management sophistication than sellers who present only consolidated P&L data.
Scroll to see more →
More practically, sellers who run the analysis 18–24 months before a process can address the bottom-quartile accounts before valuation is set. Exiting or repricing low-margin accounts improves the blended EBITDA margin, which directly improves the purchase price at a fixed multiple. A service business that improves its gross margin from 34% to 38% through customer portfolio rationalization adds meaningful EBITDA, and that EBITDA is the basis of the valuation.
The repricing and exit sequence: how to address bottom-quartile accounts before a transaction
Identifying low-margin accounts is the analysis step. Doing something about them is the operational step, and it requires a sequenced approach that protects revenue while improving the margin distribution. The worst outcome is exiting low-margin accounts abruptly, taking a revenue hit, and entering a transaction process with a declining revenue story. The best outcome is repricing accounts that value the relationship, exiting only those that reject repricing, and arriving at the process with a cleaner margin profile and a stable or growing revenue line.
Repricing and Exit Sequence for Bottom-Quartile Accounts
Step 1: Segment the bottom quartile by repricing probability
Before any outreach, assess each account: long tenure (likely to accept repricing), price-sensitive history (likely to resist), relationship is primarily with the founder (higher retention risk at pricing conversation). Do not treat all low-margin accounts the same.
Step 2: Build account-specific briefs
For each repricing conversation, prepare a brief showing: (a) cost inflation since the last price adjustment, (b) the current margin versus the business's floor, (c) the specific increase requested as a percentage. Data-backed conversations close at significantly higher acceptance rates than informal requests.
Step 3: Lead with the highest-probability accounts
Start repricing conversations with accounts most likely to accept. Early acceptances build the institutional confidence that the repricing program is working and generate momentum for harder conversations.
Step 4: Give a defined timeline
Accounts being repriced should receive 30–60 days notice before the new pricing takes effect. This is standard practice and reduces the emotional resistance to the conversation.
Step 5: Move to exit for rejections
Accounts that reject repricing after a documented second conversation are demonstrating their actual price sensitivity. Exit is the right outcome for accounts that cannot be served profitably at market rates. Execute exits over 60–90 days to allow for backfill.
Step 6: Measure and document
Track acceptance rate, revenue impact, and EBITDA improvement by quarter. This documentation becomes a diligence asset: a management team that ran a systematic repricing program with measurable results is demonstrating operational discipline.
The revenue math on exiting unprofitable accounts surprises most founders. A $600K account generating 4% contribution margin contributes $24K of EBITDA. Eliminating the account eliminates $600K of revenue but also eliminates $576K of cost. The net EBITDA impact is near zero, and the freed capacity can be deployed to a new account at market margin. At 6x EBITDA, there is no transaction value being left on the table by exiting a 4% margin account, and there may be value being created by freeing the management and operational capacity.
Scroll to see more →
Common mistakes founders make on customer margin analysis.
Frequently asked questions
Why do large customers sometimes generate the worst margins?
Large customers often negotiate the deepest discounts, require the most service customization, and generate the most overhead consumption per dollar of revenue, while the revenue scale creates a false impression of profitability. The businesses where this is most pronounced are those that grew large accounts through competitive pricing without ever building a formal allocation of their full service cost.
How does customer margin analysis differ from the standard P&L?
The standard P&L shows blended company-level margins. Customer margin analysis allocates direct costs (labor, materials) and overhead to individual customers or jobs, revealing the contribution margin at a granular level. Most businesses discover a much wider distribution than their blended margins suggest, with a small number of highly profitable accounts and a meaningful tail of near-zero or negative-margin accounts.
Work with Glacier Lake Partners
Request a Margin Analysis and Profitability Review
Most useful when preparing for a transaction or when EBITDA is growing slower than revenue.
Start a Conversation →Operating workflow scan
Find the reporting or execution workflow worth automating first.
Turn the issue in this article into a ranked AI workflow roadmap with readiness gaps and estimated time savings.
Find the first workflow →Research sources
Disclaimer: Financial figures and case-study details in this article are anonymized, composite, or representative examples based on middle market operating situations, and are not guarantees of outcome. Statistical references are drawn from cited third-party research; individual transaction and operational results vary based on business characteristics, market conditions, and deal structure. This content is for informational purposes only and does not constitute legal, financial, or investment advice. Consult qualified advisors for guidance specific to your situation.
Explore adjacent topics
