Last price
$393.25
The only name in the index where the thesis fires BOTH ways at the same time — purest disruption target, strongest incumbent counter-move.
Live quote sourced from Yahoo Finance. Prices cited in narrative below reflect the original memo date and may be stale.
Intuit is the single cleanest illustration of Sequoia's services-as-software thesis acting in both directions on one P&L at once. On one side, tax prep is the article's textbook example of a judgment-light, rule-based, per-outcome-priced workload — and Claude/GPT can already do basic returns end-to-end. On the other, Intuit has made itself the regulated execution layer inside the labs: multi-year Anthropic and OpenAI partnerships that surface TurboTax, QuickBooks, Credit Karma and Mailchimp through MCP integrations. Bulls and bears read the same facts and disagree fundamentally.
INTU is the most complex name under this thesis. It is simultaneously the most directly threatened incumbent (Claude can literally do a 1040 today) AND the most strategically positioned one (the dual AI-lab MCP integrations make Intuit the financial-execution layer inside the very products that threaten it). Your read on the stock turns on which of those two forces wins over the next 18–24 months — and unlike most names in the index, both forces are already measurable in real artefacts (Reddit demos, MCP integrations, TurboTax Live growth, Rillet customer counts) rather than just narrative.
Before the current AI wave, Intuit had already converted a real services budget (CPA labor, bookkeeper labor) into recurring SaaS revenue. TurboTax Live and QuickBooks Live are autopilot-shaped products sold at outcome price points — per return, per book — not per seat. That pre-AI flywheel is already running: TurboTax Live revenue +12% while IRS DIY volumes declined 5%+. The Anthropic and OpenAI MCP integrations are the natural extension — Intuit does not need to build a frontier model, it needs to be the place the frontier models call when the task requires filing, compliance, liability coverage, or payment rails. That is exactly the “regulated execution layer” niche the Sequoia article implicitly leaves for incumbents.
| Segment | Revenue | Growth | Direct AI threat | Moat that actually holds |
|---|---|---|---|---|
| TurboTax (consumer) | ~$5B | 8–9% | High — Claude does 1040s today | 50-state filing, e-file authorization, audit defense, liability coverage |
| QuickBooks (SMB accounting) | ~$10B | 14–15% | High — Rillet, AI-native ERPs | 100M+ SMB transaction history, payroll/payment rails, switching cost |
| Credit Karma (consumer credit) | ~$2B | 10–13% | Medium — chat-native alt offers | Largest consumer credit dataset, lender relationships |
| Intuit Enterprise Suite (mid-market) | Growing fast | ~40% ecosystem | Lower — compliance-heavy, complex | Proprietary mid-market data + AI-agent platform on top |
Intuit did not fight the labs — it embedded itself as the financial-execution layer inside them. When a user tells Claude “do my taxes,” Claude can calculate, but only Intuit can file with the IRS, guarantee accuracy under audit, handle 50-state compliance, and process the refund. If MCP volume materialises across Anthropic and OpenAI surfaces, INTU transforms from a consumer-software company into financial infrastructure — a Stripe-for-tax-and-accounting outcome.
TurboTax Live and QuickBooks Live are not AI vapourware — they are live, growing products that have been consuming professional-services budgets for years. QuickBooks Live posted +50% customer growth; TurboTax Live revenue grew 12% while IRS DIY volumes declined 5%+. Intuit is gaining share precisely in the segment that pure-AI cannot fully replace because it requires human liability cover for complex situations.
200M+ tax returns, 100M+ SMB transaction histories, decades of filing patterns across every state and filing status. Credit Karma layers in the largest consumer credit dataset in the US. Rillet has hundreds of customers; QuickBooks has millions. A startup can match the UX; it cannot manufacture 40 years of IRS filing history.
IRS e-file authorization, CPA network, 50-state filing licenses, payment-processor status, and audit-defense liability coverage are all regulatory artefacts that compound slowly. A frontier model cannot ship any of them by training longer. This is exactly the kind of “execution layer” capability Sequoia frames as the incumbent opportunity.
Intuit Enterprise Suite posted ~40% ecosystem-revenue growth with new contracts up ~50% QoQ — that is Rillet’s TAM but inside Intuit’s distribution, customer base, and data advantage. If IES holds this trajectory, the “Rillet eats QuickBooks” bear case weakens substantially.
March 16 2026 the founder and the entire executive team terminated all pre-scheduled stock-sale plans and the company accelerated a $3.5B buyback at ~$361 (down 56% from the $813 peak). Insider behaviour rarely aligns so cleanly with a “the market is wrong” read.
Simple W-2 filers are roughly 40% of TurboTax volume, and the work is precisely what frontier models excel at: rule application to structured documents. The Reddit “do my taxes” demo is not a future scenario; it is a current user behaviour. Filing infrastructure still protects Intuit, but pricing power on the simple-return tier is under immediate pressure.
QuickBooks at ~$10B is twice the size of TurboTax. SMBs switch accounting systems faster than enterprises, Rillet's AI-native architecture is genuinely superior for cloud-native companies, and Sequoia itself is an investor. QuickBooks growth has already decelerated into the low-teens; if that deceleration steepens, the core of INTU's P&L is affected, not just the headline consumer product.
Making Intuit's tools callable from inside Claude means users can access Intuit without visiting TurboTax.com or QuickBooks.com. Over time, Claude (or ChatGPT) becomes the interface, and Intuit becomes the backend — the “data store, not the product” failure mode. Brand equity, upsell surface, and pricing power all attenuate if Intuit is accessed through someone else's UX layer.
Fwd P/E has compressed from ~45x to ~23x, which is a meaningful reset — but it is still roughly double the multiples on sibling disruption-exposed names where the execution-layer argument also applies. The multiple now demands that the strategic thesis holds; a simple “margin of safety” case on INTU does not exist at current levels.
IRS Direct File was killed by the current administration but is a low-cost policy lever available to any future one. TurboTax has a consumer-brand overhang from the past decade of lobbying coverage that resurfaces on every free-filing news cycle. This is a headwind that periodically intensifies and never fully disappears.
INTU is the single name in the Nasdaq-100 where the Sequoia thesis and its counter-thesis both hit the same business at full force, and where both are already observable in real artefacts (Reddit demos vs MCP integrations, TurboTax Live growth vs Rillet customer counts) rather than as narrative. Read INTU as two bets layered on one ticker: a short bet on the services-as-software thesis for the TurboTax consumer tier, and a long bet on the MCP-execution-layer counter-move for the broader platform. Whichever side of that pairing compounds faster is the name’s eventual direction; right now they are visibly running neck and neck.
Positive, not Highly Positive — the business is defensible and the counter-move is real, but the thesis cuts both ways and verdict rests on MCP partnerships driving volume, not just announcements. Watch FY26 'AI-assisted' TurboTax Live unit economics and IES ecosystem growth.