Accel India.
The stability franchise: zero senior churn, a flat $650M Fund VIII, Swiggy at 35x, and a leaked LP document showing mid-vintage funds (2014: 0.15x, 2017: 0.13x DPI as of Nov 2024) that the 2025-26 exit run is now actively repairing.
GW GP Score
Accel India is the cohort's study in franchise durability: the same partner bench for a decade, disciplined flat fund sizing, and the best pre-seed machine in the country (Atoms, now co-piloted by Google for AI and Prosus for deeptech). The leaked November 2024 LP document is the most honest datapoint in Indian VC: spectacular early vintages (Flipkart 56x), thin mid-vintage DPI, and the Swiggy/Urban Company/BlueStone/Amagi realisations since are exactly the repair job LPs needed to see. The franchise question isn't talent or process; it's whether the mid-vintage book clears.
Re-up candidate: the realised-cash evidence is doing more work than the brand story.
The important signal is not the latest round; it is whether the firm can keep converting marks into distributions.
No named analyst override yet; current view is generated from the scored dossier and should be refreshed when the next primary source lands.
Mid-vintage (2014/2017) DPI repair: next leaked or disclosed print vs the 0.13-0.15x Nov 2024 snapshot.
Atoms-Google cohort follow-on rates: does the AI funnel convert?
Mid-vintage overhang: if 2014/2017 books don't clear through the IPO window, the leaked numbers define the narrative for the Fund IX raise.
Computed from current dossier sources; analyst override pending.
Stability plus repair: the cohort's most durable partnership and best pre-seed machine, with the honest mid-vintage DPI problem now visibly being fixed through the IPO window. Positive outlook on the trajectory.
Four fresh listings + Zetwerk/Zepto; Atoms option-book is early but wide
₹1,100 Cr+ realised 2024-26 at 3-35x: set against thin mid-vintage DPI in the leaked snapshot
Google/Prosus pre-seed partnerships = structurally advantaged AI + deeptech funnel
Fund VIII closed flat and fast (Jan 2025); Atoms deploys continuously
Zero senior churn 2024-26: the steadiest bench in Indian venture
Portfolio quality 25% · Exit realisation 30% · Sector positioning 15% · Capital velocity 15% · Franchise stability 15%: see the full GP Score methodology · compare this firm →
Positive catalysts and/or re-up evidence exceed the visible risks, but the score should move only after the named proof lands.
Zetwerk liquidity clears with credible OFS/block-sale evidence.
Risk-factor wording, regulator follow-up and any valuation discount embedded in the final offer.
Mid-vintage (2014/2017) DPI repair: next leaked or disclosed print vs the 0.13-0.15x Nov 2024 snapshot.
The leaked doc is a gift: you know exactly what you're underwriting. Early vintages prove the engine; mid-vintages prove the risk; the 2025-26 realisation run shows the repair is real. The zero-churn bench is the strongest continuity guarantee in the cohort.
Mid-vintage overhang: if 2014/2017 books don't clear through the IPO window, the leaked numbers define the narrative for the Fund IX raise.
Mid-vintage (2014/2017) DPI repair: next leaked or disclosed print vs the 0.13-0.15x Nov 2024 snapshot.
Coverage initiated: inaugural GW GP Score.
Catalyst exposure · latest source pressure
Recent public evidence mapped to this GP. Read this before reading the position table: it is where the score can move next.
Keep the file warm; current events cut both ways or are not yet material enough to move score.
Attribution3 positive catalyst(s), 2 pressure catalyst(s); top driver: Indian AI startup funding >4x YoY in H1 2026.
The broad private-capital backdrop is selective rather than euphoric; fundraising and mark support now need firm-specific proof.
Score pressureRaises the bar for weak-DPI managers and stale flagship vehicles; rewards funds with fresh closes or visible exits.
Separate broad PE-VC softness from venture AI strength before changing sector-positioning scores.
Anand Rathi puts Zepto's unlisted price at ₹38-40, down from ₹50-52 in March 2026, and estimates it needs ~3,000 daily orders/store (vs ~1,500-1,800 for rivals) to reach profitability, naming Eternal/Blinkit the 'safer' quick-commerce trade instead.
Score pressureSoftens near-term DPI expectations for Zepto-heavy books; watch whether the IPO prices through the caution or the brokerage view holds.
Final DRHP pricing, whether the July IPO-wave listing (~₹8,000 Cr) confirms or slips, and any second brokerage view corroborating or disputing the profitability bar.
$676M across 57 AI deals makes AI the cleanest growth pocket in the current venture tape.
Score pressureSupports sector-positioning scores only where AI exposure is backed by credible entry price, ownership and exit path.
Do not reward AI labels without DPI-compatible positions; paper-heavy AI books stay on proof watch.
Founders are reportedly skipping the OFS while early investors, led by Nexus, line up a major liquidity event.
Score pressurePotential DPI upgrade for Nexus; concentration and regulatory discount remain live until pricing and final disclosures are visible.
IPO price band, Nexus sell-down quantum, FEMA disclosure treatment and post-listing quick-commerce unit economics.
The updated filing disclosed founder summons linked to FEMA ahead of the IPO process.
Score pressureOffsets some DPI upside for Zepto-heavy books until the issue is priced and regulatory language is understood by public investors.
Risk-factor wording, regulator follow-up and any valuation discount embedded in the final offer.
The funds
~$3bn committed across 8 India funds across 3 tracked vehicles.
| Vehicle | Vintage | Size | Stage | Note |
|---|---|---|---|---|
| Accel India Fund VIII | 2025Recovery / IPO window | $650M | Pre-seed–Series A | Closed Jan 2025, flat vs VII; AI, consumer brands, fintech, manufacturing; India + SEA |
| Accel India Fund VII | 2022Correction onset | $650M | Pre-seed–Series A | — |
| Atoms (pre-seed platform) | rolling | $500K-$2M cheques | Pre-seed | 36+ startups, $200M+ follow-on raised; Google AI cohort + Prosus Atoms X deeptech cohort |
Closest booksInnoVen Capital (3 shared) · Prosus Ventures (3 shared) · Stride Ventures (2 shared)computed · E
Fund analytics · LP underwriting screen
Computed from the public dossier: vehicle cadence, realised exits, mark drift, concentration and franchise stability. All computed signals are GW estimates.
Latest vehicle closed inside the current exit/repricing window.
Next liquidityZetwerk
4 visible exit events since Jul 2024.
Latest vehicle closed inside the current exit/repricing window.
8 up / 2 flat / 0 down tracked signals.
No senior departure flagged in key people.
8 up / 2 flat / 0 down
4 events since Jul 2024
Largest tracked active exposure
Higher spread = less balanced franchise
Exit-led repricing; entry discipline decisive
Repricing began mid-deployment
IPO prep; cross-held with Peak XV & Lightspeed
At least one cited source is dated 2026.
4 visible events tracked.
14 representative positions tracked.
No departure signal structured.
Fund-level DPI/TVPI is usually not public; proxy remains estimated.
Peer median 2 · +2
Peer median 2 · -1
Peer median 74 · +2
Peer median 62 · +19
Peer median 0 · +0
Diligence agenda · how this view can change
The live research question, the proof required, and the source standard. This prevents a GP score from becoming a stale label.
The dossier is current and has near-term liquidity or fund events that can change the view quickly.
Current · At least one cited source is dated 2026.
Mid-vintage (2014/2017) DPI repair: next leaked or disclosed print vs the 0.13-0.15x Nov 2024 snapshot.
Refresh after each filing, listing, OFS, first close, final close or partner announcement.
Atoms-Google cohort follow-on rates: does the AI funnel convert?
Zetwerk: next public-market or secondary print
Evidence quality · source file health
How hard the public record is beneath this view: source mix, freshness, low-confidence fields and the next source action.
No primary source structured.
6 total sources · 0% primary
At least one cited source is dated 2026.
Newcomer: leaked Accel India LP returns doc (Feb 2026)
Find a filing, official firm disclosure, regulator table or LP document before next score action.
Source tiering follows Gravitywell policy: primary = filing/regulator/company disclosure, secondary = reputable media or research, E = Gravitywell estimate. Evidence score is a GW control metric, not an external rating.
The position book
~$3bn committed since 2008; 34 India rounds in 2025 (second-most-active). 14 tracked below.
| Company | Sector | Entered | Entry stage | Status | 12-mo signal | Latest read |
|---|---|---|---|---|---|---|
| Swiggy | Consumer & Commerce | 2015 | Series A | Public | ▲ | ₹412 Cr OFS at ~35x (cost ₹11.17/share); mostly exited |
| Urban Company | Consumer & Commerce | 2015 | Series A | Public | ▲ | ₹390 Cr realised vs ₹14.28 Cr cost: 27x; sizable stake retained |
| BlueStone | Consumer & Commerce | 2014 | Early | Public | ▲ | Listed Aug 2025; 8.12x partial realisation |
| Amagi | SaaS & Dev Tools | 2017 | Early | Public | → | Listed Jan 2026 at a discount; ₹183 Cr sold at >3x, ~10% retained |
| Freshworks | SaaS & Dev Tools | 2011 | Seed | Public | → | Nasdaq 2021: the legacy SaaS anchor |
| Zetwerk | Logistics & Infra | 2018 | Early | Unicorn | ▲ | IPO prep; cross-held with Peak XV & Lightspeed |
| Zepto | Consumer & Commerce | 2022 | Growth | Unicorn | ▲ | Quick-commerce exposure |
| Oolka | AI | 2025 | Series A | Private | ▲ | AI fintech; $14M Series A led + ~$12M follow-on 2026 |
| Knight Fintech | Fintech | 2026 | Series A | Private | · | $23.6M Series A led (Jan 2026): banking/lending infra |
| Wiom | Logistics & Infra | 2024 | Early | Private | ▲ | Affordable broadband infra |
| Posha | Deeptech & Space | 2025 | Seed | Private | · | Kitchen robotics |
| Sarla Aviation | Deeptech & Space | 2025 | Seed | Private | ▲ | eVTOL flying taxis |
| Atoms AI cohort (×5) | AI | 2026 | Pre-seed | Private | · | Picked with Google from ~4,000 pitches: 'no AI wrappers' |
| Atoms X cohort (×6) | Deeptech & Space | 2026 | Pre-seed | Private | · | With Prosus: healthcare, climate, space, longevity |
Representative tracked positions reconstructed from public disclosures: not the full book. 12-mo signal is a GW estimate of mark direction from round/exit prints (E). Ownership stakes are not shown: not reliably public.
Sector exposure · where the book leans
Share of the tracked active book by normalized sector (GW estimate): read against our sector dossiers for crowding.
New cheques · 2025-26
34 India rounds in 2025: second-most-active in the market. The structural move is at pre-seed: the Google (AI) and Prosus (deeptech) Atoms partnerships give Accel proprietary top-of-funnel that rivals can't replicate with capital alone. Fund VIII themes: AI, consumer brands for 'Bharat', fintech wealth, manufacturing.
Up to $2M + $350K Google credits + DeepMind model access each
Science-led: healthcare, climate, space, longevity
eVTOL
Realisation · are LPs getting paid?
₹1,100 Cr+ realised across four listings in 15 months, at 3x to 35x. Set against the leaked Nov 2024 DPI snapshot (2014 fund: 0.15x; 2017 fund: 0.13x), this run is precisely the distribution repair the mid-vintages needed, and it landed after the snapshot date, so the public numbers understate current DPI.
₹183 Cr sold at >3x; ~10% retained; soft debut.
₹390 Cr against ₹14.28 Cr invested: 27x partial.
₹134.62 Cr realised vs ₹16.58 Cr cost: 8.12x partial.
₹412 Cr OFS at ~35x: the franchise's marquee realisation.
Early vintages were home runs (leaked LP doc)
2014/2017 funds as of Nov 2024, before the 2025-26 realisations
Swiggy, Urban Company, BlueStone, Amagi OFS
Same partner bench a decade running: unique in the cohort
The only cohort firm with (leaked) hard DPI numbers in the public domain: bad and good. GW read: the 2025-26 exit run plus flat fund sizing plus zero churn makes the repair trajectory credible; mid-vintage DPI likely materially above the 0.13-0.15x snapshot today. E.
What they're doing
Own pre-seed as a platform: Atoms + Google (AI, DeepMind access) + Prosus (science-led deeptech): proprietary deal flow before anyone else prices it.
Flat fund discipline: $650M VII → $650M VIII protects the early-stage return math.
Harvest-and-hold: partial OFS at listing (Swiggy, UC, BlueStone, Amagi), retain upside stakes.
'Bharat' consumer thesis: tier-2/3 brands and wealth-management fintech for the newly affluent.
What can break
Mid-vintage overhang: if 2014/2017 books don't clear through the IPO window, the leaked numbers define the narrative for the Fund IX raise.
Pre-seed platform dependency on partners (Google, Prosus): strategic partners have their own agendas and option-books.
Founding-generation transition (Prakash's WTFund role is the first visible step) has no public succession plan.
Mid-vintage (2014/2017) DPI repair: next leaked or disclosed print vs the 0.13-0.15x Nov 2024 snapshot.
Atoms-Google cohort follow-on rates: does the AI funnel convert?
Founding-partner succession signals (Prakash/WTFund was the first).
Point-in-time discipline: these are the open items we could not verify at the current vintage. They get resolved: confirmed, corrected or dropped, at the next monthly refresh, never silently.
The takes · one screen, three readers
The leaked doc is a gift: you know exactly what you're underwriting. Early vintages prove the engine; mid-vintages prove the risk; the 2025-26 realisation run shows the repair is real. The zero-churn bench is the strongest continuity guarantee in the cohort.
The Atoms-Google-Prosus stack means Accel sees AI and deeptech founders before they need your term sheet. Competing at pre-seed is structurally hard now; the contestable ground is Series B+, where Accel hands off.
Atoms cohort selections (5 AI from 4,000; 6 science-led from 2,000+) are a curated map of India's frontier pipeline. Watch Sarla (eVTOL) and Posha (robotics) as category-creation probes.
Sourcing. Every figure is sourced and dated. We tier provenance: Primary (official, regulatory, exchange or company filings), Secondary (tier-1 industry research and reputable media), and GW estimate (our own reconstruction or opinion, labelled, never presented as external fact). We prefer primary where it exists, reconcile divergent prints to cited ranges, and hold every number point-in-time: dated, and never silently restated; revisions publish as dated changes.
Fact vs opinion. Facts vs opinion: market sizes, official prints, prices, named deals and agency ratings are sourced facts (Primary/Secondary). Scores, grades, purity weights, scenario paths and indicative sparkline points are Gravitywell's analytical opinion (GW estimate): labelled, not presented as external data.
GW GP Scores are research opinions, not investment advice, not a solicitation, and not an assessment under any SEBI regulation. Dossiers are compiled from public sources believed reliable; firms named did not participate and figures marked E are Gravitywell estimates. Point-in-time: dated to the vintage shown and never silently restated.