PerformLine

The removal test leaves a real product standing, which places this with the platform tier rather than the agent native tier. Strip the models and a discovery and monitoring engine remains: crawlers that find affiliate pages and social posts, a deterministic rulebook engine, and a remediation workflow with an audit trail.

The company is explicit about that architecture in its own guidance, recommending that firms mitigate model risk with human oversight and deterministic rules, which is a description of its own stack rather than generic advice. What the models genuinely carry is the reading: scoring creative, transcribing and evaluating phone calls and messages at volume, and judging whether a claim is misleading.

The newest product is wholly model dependent, since evaluating a conversational system's answers against a firm's own source documents cannot be done with rules. Same shape as the other established platforms in this category.

The published position is specific and operational rather than a reassurance. The company's own guidance names the failure modes it designs against, hallucinated output, bias and lack of explainability, prescribes human oversight paired with deterministic rules as the mitigation, and sets out what an audit ready record must contain: the inputs, the outputs, the reviewer's decision, and the specific rule a flag matched, stored with timestamps.

Tying every flag back to a named rule is a real explainability mechanism and it is the same discipline the remediation workflow enforces. Held at B because the scoring step decides which of millions of assets a human ever examines, and no threshold, confidence measure, recall figure or sampling audit of that triage is published. A reference grade on this axis names what is automated, what constrains it, and how the automated decisions are sampled.

No accuracy, precision or recall figure, benchmark or validation method was located. Flags are traceable to the specific rule that matched them, which makes any individual result explicable and says nothing about how often the system is right across a corpus. The newest product raises a measurement problem this index has not met before and it is worth recording carefully.

Evaluating what a conversational artificial intelligence platform says about a brand means sampling a non deterministic system whose answers vary by prompt, session, user history and model version, so a daily evaluation is a sample of an effectively unbounded space. Nothing published describes how many prompts are run, how they are chosen, or what proportion of the answer space is covered, which means a clean report cannot distinguish a compliant channel from an unlucky sample.

The strongest evidence surface in this competitor set and it rests on three independent legs rather than one. The company states that six of the ten largest United States banks are clients, which is a specific and falsifiable claim at the top of the market rather than a vague scale figure.

Customers appear by name with attributed quotes, including a partner marketplace operator describing oversight of more than 400 direct partners through the platform and a consumer finance brand describing proactive compliance across its fintech partners. And an independent industry awards panel named it best as a service solution at a 2026 banking technology awards programme, which is an outside evaluation of the product rather than a directory listing.

Nineteen years of operating history and a growth equity investor adding follow on capital in 2026 sit behind all of it. What is still missing is a customer publishing a measured outcome, so the quantification is of scope rather than of result.

No boundary statement was located on whether observations from one client inform rulebooks, scoring or models serving another, and the architecture makes the question unusually pointed in two directions. First, proprietary rulebooks built from violations observed across a client base that includes directly competing banks and lenders are, by construction, a shared asset derived from individual firms' mistakes.

Second, the newest product explicitly scores what conversational platforms say about a brand's competitors as well as the brand itself, so competitive intelligence is generated as a routine byproduct of a compliance workflow. Add the affiliate discovery function, which observes marketing published by firms that are not customers at all, and the platform holds a view of a market that no participant in it can see. None of that is described or bounded publicly.

No privacy statement, lawful basis, retention rule or consent framework was located, and the exposure here is larger than for a marketing review tool because of what the monitoring side ingests. The platform scores customer facing text messages, chats and phone calls, which means consumer conversations with a financial institution, containing whatever the consumer said about their circumstances, pass through a third party system.

Call monitoring carries a consent dimension that varies by state and that no public material addresses. The scope also reaches beyond the customer's own estate into affiliate and partner channels, so material is collected from firms that never contracted with the vendor. Held at C rather than D because the institution is the controller with its own notice and consent obligations, and the vendor processes on its instruction.

No information security certification, audit report, penetration testing statement or trust centre was located in this pass, and this is the most conspicuous instance of that gap in the competitor set. The vendor states that six of the ten largest United States banks are clients, and institutions of that size do not onboard a processor that ingests consumer call recordings and messages without an assurance package, so the artefacts almost certainly exist and are simply not published.

Recorded as an absence found rather than a proven absence, and flagged as a strong candidate for correction on a second pass, since the lesson from an earlier correction in this index is that compliance claims sometimes sit on product pages that a security page never repeats.

Correct technology supplier posture with rule level mapping that is more concrete than most of this category manages. The rulebooks encode named consumer finance regimes, the Truth in Lending Act and its implementing regulation, fair lending requirements and the statutory prohibition on unfair, deceptive or abusive acts and practices, rather than gesturing at compliance generally.

More substantively, the partner oversight function exists to discharge a specific named supervisory expectation: banks are held responsible for the conduct of the third parties marketing on their behalf, and this product is how that duty is evidenced at scale. Graded at B because no regulator has examined the product itself and no formal admission, supervised test or regulator run programme was located.

The company publishes commentary naming bias as a risk in artificial intelligence compliance workflows, which is engagement with the topic rather than disclosure about its own systems, and no governance framework, fairness position, testing programme or independent assessment was located. Two product specific exposures deserve stating.

Scoring phone calls and messages means transcription sits upstream of every judgement, and transcription accuracy varies with accent, dialect and speech pattern, so an agent whose speech transcribes poorly accumulates flags for reasons unrelated to what they said, with employment consequences attached.

Separately, the platform evaluates fair lending compliance in other firms' marketing, which makes the absence of any published fairness testing of its own scoring more conspicuous than ordinary silence.

No accuracy warranty, service commitment or remedy was located. The consequence structure has three tiers and the vendor sits outside all of them. A violation the system fails to catch in partner marketing becomes the sponsoring bank's supervisory finding, because the duty to oversee third parties rests on the institution.

An agent flagged by transcript scoring faces a performance or disciplinary consequence, with no correction or appeal path published for a person who is not party to the contract and cannot see the rule that caught them. And the consumer who acted on a non compliant advertisement or an inaccurate answer is further out still, having never chosen the system and having no way to know a model reviewed the material.

No provider, base model, hosting arrangement or build against buy statement was located for the scoring, transcription or evaluation layers. The omission has a distinctive edge in the newest product, where a model evaluates the outputs of named third party conversational platforms and neither side of that comparison is identified: the platforms being watched are named publicly, the model doing the watching is not.

For an institution whose own supervisor expects it to know and document which models its compliance controls depend on, that asymmetry is awkward, since the customer can tell an examiner precisely which systems were monitored and not what did the monitoring.

The integration surface is the channel estate rather than the back office, and it is broad: web crawling that discovers pages the institution did not know existed, social platforms, email, call recording systems, messaging and chat, and now the major conversational artificial intelligence platforms. Reaching into partner and affiliate channels means ingesting from properties the customer does not own or control, which is harder than integrating with systems a client administers.

Held at B because no named integration to a marketing resource management system, a content platform, a customer relationship system or a governance risk and compliance platform was located, and no public application programming interface documentation or connector catalogue was found, which is the layer an A on this axis requires.

Delivered as cloud software with no hosting region, tenancy model, residency option or subprocessor list located. The gap carries more weight than the standard version because of what is stored rather than where the buyer sits: the platform retains consumer call recordings, chat transcripts and text messages as evidence in an auditable compliance record, which is exactly the material a bank's own vendor assessment will trace end to end.

A United States only regime focus means cross border transfer questions arise less often than for a global vendor, but it does not answer where a decade of recorded consumer conversations is held, under whose control, or for how long.

No rate card, tier ladder or billing basis was located and the route to a number is a demo request. A customer testimonial describing the service as costing not a lot is a satisfied buyer's opinion, not a disclosure, and it is if anything less useful than silence because it invites an assumption about affordability that a prospect cannot check against anything.

The pricing question is harder than usual for this product because scope is the variable: a platform charging for channels monitored, assets reviewed, partners covered or volume scanned would produce very different totals for the same institution, and a bank extending oversight across a large partner programme cannot model the cost of that decision from public material.

Depth and breadth are both anchored to something checkable, which is what separates this from a coverage claim. Six of the ten largest United States banks are stated as clients, and the segment spread runs across banks, fintechs, lenders, mortgage, insurance and consumer finance brands, each addressed through separately maintained material.

There is also a multiplier most vendors do not have: because the product supervises affiliate and partner marketing, a single client relationship pulls hundreds of downstream firms under oversight, with one named customer describing more than 400 direct partners covered. Channel coverage spans web, social, email, calls, messages and now conversational artificial intelligence platforms. The clear limit, recorded rather than penalised: the encoded regimes are United States consumer finance rules and no non United States regulatory coverage was located.