Which categories are heating up heading into 2026 IPOs

12 pre-IPO sectors ranked by composite sentiment

The grid below ranks 12 sector cohorts by composite sentiment score. Each card carries the rank, sector tag, score badge, three-bar mover covering (1) active S-1 filings in the sector cohort, (2) 90d mega-round count (≥$100M round size), and (3) 90d late-stage count (≥$25M round size), plus a deep-link to the most-active company in that sector cohort (highest valuation, tiebroken by last_data_checked_at). Sector tags use the same canonical taxonomy as the live /coverage chip-strip so the deep-link target is the canonical /coverage?sector= URL.

How the composite sentiment score works

The Sector Sentiment Heatmap blends three signal layers because each one alone is biased in a different direction. The composite is S-1 × 100 + 90d mega-round (≥$100M) × 50 + 90d late-stage (≥$25M) × 15 + GSC comparison-intent impressions / 100 + /compare/* page-views / 100. S-1 filings are weighted highest because they are the leading-indicator count of imminent IPO pricing — once a company enters the EDGAR S-1 / confidential filing pipeline, the IPO timeline is on a 6-12 month runway. 90d mega-round and late-stage counts are the funding-density signal backstop; the GSC + /compare/* demand signals capture upstream search + internal-traffic interest that has not yet converted into an S-1 filing.

The cohort is the same 12-sector taxonomy used at /sectors/ and /coverage chip-strip — AI, AI Infra, AI Labs, Fintech, Space, DefenseTech, Robotics, Cybersecurity, Developer Tools, Biotech, Crypto, Logistics. Sector tags are taken from companies.sector_tags directly; a company with multiple sector_tags contributes to every sector it carries (Anthropic contributes to both AI and AI Labs; Anduril contributes to DefenseTech and AI Infra). The pre-IPO filter excludes ipo_status IN ('public', 'post-ipo') so the heatmap tracks current institutional sentiment, not historical placement.

Why a heatmap and not just a list of S-1s

An S-1-only heatmap would overweight sectors where filing velocity is highest (AI infra, AI labs) and miss the demand-side signal in sectors where filings have not yet begun but interest is rising (Robotics, Logistics, Crypto). The composite scoring weights S-1 filings highest, but the four secondary signals — 90d mega-round + late-stage counts, GSC impressions, /compare/* page-views — provide the floor of activity that lets the heatmap surface sectors where capital is concentrating even before the S-1 filings begin. The same logic underpins /compare-demand and /investor-intent, the prior two signal feeds in this trifecta.

What this heatmap does NOT include

The Sector Sentiment Heatmap is computed bottom-up from TechStackIPO first-party signal across companies, funding_rounds, gsc_snapshots, and page_views. It does not include: third-party sector classifications like Crunchbase / PitchBook verticals — these may differ from companies.sector_tags and are not reflected. Sovereign-wealth fund placements whose disclosure is patchy and not consistently captured in funding_rounds.lead_investors. Public-company S-1 / lockup-tracker signal — that signal is captured separately at /largest-tech-ipos-2026 and feeds the lockup-expiry tracker. Public-market sector comparables — those live at /compare/-vs- side-by-side pages and are surfaced via the /compare-demand matrix, not the heatmap.