Outplay

Established platform where the removal test leaves the whole product standing, and the vendor's own metering proves it. Model features are sold as monthly credit allowances that step up by tier, at two hundred and fifty, two thousand and five thousand generated emails a month, and the autonomous prospecting agent is a separately priced add on rather than part of the platform. A capability billed by the unit on top of a seat price is an addition to a product, not the product.

Strip the model layer and the buyer keeps multi channel sequencing across six channels, a full dialer with recording and parallel dialing, a shared inbox, trigger automation, split testing, a browser extension, bi directional connections to seven systems of record and the entire reporting suite. This is the established platform convention applied rather than a harsh read.

Administrative controls exist, are real, and sit almost entirely behind the top tier. System logs appear from the middle plan upward, while custom roles and permissions, team management and federated sign on are reserved for the enterprise step, so the buyer most likely to be running an unattended agent on a small budget has the fewest controls available to them.

Sequence rulesets, scheduled pauses, holiday calendars and time off settings are genuine guardrails on when automation may fire and are available more widely. Held at the middle band because the autonomous prospecting agent sold as an add on is described as researching, writing and booking meetings on its own, and no approval step, review queue, escalation threshold, guardrail or audit trail of agent actions is published anywhere for it.

No base model, provider, family, version or hosting arrangement appears on either page read. The product runs a mail writer, a sequence writer, an objection handler, an out of office detector, a revenue intelligence layer and an autonomous prospecting agent, and every one of them is described by what it does and none by what it is built on.

The metering is precise to the individual generated message while the estate behind it is undescribed, which is the same asymmetry recorded at several vendors this session: the commercial layer is specified to two decimal places and the technical layer is not specified at all.

Broad named attribution and not one quantified customer result. Five testimonials carry a full name, a role and in four cases a named employer, spanning revenue leadership at a video platform, a growth director, a business development manager and a chief revenue officer. Around fifteen customer logos appear across the two pages read, several of them recognisable technology companies in conversational intelligence, health data and revenue forecasting.

Against that, no case study, no measured outcome and no method appears anywhere, the only headline figure on the site is a platform claim of five thousand salespeople, and two specific defects sit in the evidence itself. Every one of the nine independent review badges displayed is dated to a single quarter of 2024, more than two years stale on a 2026 page.

And one testimonial block pairs a headline claiming a thirty percent rise in outbound success with a quote entirely about earning affiliate commission, so the one quantified assertion on the page has no supporting statement beneath it.

Four separately regulated channels are sold and no compliance position was located on either page read. The platform sends email, text messages, messages through a consumer messaging network offered as a paid add on, and voice calls through a built in dialer with parallel and power dialing, and each of those carries a distinct statutory regime in most jurisdictions.

No statute, consent requirement, unsubscribe obligation, suppression mechanism, calling window or complaint route appears anywhere in the material read. The one adjacent position the vendor does take is about list quality rather than consent, arguing against bundled contact databases because shared lists produce mass outreach to low quality records. Terms of service and any acceptable use policy were not read and are flagged as the documents most likely to carry an answer.

No privacy policy, cookie policy or processing addendum was reached on either page read, so this row is held at the middle band and flagged for verification rather than graded on inference. The processing in scope is substantial and would need covering: connected mailboxes at up to fifteen per user, call recordings retained for two years by published policy, transcription, prospect records stored to a hundred thousand and beyond by tier, browser extension capture from a professional network and four systems of record, and messaging through a consumer network.

One structural point in the vendor's favour is verified and relevant here, namely that the vendor sells no contact database, so the personal data on the platform originates with the customer rather than with the vendor.

The vendor takes a deliberate provenance position, publishes the reasoning behind it, and the reasoning is sound. It refuses to bundle a contact database into the platform and says why on its own homepage: combining automated outreach with lead scraping in one tool leads to mass outreach against low quality lists shared between customers, which burns the records, damages deliverability and risks blocklisting. The customer brings their own scrubbed list.

That is an architectural choice defended on operational grounds rather than an absence, and almost nothing else in this category articulates it. Off the top band because the browser extension pulls prospects from a professional network, its recruiter product, a mail client and four systems of record with no stated method, permitted scope or lawful basis, and because email validation is included with no source named for the checks it performs.

Three exposures sold openly as priced features, with no conformance position anywhere. Professional network automation covering automatic connection requests, messaging and engagement is a twenty five dollar monthly add on, alongside browser extension capture from that network and its recruiter product.

Phone numbers are offered with local presence in more than fifty countries and automatic rotation between them, which is the mechanism that shows a recipient a number matching their own area code rather than the caller's real location. Messaging through a consumer network is inserted into sequences with no position on that network's business messaging rules.

Held at the middle band rather than lower because nothing is marketed as evading detection, sending runs through the customer's own connected mailboxes, and the vendor's posture toward connected platforms is otherwise cooperative. Worth recording that the same vendor argues carefully against one deliverability risk while selling three platform conformance risks beside it.

Nothing published on either page read states whether customer content is used to train or improve any model, whether prompts are retained by any provider, or whether anything crosses between accounts. The question has real weight here because of what the platform holds: mailbox contents across up to fifteen connected accounts per user, call recordings kept for two years with transcription, prospect records, and the reply history feeding the revenue intelligence layer.

A model writes messages, handles objections and, on the add on tier, prospects autonomously, all of it grounded in that material. No stewardship commitment, non use statement or provider retention arrangement was located, and the legal documents were not read and are flagged.

Several features bear directly on whether a recipient can tell what they are dealing with, and none carries a disclosure position. Local presence numbering with automatic rotation across more than fifty countries presents a caller as local when they are not, which is the exact inverse of branded calling and the same tension recorded at other dialers in this index.

Content randomisation is offered on every tier including the free one, and its function is to vary the surface of a bulk message so it does not read as one. Generated mail, generated sequences, generated objection handling and an autonomous agent that writes and books meetings all reach real people with nothing stating a machine composed the message. No position on article 50 of the European artificial intelligence regulation appears anywhere.

One detail cuts both ways and is worth recording: the enterprise tier allows the call recording bot to be given a custom name, which is either sensible branding or a way to make an automated recorder read as a person, and nothing published says which use is intended.

Genuinely wide and honestly described. Bi directional connections are published to five systems of record, a conversation intelligence platform, two recruiting systems, a scheduler, a product analytics tool, a workflow broker, a video platform and a messaging platform with four distinct depths of integration up to private channel selection and deal rooms. A browser extension works across a mail client, a professional network and its recruiter product, and four systems of record.

Inbound website forms feed sequences directly. The detail most worth crediting is a constraint rather than a capability: the comparison table states plainly that only one system of record may be connected at any one time, which is a real limitation a buyer needs and which most vendors would leave to be discovered after purchase.

Off the top band because interface access sits above the entry tiers, federated sign on is reserved for the top plan, no webhooks, protocol server or developer documentation subdomain was located, and that single connection limit is a hard ceiling for any team running two systems.

Nothing on hosting provider, country, region, data centre or residency option appears on either page read. The only geographic signals available are indirect and inconsistent: static assets and uploads are served from a storage bucket in a western United States region, one open graph tag points at a locale specific path implying regional site variants, and independent review badges are issued for three separate world regions, which together suggest an international customer base without answering where any of their data sits. That matters more than usual for a platform storing two years of call recordings and connected mailbox contents. No security or trust page was reached and is flagged as the likely home of an answer.

No certification, audit, attestation, penetration test, trust page, status page or vulnerability route was located on either page read, and this row is flagged accordingly rather than graded as an established absence. What is verified is thin and tier bound: federated sign on through two named providers is available on the top plan only, custom roles and permissions likewise, system logs appear from the middle plan, and the enterprise tier is positioned on compliance and security without naming a single control or standard behind that phrase.

Set against a platform holding sending credentials for up to fifteen mailboxes per user and two years of call recordings, the absence of any located assurance is material, and confirming whether a security or trust page exists elsewhere on the site is the highest value verification available on this vendor.

The most completely priced vendor in the index to date, and the vendor argues the case for it on its own homepage, attacking the two category leaders by name of position for hiding their pricing and charging for the support a buyer needs to use the product. Four product lines each carry their own ladder.

The engagement ladder runs a permanent free tier at zero with a real allowance of two thousand active prospects a month and a long feature list, then published per seat prices at each of three paid steps, with monthly and annual terms and the annual saving stated. The entry plan carries an interactive slider so a buyer can price their own prospect volume across four bands rather than guess which tier they land in.

Every gating quantity is published per tier across ten feature categories, covering prospect storage, mailboxes, sequence limits, generated email allowances, transcription minutes, recording length and recording retention. Above all, eight add ons each carry their own published rate, covering warmup per mailbox, the autonomous agent, extra mailboxes, calling minutes, phone numbers, consumer network messaging, professional network automation and conversation intelligence.

Priced add ons are the single most common place this axis breaks elsewhere in the index, and here every one of them has a number. Signup is self serve with no card. Two minor gaps keep it from being flawless rather than from being top band: six of the eight add on rates carry a from prefix with no ceiling, and the additional user rate appears as a feature line with no figure.

One commercial freedom is published and the data question is untouched in the material read. Cancellation is stated as available at any time alongside upgrade and downgrade, and the free tier is stated to have no expiry, so a customer is not locked in by term. Retention figures are published for one asset class, with call recordings kept for one year on the middle plan and two on the top plan.

Beyond that, no export function, file format, bulk retrieval route, post termination window or deletion timeline was located anywhere. What a departing customer would want back is substantial: prospect records to a hundred thousand and beyond, sequence libraries, reply threads across six channels, call recordings and transcripts, and the reporting history. Terms of service were not read and are flagged as the document that would settle this.

The strongest thing here is a published architectural refusal rather than a feature. The vendor declines to bundle a contact database into the product and explains the reason on its homepage in deliverability terms: outreach tools that also scrape leads push customers toward mass sending against low quality lists shared between accounts, which burns records, drives down inbox placement and risks blocklisting.

Defending a design decision on sender reputation grounds is rare and is the same shape that earned a top grade for a dialer elsewhere in this index. Real mechanics sit alongside it, including included email validation, custom sending domains, mailbox rotation and warmup offered per mailbox.

Off the top band on four counts: warmup is a priced add on rather than included, which is the opposite treatment given by a much weaker vendor in this same session; unlimited emails are stated on every paid tier with no per mailbox ceiling published anywhere; automatic mailbox rotation is available on every tier including the free one, and rotation presented as a way to scale volume is not the same as discipline; and content randomisation is offered throughout, whose function is to vary the fingerprint of a bulk message. No complaint threshold, bounce handling policy or authentication position was located.

The market position is stated more sharply than most, including the part vendors usually leave out. Each tier carries a plain sentence naming its intended buyer, running from early stage startups with small teams through fast growing companies with larger teams to organisations needing compliance and security. Four product lines address different shapes of buyer, including a dedicated agency edition that manages and bills client accounts.

Most usefully, the homepage states who the product is not for by describing who its competitors are for, positioning the two category leaders as enterprise tools and this product as built for growing smaller businesses, which is an unusually direct account of fit. Around fifteen named customer logos corroborate the claim across technology, health data and video.

Off the top band because no customer count beyond a platform wide salesperson figure is published, and no region, country, language or industry coverage statement appears anywhere despite review badges issued across three world regions.