How Corp Dev Teams Actually Find Acquisition Targets.

Corporate development teams find acquisition targets by running a repeatable, three-part system: translating strategy into explicit search criteria, discovering companies by capability rather than industry classification, and overlaying internal referral and market-reaction signals to rank what the search surfaces.^[1][3][4] That system sits inside a broader five-stage process — strategy formation, target identification, due diligence, negotiation and deal structuring, and integration — which treats sourcing as a discrete, ongoing stage rather than a single moment of discovery.^[6]

Most public descriptions of corp dev treat "finding targets" as a black box. It isn't. The published job descriptions, product documentation, and vendor case studies that corp dev teams actually leave behind describe a fairly consistent mechanism — one that looks less like a hunt and more like an assembly line, re-run on a schedule rather than performed once a year during planning season.

Target identification, as corp dev practices it, is the stage in which a company researches, screens, and ranks acquisition or partnership candidates against a predefined strategic criteria set before any one of them enters diligence.^[6]

Corporate development runs on five defined stages, not a single search.

Corporate development is the internal function responsible for identifying, evaluating, negotiating, and integrating acquisitions, divestitures, and strategic partnerships that advance a company's growth strategy.^[8] Industry frameworks break the discipline into five sequential stages: strategy formation, target identification, due diligence, negotiation and deal structuring, and integration and execution.^[6] Target identification sits in the middle of that chain — downstream of a defined thesis, upstream of diligence spend — which is why teams that skip the thesis step tend to generate long, undifferentiated lists instead of short, strategically justified ones.^[6]

GitLab's published corporate development job description is useful here because it is one of the few public documents that spells out the day-to-day mechanics rather than the strategy slide. The role requires research into market trends, competitive dynamics, and emerging technology; building and maintaining comprehensive target lists "across strategic focus areas" using multiple data sources; and managing a deal flow tracking system so the pipeline doesn't go stale.^[5] That is the job in three clauses — research, list-build, track — repeated continuously rather than performed as a single annual exercise.

The sequencing matters more than it sounds. A team that jumps straight from a strategy slide to a target list, skipping the step of writing explicit search criteria, usually ends up re-doing the identification stage mid-diligence once the IC asks why a given company made the cut. SmartRoom's framing of corporate development as the function that decides "which markets to enter, which target companies to buy, and how to structure deals so the economics work" implies an order of operations: market, then company, then structure — not the reverse.^[8]

Industry codes miss most of the market corp dev teams actually want.

Traditional target identification starts with SIC or NAICS codes and static market maps, and that starting point systematically under-covers fragmented or adjacent markets. CorpDev.Ai's product thesis is built directly on this gap: teams describe the capabilities they need rather than the sector label, and the system builds a list of companies by what they do, not how a database happens to classify them.^[3] The output is deliberately framed as a shortlist with "a strategic reason behind every name" — a direct response to the common complaint that code-based target lists are long but shallow.^[3]

Grata's enterprise customers describe a related workaround rather than a replacement. One anonymous strategic acquirer uses an "AI Similar" tool to take a small set of known companies and expand outward into adjacent, previously invisible targets, calling the function "super helpful" for exactly this reason.^[4] Conceptor frames the same problem from the thesis side: a team writes an acquisition thesis the way it would appear in an IC memo and receives back a ranked target list, built specifically to surface lower middle-market and off-market companies alongside the obvious, already-known competitors.^[1]

Why this matters for corp dev specifically

Unlike a PE fund running a single platform search, a corp dev team is usually running several strategic focus areas simultaneously — core category, adjacency, and emerging technology — each with its own classification blind spots.^[5] A target list built on codes alone compounds that problem three times over instead of once. A company doing contract manufacturing for medical devices, for instance, might sit under three different NAICS codes depending on which product line a database analyst used to classify it — and a capability-based search that reads "what the company does" rather than "how it is tagged" will catch all three where a code-based pull catches one.

Internal referrals convert at the highest rate but cover the smallest share of the market.

Sales teams and division managers remain the single highest-converting source of acquisition leads for corp dev, because they encounter competitors and partners in the market every week — but referral volume is inherently capped by headcount and attention span. Grata's case study documents the actual workflow directly: "We use the AI Similar tool a lot to track down new companies. We get recommendations from our sales team and division managers, plug those into Grata, and use AI Similar to find similar companies, which has been super helpful."^[4] That two-step motion — a human tip, then a software-expanded search around it — has become the default sourcing loop for strategic acquirers rather than an occasional workaround.

The limitation is coverage, not quality. Referral-based sourcing surfaces a handful of credible names in a given period at most, which is the reason nearly every corp dev team eventually layers software-driven discovery over the top instead of relying on tips alone.^[4][7] Teams that depend purely on internal referrals tend to see pipeline volume collapse the moment a key salesperson changes roles or a division reorganizes — the referral channel has no institutional memory once the person carrying it leaves.

This is also where the comparison to proprietary deal sourcing in private equity breaks down slightly. A PE firm's "proprietary" channel usually runs through intermediary relationships that persist independent of any one employee; a corp dev team's referral channel is tied to individual relationships inside the business, which is structurally more fragile. (See proprietary vs. auction sourcing for how PE firms manage that distinction.)

Due diligence data increasingly shapes target selection, not just validation.

Historically, diligence happened after a target was already chosen; now, some teams use early due-diligence-style research to decide whether an entire sector merits pursuit before any specific company enters the list. PitchBook's Van Tuyle describes this shift directly: "It's easy for anyone to have preconceived notions about the quality of a potential new sector. We can inform our teams with data on how the market is reacting to a trend and how investors are capitalizing on it to help determine if it is something we should further investigate."^[2] Diligence-grade market data is moving upstream into the identification stage itself, which compresses the boundary between "which sector" and "which company" into a single analytical pass.

This matters for how corp dev teams allocate scarce analyst time. A sector-level diligence pass that rules out a thesis before a target list is even built is cheaper than building and screening fifty companies against a thesis that was wrong from the start.^[2] It also changes who gets involved earliest: deal teams that fold a diligence analyst into the identification stage, rather than waiting until a letter of intent is drafted, tend to kill bad theses before they consume a quarter of analyst bandwidth.

Deal flow tracking has replaced the spreadsheet as the system of record.

Spreadsheet-based target lists break down once a team runs more than one or two strategic focus areas at the same time. GitLab's corp dev team lists "manage and optimize our deal flow tracking system" as a distinct job responsibility, separate from research and list-building — evidence that pipeline management has become its own discipline rather than a byproduct of a CRM export.^[5] Grata's 2026 market overview makes the same point from the vendor side: technology helps corp dev teams find more targets and stay organized, but tooling alone doesn't define a winning acquisition strategy — most effective teams pair software with structured processes that formalize evaluation criteria and tie pipeline decisions back to long-term strategic priorities.^[7]

That pairing — software for coverage, structure for discipline — shows up repeatedly across the vendor and practitioner documentation. It is the difference between a team that can name forty companies and a team that can defend why those forty companies made the list, in order, with a reason attached to each one.

A worked example shows how the three layers compound inside one strategic focus area.

Consider a mid-market industrial company whose strategy team sets a 2026 priority: expand into adjacent services that extend equipment life for its existing installed base. That is a strategy slide, not a target list, and the gap between the two is exactly where most corp dev teams lose months.

Running the thesis layer first means writing explicit criteria before any company name enters a spreadsheet — revenue range, service mix (percentage of recurring versus project-based revenue), geographic overlap with existing field service depots, and customer concentration ceiling. That written criteria set is what Conceptor's product frames as writing "the thesis the way you would in an IC memo" before generating a list.^[1]

Running the discovery layer against that thesis means searching by capability — "predictive maintenance for rotating industrial equipment," say — rather than by an NAICS code for "machinery repair and maintenance," which would miss software-enabled service providers entirely.^[3] A capability search typically surfaces three populations: direct competitors the team already tracks, adjacent players doing a related service under a different label, and small regional operators with no public profile at all — the last group being where off-market value usually sits.

Running the signal layer means two things happening in parallel: the field sales team flags two regional operators they've encountered on service calls, and a quick sector-level diligence pass checks whether investor and customer sentiment around predictive-maintenance services is actually improving or just appears to be, per the PitchBook approach of informing teams with market-reaction data before committing further analyst time.^[2] Only candidates that clear both the capability search and the signal overlay make the ranked shortlist that goes to the business unit lead for review.

The output of that process is a list with ten to twenty names instead of the two hundred a raw code-based pull would have returned — smaller, but every name on it survives a specific, stated reason for inclusion.

The objections corp dev leaders raise about this system, answered.

The most common pushback is that capability-based search produces noise — too many near-matches that waste analyst time filtering false positives. In practice, the vendor documentation suggests the opposite failure mode is more common: teams accustomed to code-based pulls initially distrust a shorter, capability-ranked list because it looks too small relative to what they are used to reviewing, even though the shorter list is the one with a defensible reason attached to each name.^[3]

A second objection is that software-driven discovery can't replace the judgment of a salesperson who has actually met a competitor's founder at a trade show. That is true, and no vendor claims otherwise — the Grata workflow explicitly treats human referral and software expansion as sequential, not substitutable: the referral narrows where to look, and the software widens what gets found once a direction is set.^[4] The two are complements, not competitors for the same budget line.

A third objection, more structural: smaller corp dev teams (often two to four people) argue they don't have the headcount to run a continuous, multi-layer process across several strategic focus areas simultaneously, and default instead to running one search per year during planning season. That constraint is real, and it is also the reason the market for corp dev-specific sourcing software has grown — the mechanism the vendor case studies describe (write thesis, search by capability, overlay signal) was built specifically because the three-layer process is too labor-intensive to run manually across more than one category at a time.^[7] Mapping that universe without burning analyst hours on manual list-building is the specific problem Acquisition Atlas's sourcing engine exists to solve.

The three-layer target map most corp dev teams run, whether they name it that or not.

Across the tools and workflows described above, a consistent three-layer structure emerges regardless of which vendor or internal process a team uses:

• Thesis layer — translating a strategic goal into explicit, searchable criteria (capabilities, geography, size, customer type) rather than leaving it as a sector label.^[1][3]
• Discovery layer — software-driven expansion beyond already-known companies and beyond industry codes, typically through capability or similarity search.^[3][4]
• Signal layer — human referrals, market-reaction data, and early diligence research overlaid on the raw list to rank and prioritize it.^[2][4]

Teams that run all three layers consistently produce shortlists with a stated strategic reason behind every name, rather than a long, undifferentiated list of companies that merely share a classification code.^[3] Teams that skip the thesis layer and jump straight to discovery tend to produce exactly that undifferentiated list — more names, not better ones. Teams that skip the signal layer end up with a technically correct list that sales and the business units don't trust, because no one on the ground validated it.

The mechanics above point to one practical conclusion: target identification is a continuous, three-layer system that has to be re-run against every strategic focus area, not a project with a defined start and end date. Building and refreshing that system by hand, across multiple categories, at the cadence corp dev actually needs is the resource constraint most in-house teams hit well before they run out of strategy.

For corp dev leaders scoping how much of this to build internally versus source externally, two adjacent questions are worth separating out: how big the addressable universe actually is relative to current coverage (see the sourcing gap), and how much of the discovery layer can reasonably be automated versus kept as analyst judgment (see AI in PE). Neither question has a single right answer, but both get easier to answer once the thesis layer is written down explicitly rather than carried in someone's head.

^[1]: Conceptor, "Acquisition Target Search — Find Acquisition Targets with AI" ^[2]: PitchBook, "M&A due diligence for corporate development teams" ^[3]: CorpDev.Ai, "Acquisition Target Identification and Research" ^[4]: Grata, "Corporate Development M&A Platform & Software" ^[5]: GitLab Handbook, "Corporate Development" ^[6]: Dealroom, "Corporate Development (Corp Dev): Definition, Career Path & More" ^[7]: Grata, "Top Corporate Development and M&A Software in 2026" ^[8]: SmartRoom, "Corporate Development: Strategy, Structure & Process"

FAQ: How Corp Dev Teams Find Acquisition Targets