GetAccept

The two upper tiers carry an artificially intelligent native badge and the model features are real, covering an editor, smart content, a meeting summariser, a knowledge base and a machine learning score on each send out. The price list settles the axis anyway.

The entry tier ships with none of it, the model features on the middle tier arrive with unstated usage limits and unlimited use is sold as a separate add on, and everything the company is actually known for predates the model layer entirely: the shared room, the branded proposal, the electronic signature, the tracking, the contract store. Remove the generation and a complete and saleable product remains, which is the product this company built its five thousand team customer base on. That is the middle band by definition rather than a harsh reading.

Nothing in this product acts on a buyer without a person deciding to. A rep assembles the room, chooses what goes in it and sends it, reminders and notifications fire on rules the customer configures, and the model features draft and summarise into an editor where a human works before anything leaves. Edit after send is a genuine correction mechanism rather than a convenience, letting a seller fix a document already in a buyer's hands before signature.

Team structures with assigned roles give an administrator a way to control who can send what, and the content library plus template governance is the mechanism one named customer describes as giving control over what reps actually send out.

Off the top band because no oversight architecture is described around the model layer specifically: no review requirement on generated content, no accuracy statement on the deal score that ranks opportunities, and no audit trail of model driven changes.

The model layer is branded and packaged but not explained. Four named capabilities are sold, an editor, smart content, a meeting summariser and a knowledge base, plus a score described openly as machine learning applied to each send out. No provider, model family, version or hosting arrangement is named, and no accuracy or error rate is published for the score even though its stated purpose is to tell a seller which deals are more likely to sign.

The usage terms are the other gap and they are commercial as well as technical: the middle tier states only that usage limits apply, with unlimited use available as an add on subject to a fair use policy, and no number is attached to either the limit or the policy.

The distinguishing strength here is that the names and the numbers meet, which is exactly what a vendor graded earlier in this same session failed to do despite a larger logo wall. Quantified outcomes are attached to identified customers: a sales cycle shortened by 67 percent at one named company, a doubling of win rate at another, and about twenty hours a month saved on contract creation at a third, each linked to its own written case study.

Seven testimonials carry a full name, a job title and an employer rather than an initial. An independent review platform carries more than 1,200 reviews at an average of 4.6, which is a substantial external corroboration base. Off the top band on measurement rather than attribution: not one of those percentages carries a method, a period, a baseline or a sample, so a buyer can verify who said it but not how it was arrived at.

The sending here is transactional rather than cold, which materially changes the shape of the axis: documents and rooms go to a named counterparty already in a deal, by email or text message, usually after a conversation. That is a milder posture than prospecting by default. What is still absent is any stated position.

No consent standard for text message delivery is described, no unsubscribe or suppression mechanism is mentioned, and no regulation is named anywhere on the pricing or product surface, on a product that sends text messages to buyers as a paid feature with additional charges. Graded on the absence of a stated position rather than on evidence of a problem.

The legal surface is grouped deliberately rather than scattered, with the footer carrying a security hub that collects terms and conditions, the privacy policy, the security page and the electronic identification regulation together in one place.

The compliance claim is more precise than the usual badge: rather than asserting compliance in the abstract, the feature table states that the company follows the European data protection board's recommendations, which points at a specific and checkable body of guidance. A cookie policy is published separately and the product is sold in three languages and six currencies across European and American markets, so the obligation is real rather than theoretical.

Off the top band because none of those documents was read in this pass, so retention schedules, processor registers and data subject request routes are unestablished, and because the platform holds signed contracts and detailed buyer behaviour records for people who are not the customer.

There is no third party data business anywhere in this product, which is cleaner than most of the index. No contact database is bought, built, enriched or resold, no prospect list is supplied, and every record in the system arrives from the customer's own system of record or from the buyer who chooses to participate in the room. Recipients can upload their own files, so some of the content is contributed by the counterparty knowingly.

Off the top band on the one category of data the vendor does generate itself: detailed behavioural telemetry on named buyers, covering what they opened, which pages they dwelled on, what they watched and when, assembled into a timeline and fed into a score. Nothing published describes the lawful basis for that observation or gives the buyer a route to see or object to it.

Everything runs on infrastructure the vendor operates, with connections into other systems through published and sanctioned interfaces rather than borrowed access, and the Salesforce integration is native with the compatible editions of that platform named explicitly, which is the kind of specificity that indicates a real partnership rather than a scraped one.

The regulated part of the product carries a genuine conformance position: the European electronic identification and trust services regulation is named in the footer as its own item, and national electronic identity schemes are supported in four Nordic countries.

Off the top band because that position is named rather than detailed, with no statement of which assurance level of signature the product produces, and because the trusted certificate seal that lets a signature be validated in a third party reader is confined to the top tier.

The material this platform holds is unusually sensitive for a sales tool and the model layer sits directly on top of it. Signed contracts are stored and a paid feature extracts data from them automatically. Meeting transcripts and recordings are imported into deal rooms from four named conversation and meeting tools. A knowledge base is assembled from the customer's own content.

Nothing published states whether any of that material contributes to model improvement, whether anything crosses between customer accounts, what is retained after a deal closes, or which processors receive it. A service organisation report is offered on request, which addresses controls rather than the training question. On a product whose corpus is other companies' executed contracts, that is the gap this axis exists to find.

The impersonation record is clean and in this index that is worth stating: nothing here pretends to be a person, there is no synthetic voice, no invented persona, no manufactured research and no cold contact at all, since the buyer is a known counterparty who has agreed to be in the room. The tension sits entirely on the observation side, and it is the same shape recorded against another vendor here.

The buyer is tracked in fine detail, with document opens, page level dwell, video views and file interactions assembled into a complete activity timeline, and a machine learning score is computed on that behaviour to tell the seller how likely the buyer is to sign. Nothing published states that the buyer is told any of this is happening, and the transparency obligations that took effect in Europe in August 2026 are not mentioned. A person reading a proposal is being measured and scored without a stated disclosure.

The breadth is at the top of this index. Around twenty integrations are named individually across systems of record, meeting platforms, conversation intelligence tools, a sales engagement platform, payment processors and human resources systems, with an automations layer claiming more than five hundred further connectors. The interface is documented by surface rather than asserted, covering electronic signature, document generation, and events and webhooks.

A help centre, an academy and a product news feed support it, and the vendor publishes a plain text file for large language models in its own footer, which is an agent readable surface almost nothing else in this corpus offers. What holds it off the top band is that the depth is monetised at every connection point: every system of record integration is an add on rather than an inclusion, on both the middle and top tiers, and interface read access and write access are separately chargeable add ons. A platform sold on living inside the customer's system of record does not include the connection to it at any published price.

No hosting region, data centre location, infrastructure provider, residency option or processor register was located on the pricing or product surface read in this pass. A dedicated security page exists and is the place such information would normally sit, so this row records what is knowable now rather than a settled absence and should be re verified there.

The question carries more weight than usual for this vendor because the product is sold in three languages and six currencies with national electronic identity support in four Nordic countries, so a European buyer signing contracts through it has a specific and foreseeable reason to ask where the executed documents rest.

The security position is set out inside the feature comparison rather than only in marketing, which means a buyer can see exactly what applies at their tier. Two factor authentication is available on every plan including the entry one, a service organisation report at type two is held, the European data protection position is stated with reference to the supervisory board's recommendations, the electronic identification and trust services regulation is named, national electronic identity verification is supported in four Nordic countries, and a trusted certificate seal allows a signature to be validated in a third party reader.

Three things hold it at the bottom of the band. The service organisation report is available on request rather than served, which is the distinction that separates this from the vendors graded higher. No audit period, auditor or information security certification under an international standard is named, and no penetration testing statement, status page or processor list was located. And single sign on is not merely confined to the top tier but sold as an add on within it, so the strongest access control in the product is a separate purchase on the most expensive plan.

The published structure is genuinely good and the real bill is genuinely unknowable, and both halves matter. On the disclosure side: two per seat prices are published at 25 and 49 dollars a month, six currencies are named, a comparison table runs roughly sixty rows across three tiers marking each feature as included, absent or an add on, the five user minimum on the middle tier is printed on the price card rather than buried, and the frequently asked questions state plainly that the two upper tiers bill annually and that every plan carries a twelve month agreement period.

A free trial is available. On the other side sits the longest add on list in this index, and the item at the head of it is the one that matters most. Every system of record integration is an add on, on both paid tiers, with no price attached, on a platform whose central proposition is being used from inside the customer's system of record.

Also unpriced: the configure price quote module, single sign on, interface read and write access, conditional content, automated contract data extraction, unlimited model usage, payment and human resources connectors, and a dedicated success manager. Electronic identity verification and text message delivery both carry a stated warning that additional charges may apply. The published seat price covers a product that cannot connect to the buyer's own systems.

Terms and conditions are published and grouped with the other governing documents in the footer, so the contractual instrument exists and is easy to find, and its contents were not read in this pass. One commitment term is published plainly and it cuts against exit rather than for it: every plan, including the entry one, carries a twelve month agreement period, and the two upper tiers bill annually, so a customer's minimum exposure is a year regardless of which tier they choose.

That is disclosure and it is credited as such even though the term itself is restrictive. What is unestablished is everything else: no export format, no bulk download of executed contracts or room content, no post termination retention window and no deletion timeline. The stake is higher here than for most categories because the corpus at risk is the customer's signed agreements.

Delivery here is transactional and low volume by nature, since a room or document goes to a named counterparty in an active deal rather than to a purchased list, so the reputational exposure that this axis usually measures is inherently smaller. Automated reminders and text message delivery both run from the platform, and text messages are a chargeable feature.

Nothing is published about sending infrastructure, sender authentication, bounce handling, complaint thresholds or what happens if a customer's send outs start landing in spam, and nothing addresses message registration for the text message channel. The grade reflects a modest applicable surface with no stated position on it rather than any observed weakness.

Segmentation runs on more dimensions than anything else in this index and each one is built out rather than asserted. Three buyer roles have dedicated pages covering the account executive, revenue operations and the sales leader. Three company sizes have their own pages.

Seven industries are named explicitly, spanning technology, telecommunications and media, commercial property and broking, wholesale and retail, transport and logistics, utilities and energy, and professional services, each with its own page. The product is published in three languages, billed in six currencies, and carries national electronic identity support for four specific countries, which describes a geographic centre of gravity more precisely than a market claim would.

The five user minimum on the middle tier is a real floor. Off the top band on the pattern that has held others here down: addressing small business, mid market and enterprise simultaneously claims the whole market rather than describing a place in it, and no headcount or revenue band is named anywhere.