# The Runway-Risk Score: Reading Distress From Public Signals

*You will score any tracked company's distress risk on weighted public signals, band it as monitor, bridge, or write-off, and defend that score.*

- Canonical URL: https://www.refolk.ai/guides/runway-risk-score
- Pillar: Investing and deal sourcing
- Format: Framework
- Published: 2026-08-26
- Last reviewed: 2026-08-26
- Reading time: 12 min
- Keywords: signs a startup is running out of money, how to spot a company heading for a down round, reading company financial distress from public signals, detect startup shutdown risk early, portfolio company runway monitoring

## Key takeaways

- Cost cuts follow a fixed escalation ladder, so a perk or tool freeze is not noise but an ordinal position on a distress curve with a 30 to 90 day fuse before layoffs.
- The 90 to 180 day early edge exists only in operational and hiring behaviour; going concern surfaces at most annually and WARN gives 60 days, both near-terminal.
- AI-washing inverted the layoff signal for public names, so read the balance sheet: Goldman found companies announcing cuts carried higher debt, higher capex, and lower profit growth than peers.
- In Refolk's index, interim and fractional CFO supply is 4.5x deeper in the US (1,975) than the UK (436), so a UK company's runway problem is harder to patch and should score higher.
- Chief Restructuring Officers are single-digit rare in Refolk's index, so a tracked company retaining one is a high-confidence act trigger, not a monitor signal.
- Blind second-scoring is mandatory: if two analysts see each other's numbers before scoring, they converge artificially and the score proves nothing.

This guide is for early-stage investors, platform and talent partners, and angels who already track a set of companies and need to know which ones are quietly heading down. The job is to convert a private cash problem into a scored, trackable signal so you can move on follow-on, a bridge, or a distressed approach before it is priced in. What follows is a fixed set of weighted public signals, a way to score each one, decision bands tied to an investor action, and a blind-check protocol so two analysts land in the same place.

Most investing guides read companies on the way up: momentum, imminent-raise flags, deal-time red lines. This one reads a company you already hold or watch on the way down. And unlike the "signs of layoffs" listicles written for employees bracing for bad news, this turns the same signals into a number tied to a decision, with documented lead times attached.

## Why a runway-risk score beats a signs-of-layoffs checklist

A score beats a checklist because distress arrives on a schedule, and a schedule can be weighted. Cost cuts follow a fixed escalation ladder: leadership exhausts the less painful reductions first, then turns to headcount when those are insufficient. That ordering is exactly what makes the signals scorable rather than anecdotal.

No published, validated weighted distress-scoring model for private companies from public signals exists. The available material is unweighted listicles aimed at employees. So this framework is a constructed standard, not a citation. I am telling you that up front because a standard that overclaims is worse than none: the lead times below are documented, the weights below are my judgement calls, and you should tune them against your own portfolio outcomes.

The reason the ladder matters is that each rung carries its own fuse. A cancelled offsite is not a shutdown, but it is an ordinal position on a curve with a 30 to 90 day timer running underneath it. Read the position, not the event.

#### The distress escalation ladder

1. **Financial pressure** - Perks, travel, and offsites removed under efficiency language
2. **Hiring freeze plus cuts** - Hiring stops while tools, contractors, and travel are restricted
3. **Reorg and exec exits** - Restructuring lands and a cluster of executives departs
4. **HR execution** - Cross-department HR outreach, then the announcement

*Cost cuts run in a fixed order, so each rung tells you how far along the curve a company already is.*

Mid-2022 Meta is the textbook run of this ladder. The CEO announced a hiring freeze, said many teams would shrink, used efficiency language, and cut travel, events, and perks. By November 2022 Meta laid off about 13% of staff, roughly 11,000 people. The warning signs telegraphed the outcome well in advance. The score exists to catch that telegraph and act on it.

## The five signal dimensions and what each one proves

Score every tracked company on five dimensions. Each proves something specific, and each has a way it lies. The point of naming the lie is that a signal you cannot falsify is a signal you cannot trust.

- **Financial pressure.** Going-concern language, missed earnings, and the debt and capex profile. Proves the cash problem is real. It lies when a profitable company redeploys capital through a restructuring charge its cash flow can absorb.
- **Operational tightening.** Hiring freeze, withdrawn postings, perk and tool and contractor cuts, reorg cadence. Proves conservation is urgent when the measures co-occur. It lies as ordinary cost-discipline culture, so require simultaneity.
- **People and leadership.** Executive cluster departures, finance-focused new hires, WARN filings, recruiter exits. Proves leadership sees trouble. It lies as routine turnover, so require a cluster, never a single exit.
- **Time-to-event.** How far along the ladder the visible signals sit. Proves urgency by lead time. It lies when a slow-burn cost programme sits at one rung for months.
- **Remedy supply.** Whether fractional finance and restructuring help is reachable in the company's market. Proves how patchable the problem is. This is a geography adjustment, not a symptom.

The time-to-event dimension draws directly on documented lead times. Anchor your score to these rather than to intuition about how bad things "feel."

| Signal | Lead time before layoff | Source |
|---|---|---|
| Discretionary/perk budget freeze | 2-3 months | practitioner write-up |
| Hiring freeze plus tool/travel/contractor cuts | 30-90 days | practitioner write-up |
| New finance-focused exec headcount review | 90-180 days | practitioner write-up |
| Public financial signals (earnings, filings) | 30-60 days | practitioner write-up |
| Cross-department HR outreach | 1-2 weeks | practitioner write-up |

Read that table as a clock, not a menu. A new finance-focused executive gives you the longest runway to act (90 to 180 days), while HR outreach means you have already missed the window to be early. The disclosure calendar guarantees you are early only if you watch operations. Going concern surfaces at most annually, WARN gives 60 days, and both are near-terminal. Your edge lives in hiring, perk, and executive-hiring behaviour.

**38% - Startups that fail because they run out of cash or fail to raise**

CB Insights, cited via wisp.blog. The single largest failure cause, which is why runway is the axis worth scoring.

## The AI-washing filter that keeps efficiency out of your distress score

Before you weight anything, filter each announced cut through one test: does it point to redeployment or to distress? This is now the load-bearing check, because AI-washing has inverted the layoff signal for public names. Markets reward the AI framing, so tone tells you almost nothing.

AI-washing describes overstated claims about AI's role in layoffs. In many cases, companies blaming layoffs on AI do not yet have any automation in place to take over those functions. Meanwhile the market keeps rewarding the story: Block's shares surged 24% after cutting 40% of staff because the market heard "AI efficiency." So the announcement is a poor distress proxy. The balance sheet is the true tell.

| Marker | Points to redeployment | Points to distress |
|---|---|---|
| Cash flow vs charge | charge absorbable by cash flow | cannot absorb |
| Debt / capex profile | normal | higher debt, higher capex, lower profit growth |
| AI initiative in place | live automation | none yet (AI-washing) |

Goldman Sachs analysts found that companies announcing layoffs carried higher debt, higher capital expenditure, and lower profit growth than peers, suggesting the cuts were responses to financial distress rather than AI gains. The counter-example is capital redeployment: TD Cowen estimated Oracle's cuts freed up $8 to 10 billion in free cash flow for data-center projects. That is a company spending, not dying.

> **Watch out:** Do not read an AI-efficiency layoff as distress
>
> Many companies announcing layoffs are more profitable than ever. Check debt, capex, and profit growth before you award a single distress point, or you will short a company that is redeploying capital.

One caveat on scope: Gartner reported that 80% of companies piloting or deploying AI reduced their workforce, with near-identical reduction rates whether AI returns were strong or weak. That tells you the layoff-plus-AI pairing is now so common it carries no information on its own. The discriminators in the table are what separate signal from theatre.

## The remedy-supply adjustment: why geography moves the score

The same operational distress signal should score higher for a UK company than a US one, because the help available to patch a runway problem is not evenly distributed. This is a geography multiplier applied after the symptom is scored, and it is grounded in Refolk's index rather than intuition.

| Role | US count | UK count | US:UK ratio |
|---|---|---|---|
| Interim / Fractional CFO | 1,975 | 436 | 4.53x |
| Chief Restructuring Officer (turnaround-tagged) | single digit | single digit | ~1x |

In Refolk's index of professional profiles, interim and fractional CFO supply is 4.5x deeper in the US than the UK. A UK portfolio company with a cash problem has a thinner bench of fractional finance help to draw on, so its runway problem is harder to patch. Apply a modest upward adjustment to the operational dimension for companies in thin remedy markets.

The Chief Restructuring Officer row carries a different lesson. In Refolk's index, formal CRO profiles tagged with turnaround-management skill are single-digit rare in both the US and UK. The CRO is a scarce, deal-specific role, not a standing hire. So a tracked company retaining or hiring one is not a monitor signal - it is a high-confidence act trigger. Nobody brings in that role for a tidy quarter.

**4.5x - Deeper interim/fractional CFO supply in the US than the UK**

From Refolk's index. Thin remedy markets make the same distress signal more dangerous, so score it higher.

Finding these people is the friction point. Executive departures without a named successor, recruiters quietly leaving, or a fractional CFO appearing on a cap table are all leading signals, but they are scattered across LinkedIn and the open web. Rather than manually reconstruct who left and who arrived, describe the pattern in plain English and let [Refolk](/) return the list.

I ran this search: `Founders and C-suite executives who left a Series B or C startup in the last 90 days without a named successor` - [see the full result list](https://www.refolk.ai/s/vwarbrswek).

*Returns recent unexplained senior departures, the cluster signal that separates a real people-dimension score from routine turnover.*

## How to score a tracked company step by step

Run this procedure per company. The first two steps are one-time setup; the rest repeat on each review cycle. Total scoring time is roughly two hours per company at baseline and under two hours per review afterward.

#### The runway-risk scoring procedure

1. **Set the watchlist and baseline** - Record last raise, valuation, headcount, and sector so every later signal reads as a dated delta, not a snapshot.
2. **Wire up disclosure feeds** - Subscribe to state WARN trackers, SEC EDGAR alerts for tracked and adjacent names, and layoff trackers so alerts arrive automatically.
3. **Score financial-pressure signals** - Check going-concern language, missed earnings, and the debt and capex profile, attaching a source URL to each point.
4. **Score operational signals** - Check hiring freeze, withdrawn postings, perk and tool cuts, and reorg cadence, dating each signal.
5. **Score HR and people signals** - Check executive cluster departures, finance-focused hires, WARN filings, and recruiter exits, requiring a cluster not a single exit.
6. **Apply the AI-washing filter** - Label each cut distress or redeployment using cash flow, debt, capex, and whether real automation is in place.
7. **Compute weighted score and band** - Sum weighted dimensions and assign monitor, bridge/engage, or write-off/approach.
8. **Run a blind second-scorer check** - Have a second analyst score independently, then compare bands and reconcile the evidence if they differ.

For weighting, I anchor financial pressure and time-to-event as the two heaviest dimensions, because they carry the hardest facts and the clearest lead times. People and operational signals confirm rather than initiate. Remedy supply is a multiplier, not a base score. Use the rubric below as a starting point and recalibrate against your own hits and misses.

**Runway-risk scoring rubric (starting weights)**

```
Financial pressure (0-3) x 3  = ___   [going concern / missed earnings / debt+capex profile]
Time-to-event    (0-3) x 3  = ___   [ladder rung reached, per lead-time table]
Operational      (0-3) x 2  = ___   [hiring freeze + tool/contractor cuts, simultaneous]
People/leadership(0-3) x 2  = ___   [exec cluster, finance hire, WARN, recruiter exits]
Subtotal = ___
Remedy multiplier: x1.0 (US) / x1.25 (thin market e.g. UK) = FINAL
Bands: 0-12 Monitor | 13-24 Bridge/Engage | 25+ Write-off/Approach
Cut label: distress (full points) / redeployment (halve financial points)
```

*Score each dimension 0-3, multiply by weight, sum, then apply the remedy multiplier. Tune weights to your portfolio.*

## The bands: monitor, bridge, or write-off and approach

The score exists to select one of three actions. The band is the deliverable, not the number. Each band maps to a distinct investor move, and the transition markers below tell you when a company has crossed from one into the next.

#### The decision matrix

Horizontal axis runs from Low distress certainty to High distress certainty. Vertical axis runs from Long runway to Short runway.

| Quadrant | What it means |
| --- | --- |
| Monitor | Keep feeds live, re-score on cadence, do nothing yet |
| Engage early | Open a bridge conversation while terms are still workable |
| Bridge or protect | Offer a bridge or protect your position before pricing moves |
| Write-off or approach | Reserve down or make a distressed approach; act, do not wait |

*Cross financial certainty against remaining runway to choose the action, not just describe the risk.*

The clearest transition markers into the act bands are documented. A cramdown is the mandatory conversion of preferred shares of investors who fail to participate into a lower class of preferred or common at a 1-to-1 or worse ratio, dissipating liquidation preferences and rights. During a down round, investors may require a liquidation multiple above 1X, such as 1.5X or 2X. Both are late, priced-in markers - if you are reading them, you have missed the early edge.

The other hard act-triggers are a formal going-concern statement, a cluster of insider departures, cross-department HR outreach one to two weeks before an announcement, and the retention of a CRO. The terminal indicator is missed payroll or vendor payments: falling behind on paying employees or suppliers is a red flag that the write-off-or-approach band is already correct.

> **Rule:** A CRO retention is an act trigger, not a monitor signal
>
> Because turnaround-tagged Chief Restructuring Officers are single-digit rare in Refolk's index, a tracked company hiring one has already concluded it needs one. Move the band to write-off or approach.

```pull
Your edge lives in operations, because filings only confirm what payroll already knew.

## Frequently asked questions

### How early can I actually detect a startup running out of money?

The genuine early edge is 90 to 180 days, and it lives entirely in operational and hiring behaviour: perk and tool freezes appear 2 to 3 months before layoffs, and a new finance-focused executive triggers headcount reviews within 90 to 180 days. Public filings do not give you this. Going concern surfaces at most annually and a WARN notice is a 60-day legal minimum that is often near-terminal. If you wait for filings, you are late.

### How do I tell an AI-efficiency layoff from a distress layoff?

Read the balance sheet, not the announcement. Goldman Sachs analysts found companies announcing layoffs carried higher debt, higher capex, and lower profit growth than peers, which points to distress. Redeployment looks different: a restructuring charge that cash flows can absorb, with a live automation initiative actually in place. Block's shares surged 24% after cutting 40% of staff because the market heard efficiency, so tone is a poor proxy and the profile is the true tell.

### Why require blind second-scoring instead of just discussing the company?

Because if two analysts see each other's numbers, they converge artificially and the agreement proves nothing. Blind scoring, then band comparison, is the only way to know whether your evidence supports the read or whether one person anchored the other. Reconcile the underlying evidence only after both scores exist independently, and treat a two-band gap as a signal that your weights or your facts disagree.

### Is a WARN notice a good early-warning signal?

No. Federal WARN requires only 60 calendar days of notice for a mass layoff, and it is often filed close to the event, so treating it as early warning is a common mistake. A large employer can also file WARN for routine restructuring. Use WARN as a confirming act-trigger alongside going-concern language and funding history, not as your leading indicator.

### Should the same signal score differently for a UK company versus a US one?

Yes. In Refolk's index, interim and fractional CFO supply is 4.5x deeper in the US (1,975 profiles) than the UK (436). A runway problem is harder to patch with fractional finance help in a thin market, so the same operational distress signal should score higher for a UK company. The remedy supply, not just the symptom, shapes the risk.

---

*From the Refolk guide library. I revise these guides rather than replacing them, so the current version is always at https://www.refolk.ai/guides/runway-risk-score*