BlackLine

Full profile

BlackLine reported 4,260 customers at June 30, 2026, excluding on-premise software, so for many finance teams its AI features are a question about software already licensed. This profile covers BlackLine’s record-to-report products (account reconciliation, transaction matching, journals, task management, consolidation and intercompany) and the Verity AI features embedded in them. It does not cover BlackLine’s invoice-to-cash products except where an AI feature spans both. In the terms of What We Mean by AI, Verity is AI built into a finance platform. BlackLine describes several Verity features as agents that act.

Which AI this covers: AI built into BlackLine’s finance platform, as BlackLine documents it: generative features, agents and established machine learning or matching automation. It does not cover general-purpose assistants a team might use alongside BlackLine. See What We Mean by AI.

At a glance

BlackLine describes its cloud platform as serving mid-size and enterprise organizations across industries, and as integrating with ERP systems from Workday, SAP, Oracle, Microsoft (Dynamics 365), Sage Intacct and NetSuite. Its 2025 annual report lists the record-to-report products delivered on its Studio360 platform: Account Reconciliations, Transaction Matching, Task Management, Reporting & Analysis, Journal Entry, Journals Risk Analyser, Account Analysis, Consolidation, Compliance, Smart Close for SAP and three intercompany products. BlackLine launched Verity, which it calls a suite of AI capabilities providing "a digital workforce of embedded and auditable AI," in September 2025.

Vendor-stated capabilities

Everything in this section is BlackLine’s own description, taken from its product pages, press releases, SEC filings and published contract documents; a vendor’s description is not evidence that a feature performs as described. BlackLine’s product documentation site sits behind a customer login and was not read for this profile. Several statements below come from the FAQ sections of BlackLine’s product pages, which the pages serve as text in collapsed panels.

Close, reconciliation, intercompany and consolidation

• Account Reconciliations. BlackLine’s annual report describes a central workspace with standardized templates, review and approval workflows, stored supporting documentation and "AI-enabled summarization."
• Transaction Matching. The same filing describes a rules engine that normalizes data and identifies exceptions, errors, missing data and variances in high-volume transaction sets.
• Journal Entry. Users generate, review and post manual journal entries, with validation and approval checkpoints; BlackLine states that many postings can be fully automated.
• Task Management. Close checklists with recurring task scheduling, configurable workflow and a management console.
• Consolidation. Automated intercompany eliminations, top-side adjustments and currency translations across entities and ERPs.
• Intercompany. Intercompany Create originates transactions through rules-driven billing routes and books them to the general ledger; Intercompany Balance & Resolve flags out-of-balance trading pairs and applies automated or workflow-based resolution actions.
• Reporting & Analysis. BlackLine states the product drafts financial narratives, variance explanations and executive summaries automatically.
• Journals Risk Analyser. Monitoring of the full journal population with "AI-powered anomaly detection, trend analysis, and natural language querying."
• Smart Close for SAP. Runs inside the SAP user experience and can schedule close jobs, verify closing transactions and trigger responses such as alerts or corrections.

AI features

In its quarterly report for the period ended June 30, 2026, BlackLine describes Verity as a layer that integrates machine learning, generative AI and "agentic reasoning" into financial workflows, "under the direction and oversight of finance and accounting teams," on an architecture that separates the system processing transactions from the system that validates them. Its Verity AI page describes specialized AI agents embedded in workflows, a "deterministic engine," and users who keep final authority over outputs. In the page’s FAQ, BlackLine states that for generative AI features such as summarization, customer data is sent to a privately hosted model, is never used for training and does not leave its region of origin, and that the model services it uses are contractually barred from retaining or training on that data. For predictive machine learning, BlackLine states that it trains models only with the customer’s opt-in permission, that the models are single-tenant, and that data is anonymized and redacted of personal information before training.

On configuration, BlackLine’s agentic operations page states, in its FAQ, that rules and thresholds "can be set by humans" and that agent decisions and actions are recorded in an immutable audit trail. The same page describes its agents as able to escalate for human review and as learning "from new data and user feedback." None of the pages read says which customer role can change an agent’s settings.

Which providers process Verity data. BlackLine’s subprocessor list, effective July 1, 2026, names Anthropic PBC as a processor of Verity Accruals customer data, "providing AI-powered natural language processing and other intelligence features," alongside Amazon Web Services, Fivetran and Databricks. The same list names Google as a cloud provider whose AI tools support "BlackLine AI features" across the close, invoice-to-cash and intercompany products. BlackLine’s data processing addendum (Rev. 9.21, last modified June 20, 2023) commits it to tell customers in writing at least 30 days before it adds or replaces a sub-processor of personal data; the customer has 30 days to object on reasonable data-protection grounds and, if no alternative is agreed within two months, may terminate the affected hosted service, effective at the end of the then-current billing cycle.

Assurance, as BlackLine describes it. BlackLine’s security page lists SOC 1 Type 2, SOC 2 Type 2 and SOC 3 reports and ISO/IEC 42001 certification, refers to "semi-annual SOC Reports," and says customers obtain reports through their sales representative. Its Master Subscription Agreement commits it to make its annual SSAE-18 / ISAE 3402 audit report available to customers. The certification list on the Verity AI page names SOC 2 Type 2 and ISO/IEC 42001 but not SOC 1.

What the company’s SEC filings disclose

BlackLine is a public company, and its SEC filings are its own statements to investors under securities law: company disclosures, distinct from its marketing.

• Scale and concentration. BlackLine reported total revenues of $700.4 million for 2025, up from $653.3 million in 2024, with about 95% from subscriptions (Form 10-K for the fiscal year ended December 31, 2025). It states that a majority of its revenue comes from its Financial Close & Consolidation solutions. It reported 4,260 customers at June 30, 2026, excluding on-premise software, down from 4,394 at December 31, 2025 and 4,451 at June 30, 2025 on the same basis, and attributes the decline primarily to its prioritization of enterprise and mega-enterprise customers.
• AI strategy. BlackLine acquired WiseLayer, a developer of AI agents for judgment-based finance and accounting work, on December 15, 2025, stating that the purpose was to accelerate its own AI roadmap; total purchase consideration was $23.7 million (Form 10-K). The filing lists "AI capabilities and roadmap" and "auditability, chain of thought, and transparency" among the principal competitive factors in its market.
• AI risk factors. The annual report warns that if BlackLine’s models are incorrectly designed, or the data used to train them is incomplete or inadequate, the performance of its AI features could suffer. The June 2026 quarterly report adds that agentic AI may "operate unpredictably, exceed their prescribed authorizations, or fail to align with internal policies or legal obligations," and that errors or unintended actions could lead to financial loss or regulatory non-compliance. The company also cites competition from generic large language models and from ERP vendors integrating AI into their own offerings, which could require it to reduce prices or accept less favorable terms. It notes that the EU AI Act may affect the development and adoption of its AI features in Europe.
• Operational risk relevant to customers. BlackLine reports that it migrated all Financial Close & Consolidation customers from third-party data centers to Google Cloud Platform, that it has experienced occasional planned or unplanned downtime during the transition, and that migration carries a risk of data loss or corruption. It also discloses that close-period peaks make capacity hard to predict and that customer agreements typically include service-level credits.

Independent market evidence

• Analyst placements. No analyst placement is reported on this page. A placement that BlackLine announces would be vendor-stated unless the analyst’s own page is read.
• Regulator or standards-body mentions. No regulator or standards-body action concerning BlackLine’s AI features was found.

Independently verified outcomes

No independently verified outcomes were found. BlackLine’s filings, its product pages and press releases, and a web search on October 12, 2026 for independent studies or audits of Verity outcomes were checked; the outcome figures found are BlackLine’s own. BlackLine states that Verity Accruals reduces manual work by up to 80% and helps teams close up to three days faster, and that Verity Prepare early adopters achieved up to a 92% reduction in manual reconciliation preparation time. These are vendor-stated figures without a published method, population or baseline.

Pricing and packaging

No published BlackLine price list was found in October 2026; the company’s website sitemap lists no pricing page, and the product pages offer demonstrations rather than prices, so pricing is quote-only. The Microsoft Marketplace listing could not be read as text, so no marketplace price is reported.

BlackLine’s Form 10-K for 2025 and its quarterly report for the period ended June 30, 2026 describe the company’s commercial mechanics in general terms; they are not a price list.

• In 2025 BlackLine introduced a pricing model based on product mix, organization size and volumetrics such as transaction or entity counts; fees had historically been based on the solutions subscribed to and the number of users. BlackLine reports "platform pricing ARR," a metric it uses to track the new pricing strategy, at 11% of eligible annual recurring revenue at December 31, 2025, 13% at March 31, 2026 and 17% at June 30, 2026; it defines eligible ARR as total annual recurring revenue excluding revenue from SAP solution extensions and the public sector.
• Subscriptions carry initial non-cancelable terms of one to three years, most new contracts in 2025 carried three-year terms, and customers may reduce their subscription only at renewal. BlackLine typically invoices subscription fees annually in advance, and first-year fees are generally payable within 30 days of contract execution.
• Customers can buy through SAP, which resells BlackLine as an SAP solution extension, and through the Google Cloud, Microsoft Azure and AWS marketplaces.

BlackLine’s published Master Subscription Agreement (Rev. 8.1, December 2025), which its legal page presents as the current terms, sets these standard periods:

• Renewal. The agreement renews automatically unless the customer gives at least 30 days’ written notice before the end of the term; BlackLine must give at least 180 days’ notice if it elects not to renew.
• Price changes and reductions. BlackLine may change its fees only from the end of a term and on at least 45 days’ notice; a customer may reduce licenses or products only at the end of a term, on at least 30 days’ notice, and a renewal with reduced usage, products or length is re-priced. Payment obligations are non-cancelable and fees non-refundable except as the agreement provides.
• Usage above the order form. BlackLine may check use against the usage metrics on the order form; if use exceeds them, the parties sign an order form to establish compliance within 30 days after notice. The agreement does not say whether agent activity counts toward those metrics.
• Billing. Under the agreement (Rev. 8.1, December 2025), billing disputes must be raised within 60 days of the invoice date; delinquent invoices accrue interest at 1% per month, and BlackLine may suspend or terminate if an undisputed payment is not received within 30 days after notice.

The pages read do not say whether Verity features are included in a subscription, sold as add-ons, or metered by usage.

Known limitations

• Regional processing for Verity Accruals: a question the documents leave open. The subprocessor list shows Verity Accruals customer data processed by entities and locations in the USA only, while BlackLine Financial Close hosting spans US, EU, Asia-Pacific and Middle East regions. BlackLine’s data processing addendum says customer data is hosted in the region the customer requests when the agreement is signed, that BlackLine will not host it elsewhere except with the customer’s prior authorization or as needed to provide the service or comply with law, and that BlackLine and its sub-processors may process it from outside the hosting region "in accordance with the Principal Agreement and this DPA," with that processing authorized provided it complies with data protection laws. BlackLine’s FAQ statement, made for "features like summarization," that customer data never leaves its region of origin does not resolve how Verity Accruals serves a customer hosted outside the US.
• Preview status. The Finance Control Console, BlackLine’s central oversight tool for AI agents, was announced as a preview program, with participants gaining early access. BlackLine’s master agreement provides early-access and other evaluation offerings without warranty, SLA or support, and says they may include features that will never be released. The press release does not say whether the preview runs under those terms.
• Announced versus contracted features. The master agreement states that a purchase is based on current functionality and is not contingent on future features or on BlackLine’s public comments about them.
• Contract terms on customer data and liability. Section 5 of the master agreement lets BlackLine create "Usage Data," defined as analyses based on the customer’s use of the hosted service and "information derived from Customer Data." It states that Usage Data "may be used for customer training, to improve and enhance the Hosted Service and overall user experience, and for other research, development, diagnostic and corrective purposes in connection with the Hosted Service and other BlackLine offerings," and that any other use is permitted only in aggregated or other de-identified form. Section 6.2 reserves Usage Data to BlackLine. BlackLine’s Verity AI page states in its FAQ, under "Generative AI (LLMs): For features like summarization," that "Customer data is never used for training models," and in its page text, for "tasks like narrative summarization," that "we process your data securely and guarantee we never use it to train generalized models"; for predictive machine learning it states that it trains models only with the customer’s permission. The pages read do not say whether the agreement’s Usage Data clause applies to data that Verity features process. The agreement caps each party’s aggregate liability at the amount paid under the order form in the twelve months before the event, subject to listed exceptions (Section 12.2 includes gross negligence, willful misconduct or fraud, and indemnification obligations), and excludes liability for lost profits, revenues, lost data and indirect or consequential damages, and makes termination and a refund the sole remedy if BlackLine cannot fix a failure to perform substantially as documented. The agreement does not mention AI features. BlackLine’s annual report adds that customer agreements typically contain provisions designed to limit its exposure to product liability claims, and that existing or future laws or unfavorable judicial decisions could negate those limits.
• Data at exit. On a written request made before the agreement ends, BlackLine makes an export of customer data available within 30 days of expiration or termination; it deletes customer data, including backups, as soon as reasonably practicable and within 180 days, unless law requires otherwise.
• Platform migration. BlackLine’s own filing records downtime during its move to Google Cloud Platform and a risk of data loss in migration.
• Behavior that may change. BlackLine describes its agents as learning from new data and user feedback. The pages read do not say whether or how an agent’s behavior changes between periods, or how such changes are versioned.
• Language and edition limits. None were documented on the pages read; the product documentation that might state them is behind a customer login.

CoE analysis

The questions that matter for BlackLine are the ones the approved close pieces ask of any AI built into a finance platform: what the feature does to the record, who authorized it, and what evidence supports its output. This profile does not score or rank BlackLine.

• A drafted accrual is still an unsupported entry until finance supports it. Verity Accruals calculates amounts from data and from third-party replies its model has interpreted, and BlackLine routes each entry for approval before posting. Approval is a control only if the approver can see the source data, the third-party reply and the calculation, and could catch a wrong interpretation. See Treat AI Output Like an Unsupported Journal Entry and Review Is a Control Only When It Can Catch the Error.
• Confidence scores and reasoning help only if reviewers use them. Verity Prepare attaches reasoning and a confidence score to each recommendation. A finance team should decide in advance which confidence levels trigger deeper review, and test whether sign-off rates fall with the score.
• An agent that emails third parties is acting for the company. Verity Accruals contacts vendors and business partners. That is an authorization question as much as a feature question: who approved the agent to communicate externally, under what limits, and how the company would show those limits held. See Your AI Agent Is an Unlisted Signer. BlackLine’s explainer lists posting journal entries among the things agentic AI does in general. The Verity Accruals page says each drafted entry is approved before posting. The pages read do not state which posting rights each Verity agent holds; confirm them agent by agent.
• Read the assurance language as the vendor’s. BlackLine’s Verity AI page says Verity uses a deterministic engine "to guarantee accuracy" and that "every outcome is defensible," and the Verity Accruals page describes an audit trail that "stands up to the toughest regulatory scrutiny." Whether outputs are accurate, and whether the trail would satisfy an auditor, is a conclusion for management and the external auditor, reached on the company’s own data and periods.
• Assurance coverage is a question, not an assumption. BlackLine lists SOC 1 and SOC 2 reports and ISO/IEC 42001 certification, and its agreement promises an annual SSAE-18 / ISAE 3402 report. Whether the SOC 1 scope covers Verity features, the third parties that process Verity data, and changes in releases is not stated on the pages read. Internal Control Over Generative AI, Applied to the Close sets out why that matters for management’s own control evidence.
• Generated narrative is a claim about the numbers. Verity Narrate and Reporting & Analysis draft commentary on financial results. Commentary that reaches a board pack or a filing needs the same review as a chart; see A Chart Is a Claim.

Community Observations

Not yet published. Community Observations are added after a licensing review.

Member input under review

None yet.

Relevant diligence questions

1. Which Verity features are enabled by default in our environment, which need us to opt in, and how will we be told when a release changes a feature’s behavior or the model behind it?
2. For each Verity agent we would use, what can it do without human approval, including posting entries, sending external emails or changing records, and which role in our organization can change those limits?
3. Does BlackLine’s SOC 1 Type 2 report cover Verity features and the sub-processors that handle Verity data, and from which period?
4. Where is our data processed when we use Verity Accruals or other generative features, which model providers receive it, and does the Usage Data clause in our agreement apply to data that Verity features process, given BlackLine’s statement that, for generative features such as summarization, customer data is never used for training models?
5. How are Verity features priced: included, add-on, or metered by volume, and does agent activity count toward the usage metrics in our order form?
6. What evidence does Verity produce for each output (source data, third-party replies, reasoning, confidence score), and can it be exported and retained for our auditors?

Evidence gaps and unanswered questions

• BlackLine’s product documentation and release notes sit behind a customer login, so feature-level availability, edition, language and configuration terms could not be read.
• No published price list or readable marketplace price was found.
• The pages read do not state the general-availability status of Verity features other than Verity Prepare, or which are add-ons.
• The subprocessor list names Anthropic only for Verity Accruals and Google for AI features generally; which model providers support other Verity features is not stated.
• Whether Verity features run under the Master Subscription Agreement alone or under additional AI terms is not stated on the pages read.
• Whether agents that "learn from new data and user feedback" retrain on one customer’s data only, and how such changes are versioned, is not documented on the pages read.
• The scope of BlackLine’s SOC 1 and SOC 2 reports with respect to Verity was not read; the reports are available to customers on request.
• BlackLine states ISO/IEC 42001 certification on its security page and Verity AI page; the certificate, its scope and the certification body were not read, because BlackLine’s Trust Center did not load as readable text.
• No independent study of Verity outcomes was found.
• The subprocessor list effective July 1, 2026 still names third-party data center operators (Switch and Equinix) for one US Financial Close instance, while the 2025 annual report says all Financial Close & Consolidation customers were migrated to Google Cloud Platform. Which description is current was not resolved.

Right of reply

Organizations referenced on this page may respond under our Editorial Independence & Right of Reply policy.

Review status

This profile was last reviewed on October 12, 2026. Vendor-stated information is rechecked monthly. It is not legal, accounting or procurement advice.