Key Highlights

  • CS Disco (NYSE:LAW) trades inside a 52-week range of US$2.45 to US$9.11.
  • For the first quarter of 2026, reported on 6 May, revenue was US$41.9m, up 14% and ahead of the roughly US$40.3m analysts had modelled.
  • The bull case rests on operating leverage that is already visible. Adjusted EBITDA margin improved six percentage points year on year in the March quarter without dramatic revenue acceleration; research and development ran at US$12.9m for the quarter and sales and marketing US$14.8m.
  • With 91% of first-quarter revenue consumption-based, results move with litigation activity DISCO does not control; the company warns in its filings that the timing of legal matters makes period comparisons unreliable.

Why the market is paying attention

CS Disco (NYSE:LAW) trades inside a 52-week range of US$2.45 to US$9.11. Market capitalisation is roughly US$254m.

The reason for renewed attention is a commercial reset. In February 2026 the company, which trades as DISCO, folded its Cecilia AI tools, deposition management and timeline features into one all-inclusive platform priced on a transparent per-gigabyte basis, with no ingest fees. Management said in May that uptake had run ahead of expectations.

The next test is close: DISCO reports second-quarter results, the first full quarter judged against a raised full-year outlook and a stated goal of adjusted EBITDA breakeven in the fourth quarter.

What the company does

CS Disco sells cloud software for litigation. Its eDiscovery platform is what law firms and corporate legal departments use to ingest, process, search and review the emails, documents and chat logs that become evidence in lawsuits and investigations. Less than 10% of 2025 revenue came from outside the United States.

Revenue is consumption-based rather than seat-based: usage-based arrangements were 91% of March-quarter revenue, so results track data volumes flowing through live legal matters. The company also sells managed review services, where its Auto Review product applies generative AI to first-pass document review.

The field is crowded. DISCO's annual report names Everlaw, Relativity and Reveal Data as cloud rivals, with Nuix, Open Text, RELX and Thomson Reuters among legacy vendors and Consilio, Epiq and KLDiscovery among services providers.

Latest financial and operating performance

For the first quarter of 2026, reported on 6 May, revenue was US$41.9m, up 14% and ahead of the roughly US$40.3m analysts had modelled. Software revenue was US$34.7m, up 12%; services US$7.2m, up 25%. Gross margin was 74.2%.

Losses narrowed. GAAP net loss was US$9.6m, or US$0.15 a share, against US$11.4m a year earlier; on a non-GAAP basis, which excludes share-based compensation, the loss was US$4.2m or US$0.07 a share. Adjusted EBITDA was negative US$3.5m, an 8% negative margin against 14%.

Cash generation is the weak point. Operating outflow was US$11.7m, wider than the US$10.5m a year earlier. The quarter closed with US$17.6m of cash and US$85.4m of short-term investments, about US$103m, and no debt, down from US$114.6m in December.

The key operating metric is large-customer count: 347 customers generated more than US$100,000 over the preceding twelve months, up from 330 in December, contributing US$124m or 77% of revenue. Retention is softer. At the end of 2025 software net revenue retention was above 103% but total net revenue retention was 98%, so the installed base was shrinking slightly in dollar terms.

For 2025, revenue was US$156.8m, up 8%, with a GAAP net loss of US$44.4m and adjusted EBITDA of negative US$10.2m. Guidance for 2026 was lifted in May to revenue of US$169.25m to US$178.75m.

The catalyst that matters: one price, all the AI

The logic behind February's repricing is the investment argument in miniature. Cecilia AI had been sold in tiers on top of the core platform, forcing a separate purchasing decision on every matter. Bundling removes that friction and makes DISCO's price directly comparable with rivals per gigabyte.

Management's contention is that this lifts win rates and cuts discounting, and that higher data volumes more than pay for AI given away at the margin. The evidence is suggestive rather than conclusive: revenue from Cecilia AI and Auto Review rose more than 600% in the fourth quarter of 2025, and 41% of customers by then attributed some usage to generative AI features.

The bundle keeps growing, February also brought what the company called the first scaled agentic AI in legal technology, at no extra cost. If volumes do not follow, DISCO has surrendered a pricing lever precisely when it is asking investors to believe in breakeven by December.

The bull case

The bull case rests on operating leverage that is already visible. Adjusted EBITDA margin improved six percentage points year on year in the March quarter without dramatic revenue acceleration; research and development ran at US$12.9m for the quarter and sales and marketing US$14.8m.

About US$103m of cash and no debt buys several years. Delivery is real too: Auto Review reached Canada in April, ISO 27701 certification came in March, and the board has been rebuilt, Aaron Barfoot became chief financial officer in December 2025 and Andre Mintz, formerly Meta's global chief information security officer, joined in July.

The bear case and the risks

Usage volatility. With 91% of first-quarter revenue consumption-based, results move with litigation activity DISCO does not control; the company warns in its filings that the timing of legal matters makes period comparisons unreliable.

Retention below 100%. Total net revenue retention of 98% means the installed base contracted slightly, so growth is carried by new logos, and expansion from existing customers must now come almost entirely from data volume.

Scale and commoditisation. Against Relativity, Everlaw, Thomson Reuters and RELX, DISCO is a 577-person company spending under US$13m a quarter on research. Its annual report concedes that generic large language models and general-purpose agents may significantly alter how technology is developed and consumed.

Cash burn. Operating outflow widened to US$11.7m even as adjusted EBITDA improved. Adjusted EBITDA breakeven is not breakeven on cash or on a GAAP basis, where the quarterly loss was still US$9.6m.

Concentration. No customer exceeded 10% of 2025 revenue, but 77% comes from 347 accounts.

What investors should watch next

Second-quarter results. Guidance is US$41.5m to US$43.5m of revenue and adjusted EBITDA of negative US$4.5m to negative US$2.5m; whether software revenue lands inside the guided US$36.1m to US$37.1m range is the cleaner signal.

Net revenue retention. Total retention needs to move back above 100% for the bundling thesis to hold.

Language on fourth-quarter breakeven. Any softening of that target in August would be material.

Uptake of agentic Cecilia and Cecilia Advanced Research. Both arrive at no extra charge; what matters is billed data volume, not engagement.

Large customers and liquidity. The 347 accounts above US$100,000 and roughly US$103m of cash are the running scorecards.

Investor takeaway

DISCO enters the second half of 2026 as a measurably improved business trading well below where it began the year. Revenue growth has accelerated for four consecutive quarters, losses have narrowed on every measure, and the balance sheet holds about US$103m with no debt. Against that, the company has given up a pricing lever to buy adoption, total net revenue retention was below 100% at the last disclosure, and cash burn widened in the March quarter.

The 5 August result is the first honest read on whether February's repricing is converting into data volume or simply into lower realised prices. Until then the fourth-quarter breakeven target is a hypothesis with a near-term test date. This article is general information only and not a recommendation.

This article is for informational purposes only and does not constitute investment advice or a recommendation to buy, sell, or hold any security. Share prices can rise or fall, and past performance does not guarantee future results. Investors should conduct independent research and consult a licensed financial adviser before making investment decisions.