How to source add-on acquisitions for a buy-and-build strategy

Most buy-and-build strategies do not fail at the closing table. They fail earlier, at the point where the team realises the target universe is thinner than the investment committee was told. The thesis assumed twenty credible add-ons in a fragmented niche. The list that actually survives a first screen has six names, three of them already owned by a competitor.

Add-on sourcing is a coverage problem before it is a process problem. Add-ons have become the dominant form of private equity buyout activity in the US, where roughly three of every four buyouts are add-on acquisitions (Wall Street Prep). That means the obvious, intermediated candidates in any given niche are seen by everyone at once. The differentiation sits in finding the companies that never reach a banker's process, and in screening them against a thesis that is specific enough to be falsifiable.

In this article you'll learn:

• Why the size of your add-on universe, not your outreach speed, decides whether a buy-and-build thesis holds
• How to translate a value-creation thesis into screening criteria that a data set can actually run
• Where add-on candidates come from, and what each sourcing channel realistically delivers
• How to score candidates on integration reality instead of financials alone
• How to sequence a pipeline so the first add-on makes the next three easier
• The recurring reasons buy-and-build pipelines run dry after the second deal

Why add-on sourcing is a different discipline from platform sourcing

Platform sourcing looks for one company that fits a fund thesis. Add-on sourcing looks for a repeatable supply of companies that fit an existing operating business. The two are not variations of the same exercise.

Three differences drive everything else. First, the targets are smaller, often owner-operated, and frequently invisible in standard databases because they file minimal accounts and have no press coverage. Second, fit is defined by the platform, so the criteria are operational (customer overlap, service line, geography, systems) rather than purely financial. Third, you need volume: a thesis that calls for four add-ons needs a credible universe several times that size, because most owners are not sellers in your holding period.

The practical consequence: an add-on search that starts from a list of known competitors is almost always too narrow. It has to start from the market structure.

Start with the value-creation thesis, not with a target list

Before any screening, write down what each add-on is supposed to do for the platform. Vague answers ("scale", "synergies") produce vague criteria and unusable longlists. Specific answers produce filters.

Most buy-and-build theses reduce to one of a small number of mechanisms:

The right-hand column is the part that matters. It is the only place a thesis becomes something a data set can be queried against.

Define the target universe before you define the target list

The single most useful check in buy-and-build is a universe count. How many companies exist that plausibly match the profile, before any quality filter? If the thesis needs four add-ons and the universe holds fifteen companies, the strategy depends on convincing a quarter of an entire niche to sell. That is a red flag worth raising before the platform is bought, not after.

A workable rule of thumb: the universe should hold several times the number of companies you plan to acquire. Getting to that number requires defining the market by what companies actually do, not by the industry code they were assigned. Classification codes systematically miss specialised SMEs, because a company doing precision coating for medical devices may be filed under generic metal processing, or under nothing meaningful at all.

This is where most longlists silently lose half their coverage. A search built on NACE, WZ, SIC or GICS codes returns the companies that were classified correctly, and quietly omits the rest. Searching by business description, product language, customer type and capability finds the companies that never had the right code in the first place.

Where add-on candidates actually come from

Every channel has a different yield, a different level of competition, and a different amount of work per name. A realistic pipeline uses several at once rather than relying on the one that is easiest to run.

The combination that works in fragmented markets is straightforward: build the complete universe systematically, then use management's knowledge to rank it, then approach the owners directly. The order matters. Ranking a list that was never complete just formalises the blind spot.

Screen for integration reality, not only for financials

Add-ons are usually acquired at a lower multiple than the platform, which makes the financial case look easy. The value is realised only if the business can be integrated, and integration risk is visible in the screening data if you look for it.

1. Owner dependence: is the founder the commercial relationship, or is there a management layer? This drives both the retention structure and the realistic post-close timeline.
2. Customer concentration: a target with one dominant customer transfers a risk into the platform rather than diversifying it.
3. Systems and data maturity: an add-on with no usable financial reporting will absorb months of platform capacity.
4. Overlap versus addition: real overlap creates cost synergies and integration pain at once. Pure addition is easier to integrate and delivers less.
5. Succession situation: in fragmented SME markets, an approaching succession is often the single strongest predictor of willingness to talk.

Scoring candidates on these dimensions before outreach changes the conversation with the investment committee. The question shifts from "can we find targets" to "which four of these twenty-eight are worth a first call".

Sequence the pipeline so the first deal makes the next ones easier

The order in which add-ons are acquired is a strategic decision, not an accident of who says yes first. A first add-on that is small, clean and quick to integrate builds a track record the platform can point to in the next owner conversation. A first add-on that is large, messy and transformational absorbs management attention for a year and stalls the programme.

Keep the pipeline live rather than running a fresh search for every deal. Owners who said no eighteen months ago are a well-qualified target group, particularly when a succession question has moved closer in the meantime. Refresh the universe periodically instead of rebuilding it: ownership changes, new entrants and financial deterioration all create timing that a static list never surfaces.

Why buy-and-build pipelines run dry after the second deal

The pattern is consistent enough to be predictable:

• The initial universe was built from known competitors, so it was exhausted after two or three transactions.
• The search was tied to industry codes, so specialised niche providers were never in scope.
• Criteria were financial only, so the shortlist collapsed once operational fit was tested.
• Sourcing stopped while the team integrated the first add-on, and restarted from zero months later.
• The platform's own geography was over-weighted, and cross-border candidates were never screened.

Each of these is a coverage or continuity failure, not a negotiation failure. They are fixable at the search stage, and expensive to fix later.

How StrategyBridgeAI supports add-on sourcing

StrategyBridgeAI is built for exactly this problem: finding the companies that a code-based search does not return, and turning a thesis into a defensible list.

• **Global company data**: around 50 million public and private companies across more than 100 countries, with qualitative company data, financial figures and KPIs, company structures and contact data.
• **Best-in-class longlists**: target and buyer searches driven by what a company actually does rather than by rigid industry codes, so niche SMEs that classical databases miss stay in scope. Existing longlists can be uploaded and enriched.
• **Outside-In business analysis**: competitor and peer benchmarking, SWOT, risk heatmaps and financial forecasting for candidates you want to understand before making contact, delivered as board-ready PowerPoint in your own corporate design.
• **On-demand niche market reports**: current market structure, trends, barriers to entry and risks for the niche the platform is consolidating, which is what tells you whether the universe supports the thesis at all.

Teams working this way report around 80% time savings on average compared with the manual research route, which for a buy-and-build programme mostly means the pipeline can stay live while the last deal is being integrated.

In a fragmented market, the constraint is never the number of deals you can process. It is the number of companies you can actually see.

Next step

If you are building or defending a buy-and-build thesis, the fastest way to test it is to see how large the real target universe is. Book a demo and we will run your add-on profile against our data set so you can see the universe before you commit to the programme.