What Happened to the Software Buyout Boom?

Direct answer: software buyout cohort

A control-based audit of 25 large 2020–2022 software take-private transactions found no clean sponsor realizations by August 17, 2026 under its stated definition. That is an inventory of control exits, not evidence that every deal lost money or a basis for calculating cohort returns.

Original research artifact

A 25-row classification table, ten-claim ledger, control-based methodology note, machine-readable summary, and bounded fallback web edition.

What this page adds

Keep headline transaction value, current control, liquidity events, creditor transfers, and sponsor returns separate instead of labeling every IPO, refinancing, or partial sale as a completed exit.

Related research

• what happens when an index decides a company matters
• hardware startup financing
• where do online returns go
• SEO data pipeline

Four quantities must stay separate

The cohort total combines mixed company-reported headline transaction values. Some announcements use enterprise value, some use equity value, and others use another headline convention. The total is useful for sizing the selected cohort, but it is not a common-basis valuation series.

A defensible audit therefore records transaction size, current control, ownership events, and fund return as different fields rather than deriving one from another.

Liquidity occurred without clean control exits

The cohort generated IPO proceeds, partial stake sales, refinancings, dividend distributions, continuation transactions, and creditor-led ownership changes. Those events matter, but they are not interchangeable. Under the stated definition, a clean realization requires a completed sale, broad public distribution, or equivalent transfer that removes the original sponsor group’s control.

The narrower conclusion is more useful than a vintage verdict: a large stock of control investments assembled during the low-rate software take-private boom remained unresolved in conventional exit terms at the cutoff. Public labels such as “IPO,” “liquidity event,” and “exit” can obscure continuing control.

The control-based classification

Full sponsor realization sits at the top of the hierarchy. Partial liquidity records proceeds while control continues. Creditor transfers remain separate because a lender-led ownership change can extinguish sponsor control without creating sponsor proceeds. Continuing sponsor ownership applies when no higher-priority classification is supported.

SailPoint is the cleanest sensitivity case in the package. A public listing created a market price and sell-down route, but continuing sponsor voting control meant the IPO could not be treated as a completed control exit under this methodology.

• Current-control evidence takes precedence over transaction labels.
• An IPO is not automatically a full realization.
• A continuation transaction may create liquidity while preserving effective control.
• A creditor transfer is not evidence of sponsor proceeds.

The cohort is not fully seasoned

A zero-realization count at this cutoff is not evidence that zero future realizations will occur. Most acquisitions were only three to five years into sponsor ownership, while technology holding periods at exit had lengthened to roughly five years by the end of 2025.

Strategic sales, sponsor-to-sponsor transactions, public sell-downs, continuation structures, recapitalizations, and later operational improvement remain possible. The supportable conclusion is that conventional realization was delayed and ownership outcomes fragmented, not that the entire cohort was permanently impaired.

What the public package cannot establish

A cohort MOIC or IRR would require sponsor equity contributions, acquisition debt, add-on investment, operating cash retained or distributed, dividend proceeds, partial-sale proceeds, continuation-vehicle pricing, management dilution, fees, fund allocations, and current marks. Those fields are not available in the public package at the level required for a return calculation.

The package can support its stated control map and ownership-event inventory. Because it does not publish the underlying transaction source ledger, readers cannot independently reproduce every row from primary filings using this site alone. The claims should remain inside that evidence boundary.

• Do not compare current enterprise value with announced purchase price as a synthetic return.
• Do not treat every IPO or partial sale as a full exit.
• Do not infer sponsor loss from creditor stress without reconstructing prior cash flows.

Downloads

The web article is the reading layer. These files preserve the supplied source package and model.

• Row-level cohort (CSV) — Twenty-five classified transactions with announcement year, headline value, sponsor group, current-control summary, and outcome category.
• Claim ledger (CSV) — Ten bounded claims with evidence basis and confidence labels.
• Methodology (MD) — Cohort construction, value basis, control hierarchy, return boundary, and counter-case.
• Machine-readable summary (JSON) — Headline result, definitions, boundaries, and canonical route contract.
• Original fallback web edition (HTML) — The noindex standalone edition retained as a source artifact for the merged research package.