A-Portfolio

The company in one breath

Pasqal, ticker PSQL on the Nasdaq Stock Market, builds quantum computers that use lasers to trap neutral atoms and turn them into qubits, the tiny units that carry quantum information. Governments, research centers, and large companies can buy a machine, install one in a data center, or rent access through the cloud.

The pitch is simple. Pasqal wants customers to use its analog machines for narrow industrial problems today, then upgrade the same platform toward fault-tolerant computing, which means a quantum computer that can find and correct enough errors to run useful jobs reliably. That is the dream. The invoice already arrived.

Why now

Pasqal started trading on August 28 after merging with a SPAC, a listed cash shell built to take a private company public. The stock closed its first session at $19.11 and finished September 4 at $7.96, a 58.3% drop in four trading sessions. That means the debut price found almost no support. Good would look like several quiet weeks above a stable floor, because a falling stock with a tiny public float can turn price discovery into demolition derby.

The business has real technical milestones. Pasqal says it has seven quantum processing units installed and three more in production. That means hardware exists outside a slide deck. Good would look like at least 12 installed systems, repeat orders, and growing cloud use by the end of 2027.

The market is also forcing a cleaner question. Can Pasqal turn strong physics into a durable business before expensive financing and competitors eat the runway? A 30-day community scan found only one useful Reddit thread, with six upvotes and one comment. That means the stock has almost no credible crowd consensus yet. Good would look like developers, customers, and independent researchers discussing results rather than traders recycling the word quantum until it loses all nutritional value.

The numbers, translated

• 2025 revenue was €16.5 million, up 369% from €3.5 million. Revenue is money earned from customers and grants. The growth rate is huge because the starting base was tiny. Good would look like revenue clearing €40 million in 2026 and staying above 50% growth without grants carrying the plate.
• The 2025 loss was €92.4 million, up 90% from €48.5 million. The company lost about €5.60 for every €1 of revenue. Good would look like that loss falling below €2 for every €1 of revenue, then shrinking again as more machines and cloud hours sell.
• Booked and awarded business exceeded €66 million as of March. This includes orders and grants that may turn into revenue later. It means demand is larger than last year's sales, but it is not cash in the register. Good would look like at least half converting into recognized revenue within 18 months, with customers returning for more.
• Cash available at closing was about $360 million. That sounds comfortable, but $312.5 million of convertible-bond principal sits behind much of the financing. Convertible bonds are debt that lenders can swap for stock. Good would look like operating progress reducing the need for another financing before 2028.
• The bonds can cost $31.3 million a year in cash interest at the 10% rate. That annual interest bill is larger than 2025 revenue translated into dollars. Good would look like annual revenue quickly growing beyond three times the cash interest bill, so lenders do not own the economics before shareholders see them.
• Pasqal had 212.3 million shares outstanding after closing. At $7.96, that gives the company a market value near $1.69 billion, meaning investors value the whole equity at roughly 88 times 2025 revenue. Good for a moonshot this early would still be no more than about 30 times sales, unless independent technical proof and commercial orders accelerate together.
• The bonds and investment warrants can initially create about 58.6 million additional shares at $12. Warrants are coupons that let holders buy stock at a preset price. That potential count equals 27.6% of today's shares, meaning each existing share could own a smaller slice. Good would look like revenue and cash value growing faster than the share count.
• SPAC investors redeemed 90.6% of their shares before the merger. Redemption means they took their cash back instead of owning Pasqal. Only about $27.7 million stayed in the trust. Good would have been most holders choosing the business, so this vote of no confidence deserves more weight than the launch-day confetti.

What I like

Pasqal is commercial earlier than most quantum hopefuls. Seven installed machines and named industrial work matter. This is not three physicists, one laser, and a press release held together with hope.

The neutral-atom architecture has a credible scaling case. Pasqal demonstrated a 1,024-atom register with less than 0.5% average defects. That means it arranged a large field of atoms with very few empty spots. Good would look like the company turning that control into deep, repeatable calculations with error correction, not merely a prettier atom count.

The strategy makes sense. Pasqal sells analog capability now while building toward fault tolerance later. Independent comparisons put it ahead of neutral-atom rivals in commercial deployment, even when peers lead on error-correction evidence. I like businesses that collect customer scars before the science is finished. Those scars teach what the laboratory cannot.

Chief executive Wasiq Bokhari has deep-technology operating experience from Google, ticker GOOGL on the Nasdaq Stock Market, and Amazon, ticker AMZN on the Nasdaq Stock Market, plus a physics doctorate from MIT. That earns respect. Good management here means hitting technical milestones, converting awards into revenue, and refusing to finance the company like a payday loan wearing a lab coat.

What worries me

The valuation is still absurd relative to current sales. A stock can fall 58.3% and remain expensive. That sentence should be taped above every trading screen.

The capital structure is worse than the headline cash figure. Investors paid $250 million for $312.5 million of bond principal, a 20% original issue discount. The bonds pay 10% cash interest or 12% interest in additional debt, and their conversion price can reset as low as $7.80 after six months. That means weak shares can cause more stock to appear. Good financing protects the runway without laying dilution land mines across it.

Technical leadership is not settled. Independent work places QuEra, a private company, and Atom Computing, a private company, ahead of Pasqal on broader fault-tolerance evidence. Pasqal's materials-science result is promising, but its strongest quantum-advantage claim still needs peer review and independent replication. Good would look like outsiders reproducing a useful result that a classical supercomputer cannot match economically.

The newest audited operating numbers stop at December 31, 2025. That means buyers are valuing a fast-moving hardware company using information already eight months old when trading began. Good would look like detailed first-half 2026 revenue, cash use, bookings conversion, and customer concentration before anyone pretends the multiple is precise.

What would make this a buy

I want price and proof. The price door opens below $4.50, which would value the current share count near 50 times 2025 sales. That is still rich, but it stops charging full admission for a fault-tolerant future that does not exist yet. Good entry quality would also require the stock to hold that area for several weeks rather than slicing through it.

The proof door opens if annual revenue clears €40 million, the annual loss falls below €70 million, and at least 50% of the €66 million booked-and-awarded figure becomes reported revenue within 18 months. I also want a peer-reviewed quantum-advantage result on a useful industrial problem and at least 12 installed systems with repeat customers. Good would mean science, sales, and cash efficiency improving together.

This is research and commentary, never a buy signal. At $7.96, Pasqal is fascinating physics and a bad transaction.

What makes me walk away

I walk if the bond conversion price resets toward $7.80 while revenue misses €30 million in 2026. That would pair weak execution with rising dilution, the corporate version of drilling a hole while the boat takes water.

I walk if cash falls below $175 million before annual revenue reaches €50 million. That would mean roughly half the closing liquidity disappeared before the commercial engine reached useful scale. Good would look like every new revenue dollar requiring less cash to create.

I walk if the promised quantum-advantage work fails peer review, if Vela slips beyond 2027, or if customers stop at pilots and grants. Quantum computing can survive a delayed roadmap. This valuation cannot survive a science project pretending to be a product company.

I also walk if total potential dilution from bonds, warrants, and employee awards rises above 35% of the current share count without matching commercial progress. Owners should fund a category leader, not become the category's raw material.

IPO Watch

Oura, private, proposed ticker OURA, expected date not set and price range not set. Oura sells a smart ring and subscription that turn sleep, heart, temperature, and activity data into health guidance. Its September 3 filing shows $1.21 billion of revenue for the nine months through June, up 74%, meaning this is already a scaled consumer platform rather than a science fair project. Good would look like the final prospectus showing durable subscription retention, clean cash generation, and a valuation that does not treat every finger on Earth as a confirmed customer.

Earnings On Deck

Adobe (ADBE), September 10 after the close. Watch whether generative AI lifts recurring revenue while protecting margins, which measure how much money stays after costs. A good print means AI features increase paying use and keep operating margin healthy. A bad print means customers use the tools but refuse to pay more, which would also sharpen the competitive read on past feature Figma (FIG).

The verdict

Score: 49 out of 100. Bucket: pass. Pasqal earned points for real installed systems, explosive growth from a tiny base, strong scientific roots, a credible neutral-atom architecture, commercial customers, and a management team that understands deep technology. It lost more for a €92.4 million loss, an 88-times-sales valuation, brutal bond terms, 90.6% SPAC redemptions, heavy dilution, stale financials, and technical claims that still need stronger independent proof.

A candidate starts at 80. Pasqal is not close. I want to follow the company because the hardware could matter. I want no part of the stock until the financing stops shouting louder than the physics.

The DBTech read

A SPAC redemption happens when investors in the cash shell choose their money back instead of shares in the merger target. Pasqal's 90.6% redemption rate means roughly nine out of ten public SPAC shares headed for the exit before the deal closed. High redemptions leave less cash, a smaller trading float, and more room for wild price moves. Good would look like low redemptions because investors prefer the new company to cash. When almost everybody takes the refund, do not call it applause.

Sources and as-of date

• Price: $7.96 at the regular close on September 4, 2026, Yahoo Finance market data retrieved September 8, 2026 at 7:03 a.m. Eastern. This is the latest completed U.S. session before publication.
• Pasqal Form 20-F shell-company report, filed September 2, 2026. Share count, redemptions, bonds, warrants, and transaction structure.
• Pasqal Form 6-K, filed August 27, 2026. Merger completion and Nasdaq listing.
• Pasqal definitive proxy and prospectus, filed August 14, 2026. Audited financials, financing terms, forecasts, and risks.
• Pasqal investor presentation, filed June 30, 2026. Installed systems, booked and awarded business, customers, and roadmap.
• Pasqal investor page, checked September 8, 2026. Current company positioning and leadership.
• Independent neutral-atom competitive analysis, checked September 8, 2026. Commercial and fault-tolerance comparison.
• Oura Form S-1, filed September 3, 2026. IPO status and operating figures.
• Adobe investor relations, checked September 8, 2026. Earnings date.
• Community pulse covered three Reddit threads and one GitHub item from August 9 through September 8, 2026. Only one Reddit thread was actually about Pasqal. The social sample was thin, so I used it as temperature, not financial proof.
• Financial figures are as of December 31, 2025 unless stated otherwise. Market-value calculations use 212.3 million shares, the $7.96 close, and the investor presentation's 1.1644 dollars per euro conversion rate. Research and commentary only, never a buy signal.