Okta Rose 29% on a 0.8% Guidance Raise. What Got Repriced?

Okta raised its full-year revenue guidance by $25m at the midpoint — from $3.19bn-$3.21bn to $3.22bn-$3.23bn, a change of about 0.8% — and the market responded by adding close to $6.8bn of market value in a single session. The shares closed on 27 August at $173.33 against $134.42 the day before, a gain of 28.95%. On roughly 173.8m shares outstanding, the value created is about 270 times the revenue upgrade that triggered it.

That gap is the story. Okta's second quarter of fiscal 2027, reported after the US close on 26 August, was a good quarter for a company growing revenue at 10.6%. It was not a quarter in which the growth rate changed. What changed was cash conversion, the GAAP profitability picture, and — more than either — the market's willingness to believe that securing AI agents is a new demand curve for identity software rather than a new label on an existing one. Anyone asking whether the Q2 print justifies the re-rating is really asking about that third thing, because the first two do not get you anywhere near 29%.

What Okta actually reported

Revenue was $805m, up 10.6% year on year, against analyst estimates of roughly $793m. Subscription revenue was $793m, up 12%. Adjusted earnings came in at $1.05 a share against consensus of $0.96-$0.97, a beat of roughly 9%.

The margin line is the one most easily misread. On a GAAP basis, operating margin reached 13.3% against 5.6% a year earlier — an expansion of 7.7 percentage points. On a non-GAAP basis — the measure Okta guides to and the street models on — margin was 28.2% against 27.7%, an expansion of about 50 basis points, on non-GAAP operating income of $226m against $202m. Both are correct, and they describe different things. The GAAP move is largely stock-based compensation falling as a share of revenue; the underlying operating margin barely moved. Free cash flow was $227m against $162m a year ago, up 40%, taking free cash flow margin from 22% to 28%.

The backlog metrics were solid rather than spectacular. Remaining performance obligations — total subscription backlog — rose 17% to $4.858bn. Current RPO — the portion expected to convert into revenue within twelve months, and the better read on near-term demand — rose 14% to $2.585bn. Trailing twelve-month dollar-based net retention was 107%, up one point on the year. Customers spending more than $100,000 in annual contract value numbered 5,255, up 6%.

Guidance moved on three fronts, and they moved by very different amounts:

• Revenue: full-year FY2027 guidance to $3.22bn-$3.23bn from $3.19bn-$3.21bn. A midpoint increase of $25m, or 0.8%.
• Adjusted EPS: to $3.90-$3.94 from $3.79-$3.87. A midpoint increase of roughly 9 cents, or 2.3%.
• Free cash flow: to $910m-$930m from $855m-$885m, a margin of 28-29%. A midpoint increase of about $50m, or 5.7%.

Third quarter guidance was set at $813m-$817m of revenue and $0.92-$0.94 of adjusted EPS, implying roughly 10% growth. Full-year non-GAAP operating margin was guided to 26%.

Management attributed the raise to demand for securing AI agents, pointing to Okta Identity Governance and the company's agent discovery and security products as a substantial share of new bookings and a lift to average contract values in enterprise deals.

Why the market reacted the way it did

Read the three guidance lines in order and the shape of the reaction becomes legible. The revenue raise is a rounding error. The EPS raise is three times larger in percentage terms. The cash flow raise is seven times larger. Okta did not tell the market it would sell more; it told the market that what it already sells converts to cash at a materially better rate than previously modelled — and that this is durable enough to underwrite for a full year.

For a company that spent years being valued as a growth asset that could not stop spending, that is a genuine change in category. A business guiding to a 28-29% free cash flow margin is a different instrument from one converting at 22%, even though the revenue line is identical. But the size of that change should be kept in proportion, because the operating margin Okta guides to did not inflect. It moved 50 basis points, and the full-year target of 26% sits below the 28.2% just delivered, which implies no further expansion from here rather than an accelerating one. Better cash conversion justifies a re-rating of some size. It does not justify this one, and what is left over is narrative.

That second driver deserves harder scrutiny than it received. Two days before the print, on 24 August, Okta made Agent SSO generally available — bringing its Cross App Access standard into core Okta single sign-on so that an AI agent can be registered in Universal Directory, governed by access policy, and issued short-lived tokens instead of static API keys. It is a genuinely useful product and a plausible standard. It is also, by Okta's own description, included in core SSO plans at no additional cost.

The flagship agent capability, then, is not a new SKU carrying its own price. The monetisation route is indirect: agents create governance problems, governance problems pull customers into Identity Governance and privileged access products, and those attach at higher contract values. That is a real mechanism and the ACV commentary supports it. But it is an attach-rate story running through existing product lines, not a new revenue stream with its own meter — the difference between a step change in demand and a slightly better version of the cross-sell Okta has been executing for three years.

The evidence in this print points to the latter. Net retention of 107% is where expansion selling shows up first, and it moved one point. cRPO grew 14% against revenue growth of 10.6% — a positive spread, and the single best forward indicator in the release, but not the signature of a demand curve inflecting.

The Openbook read

Across our five factors, this print moves two of them, leaves two roughly where they were, and makes the fifth considerably harder to call.

Profitability — improved, but less than the headline implies. GAAP operating margin at 13.3% against 5.6% is a real 7.7-point gain, and free cash flow up 40% to $227m, against a full-year cash margin guided to 28-29%, is a real cash story. But the non-GAAP margin the company guides to moved only 50 basis points, to 28.2%, and the full-year target of 26% sits below it. This factor improves on cash conversion and on the GAAP-to-adjusted gap closing, not on underlying operating leverage stepping up. It is still the most defensible part of the story, because it is visible in reported numbers rather than inferred from bookings commentary — simply a smaller upgrade than 7.7 points of margin expansion sounds.

Growth — unchanged, and that is the point. Revenue growth of 10.6%, subscription growth of 12%, Q3 guided to roughly 10%, full-year revenue upgraded by 0.8%. Nothing in the quarter argues that Okta's growth rate is turning. cRPO at 14% is modestly encouraging, but a company cannot be scored as reaccelerating on a spread of three and a half points in one quarter of backlog.

Solvency — comfortable and improving. Free cash flow guidance approaching $920m at the midpoint gives Okta genuine self-funding capacity. This factor was not the constraint before the print and is less of one now.

Momentum — strongly positive, and mechanically so. A 29% single-session gain resets every technical measure. The honest caveat is that momentum scored immediately after a gap of this size tells you what has happened, not what persists.

Reward/Risk — the factor that deteriorated. At roughly $30bn of market value against $3.225bn of guided FY2027 revenue, Okta trades near 9.3 times forward revenue, and near 33 times guided free cash flow at the midpoint. Neither multiple is absurd for a business with 28-29% cash margins and a defensible position in enterprise identity. But both are being paid for a company guiding to 10% growth next quarter. The margin lever that might have underwritten them is largely spent: non-GAAP operating margin is already 28.2% and guided to 26% for the full year, so there is no large expansion left to come. That leaves the reward case resting on the agent-identity thesis converting into visible growth acceleration within a few quarters, and the risk case on the multiple having to be supported by 10% growth alone. What the 29% session did, in factor terms, was take the cushion out: the operational improvement is now in the price, and the narrative component is not yet in the numbers.

The net position is a company whose cash quality has genuinely improved and whose reward/risk balance has tightened at the same time — a common combination after a re-rating, and one worth stating plainly rather than resolving into a single verdict.

The read-across

Okta did not move alone. CrowdStrike reported the same evening, also beat, also raised, and also gained around 20% on 27 August — its best single session on record. Two identity and endpoint security companies re-rating hard on the same night, both attributing demand to AI agents, is not two data points. It is one: the market repricing the security complex on an agent-adoption thesis through whichever names reported into it.

A sector-wide repricing on a shared narrative is more fragile than a company-specific one, because it can be unwound by evidence that has nothing to do with either company. It also means the next security print into this window becomes the first out-of-sample test of the thesis.

The broader software read-across is more nuanced. Salesforce also reported on 26 August and also gained more than 20%, but on a different mechanism: an EPS beat and an AI partnership announcement. The common factor across all three is that the market is paying up for credible AI positioning in enterprise software regardless of whether the revenue is yet visible. Investors screening on that basis can compare the underlying growth and margin profiles on our screener rather than the narrative each company is telling.

What to watch next

Three things will settle whether this re-rating holds, and the first two have dates attached.

Q3 FY2027 results, expected in early December. The line that matters is not revenue — guidance of $813m-$817m is narrow and will likely be met — but cRPO. If agent-driven demand is real and converting, current RPO growth should widen further above revenue growth from the 14% versus 10.6% spread reported here. If cRPO growth compresses back towards the revenue line, the agent story is cross-sell rather than a new curve.

Net retention. One point of improvement to 107% is where an ACV uplift would first appear. Two or three more points over the next two quarters would corroborate management's account of what agent governance is doing to deal sizes. A flat print would not.

Whether Agent SSO stays free. Okta has taken the standards-setting route, giving the capability away inside core SSO to establish Cross App Access as the protocol. That is a sound competitive strategy and a deferred monetisation decision. Any move to price agent identity separately, or to gate governance features behind higher tiers, would be the clearest signal that the company sees a directly billable market rather than an attach lever. Until then, the agent revenue thesis rests on contract-value expansion, and contract-value expansion shows up in retention and cRPO, not in press releases.

Full factor scores and history for Okta are on the ticker page.