Docket Article
How to Choose a Platform Company for a Buy-and-Build
Platform company selection criteria for a buy-and-build: what to test before diligence, what BCG, Bain and survey evidence show, and a weighted scorecard.
Summary
- 01Choose the platform for its ability to absorb acquisitions: management depth, systems and integration capacity matter alongside the quality of the underlying business.
- 02Public evidence helps identify candidates for a first call. Revenue quality, customer concentration, profitability, system scalability and valuation still require diligence.
- 03Historical return studies offer screening hypotheses, not expected returns. The BCG/HHL results rest on a small, mostly European sample, while Bain's comparison also has a thin evidence base.
- 04Apply knockout gates before scoring, cite evidence for each score and report row scores alongside the total. Keep an unknown option when evidence is missing.
- 05Platform sourcing needs depth on a few candidates; add-on sourcing needs broad coverage. Map the region and niche before closing to test whether enough credible add-ons exist.
Inside this article
Platform vs add-on: the definitions
A buy-and-build has two kinds of acquisition, answering different questions. The platform company is the first business the sponsor buys and the one that will absorb everything that follows. An add-on acquisition is bought by, and folded into, that platform.
Bain's definition sets a bar: buy-and-build is "an explicit strategy for building value by using a well-positioned platform company to make at least four sequential add-on acquisitions of smaller companies" [1]. BCG and HHL Leipzig, in the study most quoted on returns, use a looser one: deals "in which a PE firm buys a company and then builds on that 'platform' through add-on acquisitions" [2].
An add-on is judged on what it brings to the platform (customers, geography, capability). A platform is judged on whether it can carry the add-ons. A business can be an excellent add-on and a poor platform.
This article is a framework for screening, not investment advice.
Why the platform is chosen first
The platform sets three things that every later add-on inherits: the management team that will negotiate and integrate the deals, the systems the add-ons will be migrated onto, and the cost and governance base against which synergies are measured.
On the first, Bain's 2019 report sees a large benefit in starting with "a strong existing management team that has already demonstrated its ability to pull off acquisitions," and "it's also critical to determine if the platform is stable enough to support what the fund wants to do with it" [1]. The same report notes that "the platform company usually makes the add-on acquisitions—not the PE fund."
Survey evidence on investment choice in general points the same way. In Gompers, Kaplan and Mukharlyamov's 2015 NBER working paper, 65 private equity respondents ranked "business model / competitive position" highest (mean 4.6), followed by the management team (3.8), the sponsor's ability to add value (3.6) and valuation (3.5), with industry/market (3.2) and fit with the fund (2.3) last [3]. The same paper reports that almost 70% of the PE investors surveyed invest in the existing management team rather than recruiting their own before the deal, while 31% do recruit their own senior team.
Two implications follow. The business comes first even for a platform: a strong team cannot rescue a weak competitive position. And most sponsors inherit the team, so the pre-deal read on whether incumbents can run an acquisition programme matters more than a plan to replace them later.
Platform selection criteria
The table lists recurring qualities, what a buyer can check from public evidence before a CIM or management meeting, and what only diligence settles. The pre-diligence column tells you which candidates deserve a first call.
| Criterion | What it means for a platform | Checkable before diligence | Settled only in diligence |
|---|---|---|---|
| Business model and competitive position | Recurring or repeat demand, defensible local or niche position, room for organic growth | Service mix and end markets from the website; licences and certifications; tenure in the market from registry formation date | Revenue quality, churn, pricing power |
| Management depth | Leaders below the owner who can run sites, finance and integration | Named leaders beyond the founder; finance or operations hires visible in job postings; multiple locations already managed | Who actually makes decisions; key-person dependence |
| Back office and systems | Finance, HR, IT and reporting that can absorb another business | Job postings naming an ERP, payroll or field-service system; a controller or CFO role; centralised hiring | Close process, data quality, system licences and scalability |
| Integration capacity and record | Has bought and absorbed businesses before, or has a team that has | Prior acquisitions in press releases, registry mergers, rebranded locations, "formerly known as" pages | Whether the past acquisitions were actually integrated or left standalone |
| Standalone profitability | Earns enough on its own to fund integration costs and service debt without add-on synergies | Rarely visible for private companies; at most a revenue range | Quality of earnings, margins, working capital |
| Customer base quality | No single customer, payer or contract that can sink the plan | Government awards in public spending data; named anchor clients on the website | Customer-level revenue, contract terms |
| Valuation headroom | Price leaves room for add-ons to be accretive rather than paying up front for the whole plan | Nothing reliable pre-diligence | Purchase price, structure, leverage |
"Checkable" means you can find evidence, not that the evidence settles the question: a controller job posting shows intent, not a functioning close. The back-office row is also where thesis and platform meet. Under a back-office professionalization thesis, a platform that already runs a clean back office is worth more than one that needs building.
An add-on is judged on what it brings to the platform (customers, geography, capability). A platform is judged on whether it can carry the add-ons. A business can be an excellent add-on and a poor platform.
What the studies say
BCG and HHL Leipzig (2016)
The BCG/HHL headline: buy-and-build deals generated "an average IRR of 31.6% from entry to exit, compared with an IRR of 23.1% for standalone deals" [2].
Read the sample carefully:
- The authors began with 9,548 deals exited from 1998 through 2012, analysed 2,372 of them, and narrowed to 121 deals, including 48 with add-ons, "that provided the most complete data on IRR and on the levers employed to generate performance."
- About 90% of the 121 deals involved European companies, and no US companies were included, because, the authors note, US regulations do not require the disclosures that provide the necessary detail.
- A footnote states that "IRR figures are from our sample of 121 deals." The page does not separately state the sample behind each sub-result.
On platform size, the study defines small platforms as an enterprise value under $70 million, medium as $70 million to $290 million and large as over $290 million. Small platforms "generated an average IRR from entry to exit of 52.4%," compared with 34.6% for medium and 12.5% for large. The authors add that "the larger the platform, the more acquisitions it makes."
On the number of add-ons, deals with one or two add-ons averaged 35.5% IRR, against 23.1% for standalone deals and 19.9% for deals with more than two add-ons.
The study's profile of the most successful deals is a useful screening hypothesis: a deal with a PE sponsor with operational and buy-and-build experience, offers "an operationally efficient and scalable platform," is small or medium-sized, makes one or two add-on acquisitions, targets add-ons in its core industry and operates in "a low-growth, low-profitability, highly fragmented industry." It also reports that the contribution to added value from operational improvements "climbed steadily, from 18% in the 1980s to 48% in 2012."
How much weight to give it:
- Indicative, not a benchmark. The return results rest on 121 deals, 48 with add-ons; split by platform size, each bucket is a fraction of 48. Read the small-platform figure as "small platforms did well in this sample," not as an expected return.
- Old and mostly European. Exits ran through 2012; financing and competition for add-ons have changed since.
- A different population from Bain's. BCG's best results come from one or two add-ons; Bain's definition starts at four.
Bain (2019 and 2024)
In 2003, 21% of all add-on deals were at least the fourth acquisition by a single platform [1]. The 2024 report states that its 2019 report found more than 30% of add-on deals were at least a platform's fourth acquisition, and that "that number is close to 50%" today [4].
The 2024 report also compares, in a set of 44 buy-and-build deals made between 2010 and 2019, those relying on multiple arbitrage had an average multiple on invested capital of 1.4x, while those with "a strategic rationale driving accelerated organic growth or meaningful margin improvement had an MOIC of 2.2x" [4]. Forty-four deals is a thin base: a direction, not a rule, matching the report's view that "success ultimately relies on M&A expertise, thoughtful integration."
Academic work
Academic work on platform selection is thinner. Borell and Heger's 2013 ZEW discussion paper on buy-and-build strategies (ZEW Discussion Paper 13-094; also SSRN 2356191) reports in its abstract that investors pair platforms with lower capacity utilization and lower returns with add-ons that have higher utilization and higher returns, and that the strategies improve profitability only for firms with increasing industry-adjusted utilization. That is a pairing finding, not a selection rule, but it supports testing a platform's spare capacity and an add-on's utilization together. Kim's 2025 working paper on hospital buyouts [5], not yet peer reviewed, classifies targets into platforms and add-ons; its abstract reports that platform hospitals "experience significant and persistent profitability gains, whereas local add-ons exhibit severe contractions in patient volume and profitability," which the author attributes to deliberate reallocation toward the platform. One industry and one working paper, but the point travels: platform and add-ons play different roles.
Where buy-and-build gets harder
Bain's 2024 report explains what changed. "The cost of financing today is approximately double what investors were modeling in 2020," higher rates are "putting downward pressure on asset prices," and "in the absence of a multiple-arbitrage boost, those that had flimsy strategic rationale from the start are in a particularly difficult place" [4]. The same report observes that "more sponsors are hustling to try to complete buy-and-build strategies than are trying to launch new ones."
Three consequences for platform selection:
- Standalone economics matter more. If the exit cannot rely on a higher multiple for the combined company, the platform has to earn its return through organic growth and margin.
- Integration capacity is no longer optional. Four add-ons with no integration team means four sets of systems and pricing. The add-on count rises; the synergies do not.
- Add-on supply is contested. When close to half of add-on deals are a platform's fourth or later acquisition, platforms compete for the same targets. Management with relationships among likely sellers has an advantage the financials do not show.
Platform red flags
Reasons to downgrade a candidate as a platform, even if it remains a good add-on:
- Founder-only management. Every customer relationship, pricing decision and hire runs through one person.
- Customer concentration. One customer, payer or contract whose loss would break the debt case. Rarely disclosed, so test it early.
- Improvised systems. Multiple accounting files, spreadsheet scheduling, no single payroll. Each add-on makes this worse.
- No integration record. Past acquisitions still run under separate names with separate books are not evidence of integration capacity.
- Owner-held licences or relationships. If an essential licence, certification or referral relationship sits with an individual, it may not transfer.
- A thesis that only works through multiple expansion. If the case collapses when the exit multiple equals the entry multiple, the platform is not carrying the plan.
Worksheet: platform scorecard
- 01Apply knockout gates
Exclude candidates if the owner will not stay through transition and has no named successor, an essential operating right cannot transfer, or revenue cannot carry finance and integration.
- 02Set weights and cite evidence
Adjust the template weights to the thesis and record why. Score each criterion from weak to strong, citing its evidence.
- 03Calculate the weighted score
Sum weight multiplied by score, then divide by 5 to place the result on a 0-100 scale.
- 04Show rows and unknowns
Report row scores with the total so the drivers remain visible. Keep missing evidence unknown instead of assigning a default score.
A candidate that fails any gate is not scored as a platform.
A template. Weights put business model and management highest, in line with the survey ranking above; adjust them to your thesis and record why. Scores run 1 (weak) to 5 (strong), each citing its evidence. Weighted score = sum of (weight × score) ÷ 5, giving a 0-100 scale.
| Criterion | Weight | Evidence required for a 4 or 5 | Evidence source field |
|---|---|---|---|
| Business model and competitive position | 20 | Recurring or contracted demand; organic growth visible in locations or hiring | URL, retrieved date, quote |
| Management depth | 20 | At least two named leaders below the owner running functions or sites | URL, retrieved date, quote |
| Back office and systems | 15 | Named finance lead; single ERP/payroll evidenced | URL, retrieved date, quote |
| Integration capacity and record | 10 | At least one prior acquisition absorbed under one brand and one set of systems | URL, retrieved date, quote |
| Standalone profitability | 15 | Profitability evidenced in a document, not inferred | Document, date |
| Customer base quality | 10 | No customer above the mandate's concentration limit | Document, date |
| Valuation headroom | 10 | Expected price leaves add-ons accretive at the entry multiple | Model reference |
Before scoring, apply knockout gates. A candidate that fails any gate is not scored as a platform:
- Owner unwilling to stay through a transition period, with no successor named.
- A licence, permit or contract essential to operations that cannot transfer.
- Revenue below the floor at which the business can carry a finance and integration function.
Illustrative comparison
The two companies below are fictional, and every figure is invented for illustration.
Northgate Mechanical (fictional), a commercial HVAC service business with four branches, a general manager and a controller below the owner, one prior acquisition rebranded and moved onto its field-service system, and no single customer above the mandate limit. Its likely price is full.
Pinecrest Building Services (fictional), a single-site business with stronger margins, a well-spread customer list and a lower likely price, but the owner runs operations and finance personally and the books are kept in two separate accounting files.
| Criterion | Weight | Northgate score | Northgate weighted | Pinecrest score | Pinecrest weighted |
|---|---|---|---|---|---|
| Business model and competitive position | 20 | 4 | 80 | 4 | 80 |
| Management depth | 20 | 4 | 80 | 2 | 40 |
| Back office and systems | 15 | 3 | 45 | 2 | 30 |
| Integration capacity and record | 10 | 4 | 40 | 1 | 10 |
| Standalone profitability | 15 | 3 | 45 | 4 | 60 |
| Customer base quality | 10 | 3 | 30 | 4 | 40 |
| Valuation headroom | 10 | 2 | 20 | 4 | 40 |
| Total (sum ÷ 5) | 100 | 68 | 60 |
Northgate scores higher despite weaker economics and a fuller price, because it has the people and systems add-ons depend on. Pinecrest wins every economic row and loses every capacity row: not a worse business, but a better add-on.
Report row scores alongside the total, since a weighted total hides what drove it. Scores without a cited source are opinions; keep an "unknown" option rather than defaulting a missing row to 3.
Sourcing platforms vs sourcing add-ons
Platform and add-on origination are different research problems; one list tends to serve neither.
| Platform search | Add-on search | |
|---|---|---|
| Universe | Narrow: a handful of candidates large and organised enough to carry a programme | Broad: every business in the platform's footprint that fits its service line |
| Main question | Can this company absorb others? | Does this company add customers, geography or capability to the platform? |
| Disqualifiers | Thin management, improvised systems, no integration record | Overlap with the platform's existing customers, incompatible service model |
| Evidence emphasis | People, systems, prior deals | Location, service mix, size band, owner situation |
| Output | A short, deeply researched list | A long, mapped list ranked by fit and timing |
The platform search needs depth on a few records; the add-on search needs coverage. A market map of the platform's region and niche, built before the platform closes, shows how many add-ons actually exist and whether "at least four" is realistic. It also tests the platform thesis: if the map turns up too few credible add-ons, the platform is being bought for a programme that cannot be executed.
Docket's Triage, Scout and Audit agents turn a sponsor's platform and add-on criteria into questions answered from retained, quoted sources.
- Ask whether the company can absorb others.
- Focus on a handful of candidates large and organised enough to carry a programme.
- Prioritise evidence about people, systems and prior deals.
- Disqualify thin management, improvised systems and no integration record.
- Ask whether the company adds customers, geography or capability to the platform.
- Cover every business in the platform's footprint that fits its service line.
- Prioritise location, service mix, size band and owner situation.
- Disqualify customer overlap and an incompatible service model.
Map the platform's region and niche before closing to test how many credible add-ons exist and whether the programme can be executed.
Sources
- Buy-and-Build: A Powerful PE Strategy, but Hard to Pull Off (Global Private Equity Report 2019) — Bain & Company (2019), https://www.bain.com/insights/buy-and-build-global-private-equity-report-2019/, accessed 2026-10-28
- Building a Stronger Buy-and-Build (Global Private Equity Report 2024) — Bain & Company (2024), https://www.bain.com/insights/building-a-stronger-buy-and-build-global-private-equity-report-2024/, accessed 2026-10-28
- Private Equity's Power of Buy and Build — Boston Consulting Group with HHL Leipzig Graduate School of Management (2016), https://www.bcg.com/publications/2016/private-equity-power-of-buy-build, accessed 2026-10-28
- Gompers, Kaplan & Mukharlyamov, What Do Private Equity Firms Say They Do? — NBER Working Paper 21133 (2015), https://www.nber.org/papers/w21133.pdf, accessed 2026-10-28
- Kim, When Buy-and-Build Breaks: Serial Private Equity Acquisitions in the Hospital Industry — SSRN working paper 5416094 (2025), abstract read via Crossref, https://doi.org/10.2139/ssrn.5416094, accessed 2026-10-28
- Borell & Heger, Sources of Value Creation Through Private Equity-Backed Mergers and Acquisitions: The Case of Buy-and-Build Strategies — ZEW Discussion Paper 13-094 (2013), abstract read on RePEc, https://econpapers.repec.org/RePEc:zbw:zewdip:13094; SSRN 2356191, https://doi.org/10.2139/ssrn.2356191, accessed 2026-10-28
External Sources (5)
Published by
Docket Research
Docket provides deal-origination research software and managed research for private equity teams, screening companies against a mandate and showing the evidence behind each finding.
Editorial note
Researched and written by Docket Research with AI-assisted research tools; sources are cited inline and checked before publication. Report an error: https://docket.capital/evidence-standards#corrections.