Summary

  • Docusign says IAM represented 15.1% of total ARR at 31 July 2026, up from 12.6% in April and 10.8% in January. It reports total ARR only annually, so the April and July product percentages have no disclosed contemporaneous dollar denominator.
  • The January figures imply approximately $353.4 million of IAM ARR and $2.919 billion of non-IAM ARR by BTW arithmetic. Applying Docusign's fiscal-2027 guides produces a conditional year-end IAM range of roughly $639 million to $678 million, while the complementary non-IAM pool is roughly flat.
  • Those calculations do not establish quarterly IAM-dollar growth. ARR assumes renewal, product value is allocated inside multi-product contracts, and IAM is broader than the new AI-agent features. A quarterly product bridge is the missing evidence.

The number 15.1% looks self-contained. It is not. A percentage always carries a numerator and a denominator, even when only the ratio is printed. Docusign disclosed the ratio for Intelligent Agreement Management at the end of July but withheld the current total ARR in dollars. That choice makes the strategic direction visible while leaving the speed and composition of the move unresolved.

The direction is striking. IAM accounted for 10.8% of total annual recurring revenue at 31 January 2026, 12.6% at 30 April and 15.1% at 31 July. The share gained 1.8 percentage points in the first quarter and another 2.5 points in the second. Across six months it rose 4.3 points, or almost 39.8% relative to its January share by BTW's calculation.

That 39.8% is growth in the ratio, not growth in IAM dollars. The distinction is not pedantic. A product's share can rise because its own annualized contract value increases, because the rest of the portfolio shrinks, because multi-product value is reallocated, or because several of those forces occur together. Without the current total, none can be measured from the reported percentage alone.

January is the one complete snapshot

At the fiscal year-end, Docusign published both sides. Total ARR was $3.272 billion at 31 January, and IAM represented 10.8%. Multiplying the two gives approximately $353.376 million of IAM ARR. The complement, 89.2% of the total, was approximately $2.918624 billion. Both figures are BTW arithmetic, although management separately described IAM as exceeding $350 million.

That snapshot is useful because it anchors scale. It says IAM had become material but still sat inside a much larger eSignature, contract-lifecycle and related subscription base. It does not license an investor to carry the $3.272 billion total forward unchanged.

Doing so would turn July's 15.1% into $494.1 million. The multiplication is easy and the premise is unsupported. Total ARR could have expanded or contracted during the half-year. The mix also moved at both checkpoints. Docusign expressly says it reports total ARR annually at fiscal year-end, so the April and July denominators are absent by design, not hidden elsewhere in quarterly revenue.

Revenue cannot substitute for them. Docusign recorded $875.746 million of revenue in fiscal Q2, 9% more than a year earlier, with foreign exchange contributing about 1.3 percentage points. Revenue is recognized under accounting rules over service periods. ARR annualizes active contracts at a measurement date. One is a flow through the income statement; the other is a management operating measure built from contract values and assumptions. Their units may both be dollars, but their clocks are different.

Deferred revenue and remaining performance obligations are not substitutes either. Docusign tells readers to view ARR independently of revenue, deferred revenue and RPO. ARR assumes any active contract expiring within the next 12 months renews on its existing terms, and it excludes non-recurring streams recognized at a point in time. That renewal assumption makes ARR useful as a run-rate view. It also prevents the measure from being treated as cash already collected or work already contracted beyond cancellation and renewal risk.

The product boundary is allocated, not naturally separate

The missing denominator matters more because IAM is not necessarily sold on an island. For contracts spanning multiple product lines, Docusign allocates support contract value to each product offering according to its proportional share of the total contract value. It annualizes committed value by dividing by the subscription term in months and multiplying by twelve.

This is a reasonable way to build a product metric from bundled agreements, but it creates an analytical boundary. A change in the IAM percentage can reflect a new module, a broader renewal, a migration from a prior package, a change in bundle composition or the proportional allocation inside a contract. The public figure does not decompose those sources.

Docusign also uses fixed exchange rates set at the beginning of each fiscal year for foreign-currency contracts and adjusts previously reported ARR annually for comparison. That limits some currency noise within the year, but it means the annual comparison base may be reset. Any reconstruction should preserve that methodology rather than mixing the ARR guide with reported-revenue foreign-exchange effects.

Product integration makes the distinction practical. Docusign integrated Agreement Manager and other IAM capabilities into Docusign CLM. The company launched an AI assistant for reviewing and drafting agreement language, pre-built agents for intake and renewals, an Agent Studio, agent support in Workflow Builder and a Model Context Protocol server. These releases expand the platform. They do not establish that every dollar assigned to IAM was separately invoiced for an AI agent.

IAM is a product family and strategic architecture, not a synonym for AI revenue. It covers a repository, workflows, agreement analysis and actions across the lifecycle. An organization can adopt IAM capabilities without buying every agentic feature; an existing customer can migrate into a package whose contract value is allocated across several lines. The 15.1% mix should therefore be read as management's product classification under its disclosed method, not a clean revenue ledger for generative AI.

Customer counts do not supply the missing bridge

The operating context supports adoption without resolving dollars. Docusign had more than 1.9 million total customers at July, including about 289,000 direct customers. It counted 1,296 customers with annualized contract value above $300,000, up from 1,137 a year earlier. In April, management said 40,000 customers were investing in its IAM roadmap.

These figures have different populations and definitions. Total customers include very small businesses served digitally as well as direct enterprise accounts. The $300,000 cohort is classified by annualized contract value across the customer relationship, not disclosed IAM spend. The 40,000 statement concerns customers investing in IAM, but the company did not pair it with IAM ARR per customer, cohort retention, new-logo counts or the division between paid migrations and added modules.

It would therefore be wrong to divide an inferred IAM amount by 40,000 and call the result average IAM ARR. The customer count was disclosed at April while the only complete product-dollar snapshot is January. The two dates differ, the scope of the word “investing” is not a billing schedule, and multi-product allocations remain in the measure.

What the counts do show is a credible commercial surface. Docusign has a huge installed base, a growing large-contract cohort and distribution through direct sales, partners and self-service. IAM can grow by converting existing eSignature relationships as well as by winning new customers. That makes the missing split between migration, cross-sell and new-logo ARR more important, not less.

Guidance reveals the destination more than the journey

Docusign supplied enough year-end guidance to build a conditional endpoint. It expects fiscal-2027 total ARR growth of 8.5% to 9.0% and says IAM should represent approximately 18% to 19% of total ARR exiting the fourth quarter. Applying the total-growth range to the $3.272 billion January base gives approximately $3.550 billion to $3.566 billion of total ARR at the next year-end, before any annual adjustment to the comparison base.

Combining the low total with an 18% share and the high total with a 19% share gives approximately $639.0 million to $677.6 million of IAM ARR. The complementary non-IAM pool spans about $2.876 billion to $2.925 billion across the outer corners. Compared with the reconstructed $2.919 billion non-IAM amount at January 2026, that is roughly a 1.5% decline to a 0.2% increase.

The implication is economically important: if both guides are met and definitions remain comparable, nearly all of the net ARR expansion could sit in the IAM bucket while the non-IAM pool finishes roughly flat. The implied IAM endpoint would be about 80.8% to 91.8% above the reconstructed January amount.

None of those product-dollar figures is guidance issued by Docusign. They are BTW scenario arithmetic combining two management ranges. The lowest total need not coincide with the lowest IAM percentage, and the highest total need not coincide with the highest percentage. Currency-base adjustments, packaging and allocation can also affect the comparison. The range is a map of what the guides jointly permit, not a promise about either product pool.

It also says nothing precise about July. A business can travel from 10.8% to an 18%-19% endpoint along many paths. IAM dollars could accelerate smoothly, arrive through large fourth-quarter renewals, or move in steps when customers migrate. Non-IAM dollars could grow early and flatten later, or contract before recovery. The July ratio locates the mix; it does not reveal the path's cash, contract or revenue profile.

Profitability belongs on another ledger

The quarter's economics improved at company level. GAAP operating income rose to $117.621 million from $65.227 million a year earlier. Operating cash flow was $334.546 million and free cash flow was $295.8 million, a 34% margin. These numbers show that the platform shift is occurring inside a profitable and cash-generative company.

They do not provide IAM unit economics. Docusign does not disclose product-level revenue, gross profit, sales expense or cash collection for IAM. Hosting costs increased partly to support IAM expansion and the migration of customer data to cloud storage, but that cost statement also covers shared infrastructure. Company gross margin cannot tell whether an incremental IAM contract is more or less profitable than a renewal of a mature eSignature package.

The same separation applies to repurchases. Docusign spent $306.5 million buying back shares during the quarter. Capital return may affect per-share outcomes and the cash balance; it does not validate the product allocation or fill the total-ARR denominator. Product adoption, income-statement profitability and capital allocation are three distinct questions.

The market has a ratio and needs a bridge

Docusign's disclosure is not empty. Three successive IAM percentages establish that management is moving more annualized contract value into the platform category. The year-end target is ambitious enough to imply a substantial change in product mix. The launches and integrations explain why customers may buy more than an electronic-signature seat.

The evidence remains incomplete at the point investors most need it. A rising share can coexist with several different outcomes for the mature franchise, and the current quarterly total is the number that distinguishes them. Publishing only the product percentage creates precision around mix while leaving the economic mass unmeasured.

The clean remedy is small: disclose quarterly total ARR, IAM ARR and non-IAM ARR under a stable product-allocation policy, then explain material classification changes. A bridge from opening balance through new logos, expansions, migrations, contractions and churn would be better still. Until that arrives, 15.1% is a useful direction marker—not a quarterly dollar result.

Sources