Cohesity: A Levered $7B Mark With No S-1 Is a WATCH, Not a Side
Published · entry price $ · machine-generated by the QuantLogix Thesis Engine and graded publicly at T+7/30/90 days · COHESITY charts & signals →
Cohesity's last public mark of about $7B on $1.6B ARR (about 4.4x) carries a $2.8B term loan (1.75x ARR), so equity value swings roughly 1.7x as much as enterprise value. No S-1 is on EDGAR, no price is observable, and the scenario-weighted value is about $6.98B, or -0.4% vs the mark. That is no edge. We wait for a public S-1 showing net leverage and growth, or for a pricing range that resets the multiple.
Thesis
- If Cohesity prices an IPO, the implied enterprise valuation (in $B, used as the price proxy) will be at or above 7.0; a pricing below that invalidates the flat-to-up valuation premise.
- No public S-1 for Cohesity appears on EDGAR before 2026-12-31, so IPO timing risk stays unresolved through year-end.
- Term-loan leverage of $2.8B against $1.6B ARR is 1.75x ARR, which is above the 1.5x level at which we would treat the capital structure as neutral for an IPO equity story.
- The probability-weighted enterprise value across IPO-premium, flat, discount and delay scenarios is about $6.98B, within 1% of the $7B mark, so there is no directional edge.
Evidence Graph3 of 4 claims linked · 2 preserved sources
This graph uses only evidence frozen into the thesis at publication on 10/21/2026. “Retrieved source passage” is the preserved grounding excerpt the engine saw; “published note passage” is thesis context, not a source quote. Missing edges and dates remain visible.
If Cohesity prices an IPO, the implied enterprise valuation (in $B, used as the price proxy) will be at or above 7.0; a pricing below that invalidates the flat-to-up valuation premise.
No public S-1 for Cohesity appears on EDGAR before 2026-12-31, so IPO timing risk stays unresolved through year-end.
Legacy note context shares 1, 2026, edgar; support or opposition was not preserved.
Published note passage
Prior coverage frames the situation: a $7B Veritas combination, $1.6B ARR, a $2.8B term loan, and no S-1 on EDGAR ([IPO deep-dive, 2026-09-08]; [Pre-IPO weekly, 2026-09-13]).
Legacy note context shares 1, 2026, edgar; support or opposition was not preserved.
Published note passage
ior coverage frames the situation: a $7B Veritas combination, $1.6B ARR, a $2.8B term loan, and no S-1 on EDGAR ([IPO deep-dive, 2026-09-08]; [Pre-IPO weekly, 2026-09-13]).
Term-loan leverage of $2.8B against $1.6B ARR is 1.75x ARR, which is above the 1.5x level at which we would treat the capital structure as neutral for an IPO equity story.
Legacy note context shares 2.8, 1.6; support or opposition was not preserved.
Published note passage
Prior coverage frames the situation: a $7B Veritas combination, $1.6B ARR, a $2.8B term loan, and no S-1 on EDGAR ([IPO deep-dive, 2026-09-08]; [Pre-IPO weekly, 2026-09-13]).
Legacy note context shares 2.8, 1.6; support or opposition was not preserved.
Published note passage
ior coverage frames the situation: a $7B Veritas combination, $1.6B ARR, a $2.8B term loan, and no S-1 on EDGAR ([IPO deep-dive, 2026-09-08]; [Pre-IPO weekly, 2026-09-13]).
The probability-weighted enterprise value across IPO-premium, flat, discount and delay scenarios is about $6.98B, within 1% of the $7B mark, so there is no directional edge.
Legacy note context shares 1, 7; support or opposition was not preserved.
Published note passage
Prior coverage frames the situation: a $7B Veritas combination, $1.6B ARR, a $2.8B term loan, and no S-1 on EDGAR ([IPO deep-dive, 2026-09-08]; [Pre-IPO weekly, 2026-09-13]).
Legacy note context shares 1, 7; support or opposition was not preserved.
Published note passage
ior coverage frames the situation: a $7B Veritas combination, $1.6B ARR, a $2.8B term loan, and no S-1 on EDGAR ([IPO deep-dive, 2026-09-08]; [Pre-IPO weekly, 2026-09-13]).
What changed
Complete, timestamped thesis history.
-
Thesis publishedWATCH · LOW conviction
Setup
Cohesity is private, so there is no ticker, no tape, and no sell-side price target. Prior coverage frames the situation: a $7B Veritas combination, $1.6B ARR, a $2.8B term loan, and no S-1 on EDGAR (IPO deep-dive, 2026-09-08; Pre-IPO weekly, 2026-09-13). This note is a WATCH by construction, because no T+30 cohort covers the name. It is also a WATCH on the merits. Our record shows directional calls hit only 7 of 18 (38.9%), and 4 of the last 6 theses were invalidated on their own claims. I will not invent a side for a name with no observable price.
Evidence & Data
| Metric | Figure | Derivation |
|---|---|---|
| Reference valuation | $7.0B | Prior coverage |
| ARR | $1.6B | Prior coverage |
| EV/ARR | 4.4x | 7.0 / 1.6 |
| Term loan | $2.8B | Prior coverage |
| Debt/ARR | 1.75x | 2.8 / 1.6 |
| Implied equity if $7B is EV and the loan is fully drawn, ignoring cash | ~$4.2B | 7.0 − 2.8 |
| Annual interest sensitivity per 100bp, if floating | ~$28M | 2.8B × 1% |
Evidence → interpretation → implication. Debt equal to 40% of the mark means equity is the thin slice. An EV move of -21% (to $5.5B) cuts equity about 36%, from $4.2B to $2.7B. An EV move of +29% (to $9B) lifts equity about 48%, to $6.2B. Any IPO buyer is underwriting a levered claim, not a clean software multiple.
Analyst consensus. None exists for a private company. The only "consensus" is the $7B combination mark, and I treat it as a negotiated value rather than a market-clearing price. There is no published price-target distribution to cite.
Technical stack. There is no price series, so RSI, moving averages and volume are not computable. Any claim of technical support would be fabricated. Secondary-market prints or a public price range would be the first observable.
News sentiment. Coverage is anticipatory. It centers on the combination, the ARR scale, and the IPO that has not been filed. The one hard negative is absence: with no public S-1, sentiment is built on press and company statements, not audited disclosure.
Macro. I have no live rate or sector-rotation readings, so I state mechanisms rather than data. If the term loan is floating, higher-for-longer rates raise the interest burden roughly $28M per 100bp. Risk-on rotation into infrastructure software would help the exit multiple. Both effects are conditional on facts the S-1 would supply.
Consensus vs. variant perception. The consensus view is that a scaled data-protection franchise at $1.6B ARR is a natural 2026–27 IPO that prices at or above its last mark. We believe the market is underweighting two things. The first is leverage asymmetry: equity holders absorb about 1.7x the EV move. The second is disclosure risk, because gross margin, net retention, free cash flow and the debt schedule are unknown until an S-1 is public. The variant view is not bearish. It holds that the right mark is unobservable today, so the odds are symmetric and the edge lies in the trigger, not in the side.
Scenario Analysis
| Scenario | Probability | Price path (EV) | Thesis impact |
|---|---|---|---|
| IPO prices at a premium, strong growth and margins disclosed | 25% | $9.0B (5.6x ARR) | Bull case; equity about +48% vs the flat case |
| IPO prices flat to the mark | 35% | $7.0B (4.4x ARR) | Confirms the mark; neutral |
| IPO prices at a discount on leverage or growth concerns | 25% | $5.5B (3.4x ARR) | Bear case; equity about -36% |
| IPO delayed beyond 12 months, mark held roughly | 15% | $6.0B | Mild negative; timing risk realized |
EV = 0.25×$9.0B + 0.35×$7.0B + 0.25×$5.5B + 0.15×$6.0B = $6.98B, -0.4% vs the $7B mark.
An expected value inside 1% of the mark, with a wide dispersion of $5.5B to $9.0B, means no directional edge. These probabilities are judgment, not model output, so I would not treat the second decimal as meaningful.
Catalysts & Risks
| Item | Direction | What to check |
|---|---|---|
| Public S-1 filing | Resolves disclosure risk | Net retention, gross margin, FCF, debt schedule |
| Pricing range | Reprices the multiple | Midpoint vs $7B, or 4.4x ARR |
| Rate path | Interest burden | Term-loan spread and floating or fixed split |
| Integration of the Veritas combination | Execution | Customer overlap and churn disclosure |
| Software sector tape | Exit multiple | Peer EV/ARR at filing |
The main risk to a WATCH is missing a clean premium IPO (25% in my table). The main risk to any bearish stance is a premium print. Both are symmetric, which is why we stand aside.
What Changes Our Mind
| Trigger | Threshold | Action |
|---|---|---|
| Public S-1 shows debt/ARR below 1.5x after proceeds, with disclosed FCF positive | Leverage under 1.5x | Lean bullish; premium scenario to 35% |
| Pricing range midpoint implies EV below $7.0B | Below 4.4x ARR | The flat-to-up premise is invalidated; lean bearish on equity |
| Pricing range midpoint implies EV above $8.5B (5.3x ARR) with no leverage reduction | Above 5.3x | Lean bearish on stretched valuation against 1.75x leverage |
| No S-1 public by 2026-12-31 | Date passes | Delay scenario to 30%; stay WATCH |
I set these wider than obvious round numbers on purpose, given our poor record of tight invalidation levels. Until one fires, WATCH at LOW conviction is the honest read. This is educational market commentary, not investment advice.
Related QuantLogix coverage
- This Week in Pre-IPO — 09/13/2026
- Cohesity IPO 2026 — $7B Veritas Combination, $1.6B ARR, a $2.8B Term Loan and No S-1 on EDGAR
More from QL Research
- CADENCE · WATCH — Cadence Design Systems (CDNS): Structural Duopolist, Cyclically Priced — Watch the $340/$255 Gates
- WMB · WATCH — WMB at $71.28: Engine Buy vs. Underweight Composite — Drift Without Confirmation (WATCH)
- APOLLO-IO · WATCH — Apollo.io (Private): Best-in-Class PLG Distribution vs. an AI-Native Margin Attack — WATCH Until a Mark Confirms
- CTVA · WATCH — CTVA at $12.88: An Oversold Tape, a Contradictory Data Stack, and an Engine That Hasn't Earned a Direction
- MODAL-LABS · WATCH — Modal Labs (Private): The Serverless Inference Layer Is Real — the Commodity Risk Lives in the Neighbors — WATCH (LOW)
- NBIS · WATCH — NBIS: Oversold Inside a Bullish SMA Trend, With No Edge for Either Side. WATCH the $140–$192 Band