Dynamo Software

Dynamo was selling relationship management, fundraising and investor reporting for more than two decades before any model shipped. Portfolio monitoring and fund accounting predate the models too, and stripping the models out leaves every one of those functions running exactly as before. The AI line is embedded convenience across that existing system. It handles tagging, extraction into platform fields and summarization, plus scoring and conversational querying. Those are real capabilities, but a firm running its operations on Dynamo does not depend on a model to do so.

The published material describes what the models produce without describing what checks any of it. Extraction writes structured values from offering documents and partnership agreements directly into platform fields that downstream reporting reads, and no validation, confidence threshold, exception queue or review step is described anywhere.

Relationship scoring is the sharper case: a machine judgement is attached to a named human counterparty and shapes which investors receive attention during a fundraise, with nothing published on how a score is derived, whether it is visible to the person scored, or how it is contested.

The one quantified figure attached to the AI line measures time saved rather than accuracy. Time saved is a productivity claim, and it says nothing about how often an extracted value or a generated summary is correct. No accuracy rate, error rate by document type or validation methodology is published. There is no public account of drift monitoring, a revalidation cadence or model documentation either. For extraction writing directly into fields that feed investor reporting and fund accounting, the absence of any published error measure is the gap that matters most.

The AI line carries its own reported outcome, more than eighteen hours saved per user each month, drawn from client reported metrics. Scale is stated as more than a thousand clients holding over $10 trillion in assets under management, and a stated five thousand users shaped the version three release. The company also holds top rated placements across three product categories on a review platform that verifies reviewer identity.

That is evidence generated by customers rather than by marketing. Four consecutive annual general partner surveys plus quarterly research add a body of original primary research. No client is named against a result.

Nothing published says whether client documents, deal pipelines or investor records inform models serving other clients. There is also no statement on whether any aggregate dataset is built from platform activity, or whether firms can decline participation. The exposure is specific rather than theoretical.

A platform with the deal pipelines, investor relationships and fundraising activity of a thousand competing managers sits on the most commercially sensitive information in the asset class. Relationship scoring in particular implies learning about counterparties who appear across many client books. Nothing suggests misuse and nothing sets a boundary.

Privacy coverage reaches Europe and California. The company maintains a supplemental privacy notice for the European Economic Area and states compliance with California's consumer privacy statute. It also publishes a contract addendum covering DORA's obligations on information and communications technology providers, including the heavier requirements that attach to providers supporting critical functions.

Access controls are described concretely, with role based permissions, multi factor authentication and encryption in transit and at rest. What is still absent is any account of retention periods, deletion on termination, or how investor personal data inside the portal is treated distinctly from firm data.

A trust center exists, offering current certificates, completed standard security questionnaires and downloadable diligence documents, plus a subscription for program updates. The company states annual SOC Type II audits with reports released to clients, and a SOC Type I report is also attributed to it. Encryption in transit and at rest, role based access and multi factor authentication are described, along with dedicated information security staffing. Two limits matter.

The trust center content sits behind an access request rather than being publicly readable. A 2022 ransomware incident affecting client personal data also entered the public record through a client's own regulatory breach notification, not through the company's disclosure.

The company has put a named regime into contract language and said so publicly, a specific and current commitment rather than a general claim. It publishes a contract addendum for clients subject to DORA, covering the obligations that fall on technology providers to financial entities. That includes the additional terms for providers supporting critical functions. No license is held or required, and the company states its position as a technology supplier plainly. Nothing comparable addresses United States adviser books and records obligations.

Two distinct exposures sit here and neither is addressed. Relationship scoring ranks human counterparties and influences who gets contacted during a capital raise, which is a machine judgement about people with an economic consequence, and no fairness testing, factor disclosure or appeal route is published.

Call transcript analysis carries the accent and dialect problem that applies to every speech product, across an investor base that is international by definition, with no per language or per region accuracy stated. Extraction accuracy will also track document format, so smaller and non United States managers are the most likely to be read wrongly.

No service level, warranty or correction obligation is published. Nor is there a remedy for a wrong extracted value, a misleading summary or a mistaken relationship score, and no published disclaimer places responsibility back with the firm. The consequence path is real. Extracted terms from partnership agreements and offering documents feed fund accounting and investor reporting, where an error reaches limited partners. The resilience addendum for European clients is the closest thing to a published obligation, and it addresses continuity of service rather than the correctness of what the models produce.

One published line reads like a supply chain disclosure and is not one. The claim that no external AI platform or tools are required describes what a client is required to buy, not what the vendor runs. It is easy to read as a statement that no outside model is involved, when nothing says so. No provider, model family or hosting arrangement is named for extraction or summarization, or for scoring and conversational querying.

Nothing states whether partnership agreements, offering documents or call recordings pass to a third party during processing. At least one close competitor answers the same question explicitly, which shows the disclosure is available.

Integration breadth is both internal and external, and the external partners are named. The platform connects to the two dominant corporate mail and calendar systems and to four established private markets data and sourcing providers. It exposes a secure programming interface and ships a mobile application. An investor facing portal carries electronic consent workflows for tax document distribution, and the company adds a holdings enrichment component. Internally the platform spans front, middle and back office, including fund accounting. That back office reach came through an acquisition rather than a marketing claim.

No hosting regions are offered, no residency commitment appears and no single tenant option is described. Two major public cloud platforms are named as the hosting foundation, with redundancy and disaster recovery described in general terms. The question is live rather than academic for this client base. Public breach notification records from 2022 place client data on United States based servers. The company now contracts with European clients under DORA, a regime built specifically around operational resilience and concentration risk in technology providers.

No rates, tiers or unit of charge are published and the route to a number is a demo request. The platform spans at least seven functional modules across front, middle and back office, and nothing published indicates which are separately licensed, whether the artificial intelligence line carries its own charge, or how pricing scales between a small emerging manager and a firm with tens of billions under management. A buyer cannot construct even a rough shape of a deal from public material.

Both sides of the private capital relationship are served plus the service providers between them, at more than a thousand clients and over $10 trillion in assets under management, with global offices. Strategy coverage is enumerated across private equity, venture capital, private credit, real estate, infrastructure, hedge funds and funds of funds, and the research programme addresses general partners, limited partners, hedge funds, emerging managers and fund accountants as distinct audiences, which is a reasonable proxy for who actually buys. A fund accounting business acquired in 2018 extends the reach into the back office rather than stopping at the front.