What Actually Predicts Whether a Startup Gets Acquired — or Dies?
We scored 75 engineering-software startups across seven dimensions, then separated the acquisitions from the bankruptcies. One dimension blew the doors off. It wasn't technology.
AI Answer
ThreadMoat analyzed 75 exited engineering-software startups (63 acquired/IPO, 12 bankrupt/shut down) across 7 scoring dimensions. Growth Metrics is the #1 predictor of acquisition vs. failure (gap: 1.64 on a 5-point scale), followed by Funding Efficiency (1.44) and Industry Impact (1.15). Technology Differentiation is the weakest discriminator (gap: 0.46) — good tech is table stakes. The average acquired company scored 3.67; the average failure scored 2.68. Hyperganic (WS 4.11) is the most instructive outlier: it outscored the acquired average in 6 of 7 dimensions but failed — its weakest score was Funding Efficiency (3.0).
Every founder believes their technology is the reason they'll win. Every pitch deck opens with a technical differentiation slide. And in the industrial-AI space, the technology usually is good — these are hard-science companies solving real physics problems.
But when we ran the numbers on which startups actually get acquired versus which ones go bankrupt, technology turned out to be the worst predictor of the seven dimensions we track. Not the best. The worst.
The thing that actually separates exits from failures isn't even close.
75 companies. Two outcomes. Seven dimensions.
ThreadMoat scores every startup in its dataset on a 1–5 scale across seven dimensions: Market Opportunity, Team & Execution, Technology Differentiation, Funding Efficiency, Growth Metrics, Industry Impact, and Competitive Moat. These are analyst-assigned scores based on performance data, not founder self-assessments.
Of the 900+ companies we track, 75 have reached a terminal outcome — 63 acquired or IPO'd, 12 shut down. Same scoring framework, same analyst, same market window. Two very different endings.
| Dimension | Acquired avg | Failed avg | Gap | |
|---|---|---|---|---|
| Growth Metrics | 3.43 | 1.79 | 1.64 | ← Blowout |
| Funding Efficiency | 3.49 | 2.05 | 1.44 | |
| Industry Impact | 3.67 | 2.52 | 1.15 | |
| Competitive Moat | 3.35 | 2.48 | 0.88 | |
| Market Opportunity | 3.72 | 2.86 | 0.86 | |
| Team & Execution | 3.87 | 3.06 | 0.81 | |
| Technology Differentiation | 3.77 | 3.31 | 0.46 | ← Table stakes |
Look at the bottom row. The companies that went bankrupt averaged 3.31 out of 5 on technology. That's not a failing grade. That's a solid product built by competent engineers solving a real problem. They died anyway.
Now look at the top row. A gap of 1.64 on a 5-point scale is enormous. Failed startups averaged 1.79 on Growth Metrics — that's effectively zero growth at the time of scoring. The acquired companies? 3.43 — not even exceptional by dataset standards.
You don't need hypergrowth. You need visible growth. Flat is fatal.
The silent killer
Funding Efficiency at 1.44 is the dimension nobody talks about in pitch meetings but every acquirer runs the math on.
Here's the arithmetic: an acquirer's willingness to pay is heavily discounted by how much capital the target has already consumed. A company that raised $60M to build $50M of enterprise value is a terrible deal for both sides — the founders are underwater and the acquirer is overpaying for proven traction that was expensive to produce.
The startups that got acquired built value on relatively modest capital. The startups that failed burned through venture money without commensurate revenue or product progress. This is the dimension that separates the good exits from the bad outcomes more than almost anything else — and it's the one founders have the most direct control over.
The Hyperganic problem
The most instructive data point in the whole dataset is a company that should have been acquired.
Hyperganic scored a 4.11 weighted score — higher than the average acquired company (3.67). It beat the acquired average in six of seven dimensions:
- Market Opportunity: 4.80 (vs. 3.72 acquired avg)
- Technology Differentiation: 4.20 (vs. 3.77)
- Competitive Moat: 4.30 (vs. 3.35)
- Growth Metrics: 4.10 (vs. 3.43)
- Industry Impact: 4.50 (vs. 3.67)
- Team & Execution: 3.90 (vs. 3.87)
One dimension lagged: Funding Efficiency — 3.0 (vs. 3.49 acquired avg).
Hyperganic shut down in 2024 and pivoted to Leap71. A company with a 4.8 on market opportunity, 4.5 on industry impact, 4.2 on technology — dead. The technology was clearly there. The market was there. The team executed. But the burn rate outran the traction, and when the music stopped, no acquirer stepped in at a price the cap table could absorb.
One dimension. That was the gap between a strong acquisition outcome and a shutdown.
What this means
If you're investing in engineering-software startups and the company scores well on technology but shows flat growth and capital-heavy operations, the base rate says this company is more likely to fail than to get acquired. Growth Metrics and Funding Efficiency are the two dimensions that separate outcomes. Technology and market size are entry tickets — they don't predict survival.
If you're building in this space, your moat is not your algorithm. It's the proof that customers are paying for it, that each dollar of venture capital generates more than a dollar of enterprise value, and that your growth rate is compounding. A killer demo and a strong patent portfolio do not make you an acquisition target. Revenue velocity does.
If you're acquiring, the best deals in this dataset are companies with strong Growth Metrics and strong Funding Efficiency — real traction built on modest capital. That combination means you're buying proven demand at a reasonable basis, not overpaying for overcapitalized potential.
The uncomfortable bottom line
Good technology in a big market is the minimum bar for existing in engineering software. Every serious startup in this dataset has it. The ones that failed had it too.
What separates the acquisitions from the shutdowns — across 75 companies, seven dimensions, and three years of analyst scoring — is whether you grew, and whether you did it without lighting money on fire.
The founders who died were building real products. They just weren't building businesses.
This analysis comes from the ThreadMoat Exit Benchmarks dashboard.
63 acquisitions and 12 failures, scored across every dimension — filterable by investment category, lifecycle stage, and exit outcome. The data behind this article is live, interactive, and updated as the market moves.
Scores are analyst-assigned as of August 2026. Sample: 63 acquired/IPO, 12 bankrupt/shutdown. All companies are in engineering software and industrial AI.