Selling.com

Model features are enumerated and the product is bought for the database.

What exists is a reasonable set: workflow automation, a generated email sequence builder, a generated individual email writer, call coaching, a persona builder and voicemail transcription analysis. Users confirm using them, with one describing the persona builder and email personalisation as noticeably improving engagement rates and another describing generated messages for social platform outreach as a time saver.

What places this mid band is where the value actually sits. Across 151 independent reviews the overwhelming subject is data: accuracy, coverage, verification, bounce rates, whether a contact still works at the company. The aggregated themes users like are led by ease of use and accurate contact information; the themes they dislike are led by inaccurate data and outdated contacts. The model layer barely features in either direction. A customer who disabled every generated feature would still have the product they bought.

The positioning confirms it. The vendor's own description leads with 97 percent accurate contact data and lists model based automation fourth, framed as eliminating repetitive work rather than as the reason to buy.

One review reports a discrepancy in the voicemail transcription analysis, which is the only user level accuracy observation on the model layer located.

Ask what the generated email features add over the sequence templates.

The controls that exist are targeting controls, and the sending layer is undescribed.

What users configure is genuine and they describe it in detail. Suppression lists are a named filter. Domain and name exclusion is available and praised. An impression list feature handles exclusions. Inclusion and exclusion by industry, job title, department and seniority is repeatedly cited as the reason the tool is useful. Those are real constraints on who enters a campaign, applied before anything sends, and for an outbound product that is the control that matters most.

Workflows then run automatically. One reviewer describes systems that automatically feed prospects into multi touch sequences and transfer engaged prospects to the record system, another describes setting up workflows to build target lists and send automated mail. Generated messages enter those sequences.

What is not described anywhere retrievable is oversight of the automation itself. Two passes located no approval gate before a generated message sends, no statement of whether a human reviews generated copy, no audit trail, and no administrative permission model beyond a general reference to user, role and access management in a directory feature list.

The vendor's own surfaces could not be examined, so this reflects what a buyer finds without a sales conversation.

Ask whether generated emails require human approval before a sequence sends them.

Six model features are named by function and nothing is named behind any of them.

The capabilities listed are a sequence generator, an individual email generator, call coaching, a persona builder, workflow automation and voicemail transcription analysis. Two passes across the vendor supplied product description, the directory listings and 151 reviews located no model provider, no model name, no version, no statement of what customer or prospect data is transmitted for inference, no training or retention position and no evaluation or accuracy measure for any generated output.

The transcription feature deserves specific mention because a user found the gap independently. One reviewer reports a discrepancy in the voicemail transcription analysis, which is a customer encountering an accuracy problem with no published accuracy claim available to check it against and no error rate to set expectations. Call coaching has the same shape: advice given to a representative about their own calls, with no basis stated for the advice.

The contrast with the vendor's data claims is the point. This company publishes a specific numeric accuracy figure for its contact database and stakes its positioning on it, then publishes nothing at all about the models that write the messages that database feeds.

A buyer running a governance review would find function names.

Ask which provider processes call recordings and generated message content.

The evidence base is substantial, independent, and sits entirely outside the vendor's control.

The principal enterprise review platform carries 151 reviews at 4.6 with the distribution published rather than summarised: 114 at five stars, 27 at four, 4 at three, 1 at two and 2 at one. That spread matters more than the average, because it shows a small but real cluster of severe dissatisfaction alongside broad approval, and it lets a buyer see both. Market segmentation is published at roughly 48.6 percent mid market. The vendor has been serving customers since 2019.

Individual reviews carry figures with names and roles attached: a director of sales reporting four times pipeline generated, a client success representative reporting a 45 percent open rate against a stated national average, a director reporting outbound campaign launch time reduced by two days, and repeated cost comparisons placing the product at a quarter to a tenth of an established competitor's quote.

What holds this below the top band is that all of it is third party. Two passes could not load the vendor's own website, and no vendor published case study, methodology or measurement window was located anywhere. The 97 percent accuracy figure that anchors the entire proposition has no published basis.

That inversion is unusual and worth noting: the independent evidence is stronger than the vendor evidence.

Ask how the 97 percent accuracy figure is measured and over what sample.

Suppression is a first class feature and the compliance documentation could not be found.

What exists in the product is real. Suppression lists appear as a named search filter rather than an afterthought. Domain and contact exclusion is available and users cite it as a reason the tool works for them. A directory feature listing includes call scrubbing under automation, which in a dialer context means screening against do not call registers. Real time verification before send reduces contact with stale records.

What could not be located across two passes is any published compliance position: no privacy policy, no processing agreement, no consent basis for the database, no data subject rights process, no regional privacy statement and no calling hours or register documentation. The vendor's own website did not load, so this reflects retrieval failure rather than confirmed absence, and a buyer with a browser may find these documents.

The reason it matters here more than for most records is the combination on offer. This vendor supplies 75 million mobile telephone numbers and ships an auto dialer to call them. Mobile numbers plus automated dialing is the precise configuration United States telephone consumer regulation governs most tightly, and the consent basis for holding those numbers is the first question a regulated buyer's counsel will ask.

Ask for the consent basis for the mobile number database and the register screening method.

The scale of personal data held is established and the vendor's own privacy position could not be retrieved.

What is established from the vendor's supplied description and independent reviews: the platform holds 220 million contact profiles across 11 million accounts, including 75 million mobile telephone numbers, and enriches customer records against them. Users describe using it to verify whether an inbound lead is a real person at a real company, which is a resolution use rather than a prospecting one. Call recording is a listed capability. That is a substantial holding of personal data about people who have no relationship with the vendor.

What exists on the control side is customer facing rather than data subject facing: suppression lists, exclusion filters and real time verification, all of which serve the buyer rather than the person in the record.

Two passes could not load the vendor's website, and no privacy policy, processing agreement, subprocessor list, retention schedule, residency statement or data subject rights process was located through any other route. That is a retrieval limitation rather than a confirmed absence, and the grade reflects it as such rather than treating an unreachable document as a missing one.

For a company whose entire product is other people's contact details, the privacy documentation is the material a buyer's counsel reads first.

Ask how an individual in the database requests access or erasure.

A specific numeric accuracy claim anchors the product, and the independent review base contradicts it in volume.

The claim is 97 percent accurate contact data with real time verification, across 220 million profiles and 75 million mobile numbers. The review platform's own aggregated dislike themes read as a direct rebuttal: inaccurate data leading to less valuable leads across seven reviews, high bounce rates and outdated contacts across five, outdated information across four, data inaccuracy including high bounce rates and regional discrepancies across three, further inaccuracy and regional discrepancies across three, missing emails across three, and outdated data including incorrect addresses across three. Six separate reviews praise the 97 percent figure, so the picture is genuinely divided rather than uniformly negative.

Two individual reports are specific enough to record. A sales development representative rating the product at zero states the bounce rate runs 40 to 50 percent. A head of lead generation states that after enabling the real time verification setting, a manual verification pass found many contacts outdated, no longer at the company, in changed positions or in the wrong location.

Provenance is absent entirely. Two passes located no named upstream supplier, no licence, no collection method and no consent basis for any of the 220 million records. Multiple reviews describe the browser extension surfacing contact details alongside a professional network profile, and the vendor describes that extension as ungating contact data, which points at an origin without establishing one.

Regional coverage is reported as materially weaker outside North America.

Ask which suppliers the database draws on and how the 97 percent figure is verified.

The professional network relationship runs through the product repeatedly and its basis is not established.

What users describe: a browser extension that displays contact details alongside a professional network profile while the seller is looking at it, syncing with that network to add prospects manually during list building, and a database whose leads reviewers describe as correlating in parallel with specific profiles on that network. One reviewer describes access to leads verified against that network with viable addresses and direct telephone lines. The vendor's own description of the extension is that it ungates contact data.

A sales navigator integration is listed among the product's three recorded integrations, and that platform does operate a sanctioned partner programme, so a listed integration may well be licensed. The two possibilities look identical from outside: a partner integration surfacing permitted data, or an extension operating over pages the vendor does not own.

Two passes located no statement of interface access, no partnership announcement and no acknowledgement of the question. The vendor's own site did not load, so material addressing this may exist and was not reachable.

Beyond that surface the record is clean, with no social automation, no account renting, no undetectability marketing and no evasion claim located anywhere.

The honest position is insufficient evidence rather than a finding.

Ask whether the browser extension operates under a licensed partner integration.

Model output reaches prospects and coaches representatives, with nothing published on either side.

On the model side, two passes located no provider, no training or retention position, no accuracy figure, no evaluation and no error rate for the sequence generator, the individual email generator, the persona builder, the call coaching or the voicemail transcription. One reviewer independently reports a discrepancy in the transcription analysis, which is the only quality observation available and it comes from a customer rather than the vendor.

Call coaching deserves separate mention. Advice generated about a representative's own calls influences how that person sells, and where it is wrong the cost is absorbed silently by a seller who assumes the tool knows better. Nothing describes what the coaching is based on or how often it is right.

On the stewardship side nothing was located: no encryption statement, no access control description, no retention schedule for call recordings, no incident response or breach notification process, no named security contact and no vulnerability disclosure route.

The holdings make the absence consequential. This platform stores 220 million contact profiles, 75 million mobile numbers and call recordings of conversations with prospects.

The vendor's own website could not be loaded, so documentation may exist beyond retrieval.

Ask what the call coaching recommendations are trained on and how accuracy is measured.

The sender is genuinely the customer and one user's description of why that matters is worth recording.

Mail leaves through the customer's own connected mailboxes, with multiple mailbox integration a named capability. A reviewer running an agency describes the mechanism approvingly: the system lets you add a mail account and market directly from the client's inbox, as if you are manually sending from your personal work address. That is the value proposition stated plainly by a user, and it sits adjacent to the undetectability pattern recorded elsewhere in this sweep without being marketed as such. The intent is deliverability and familiarity rather than concealment, and the distinction is real, but the effect on the recipient is that an automated sequence is indistinguishable from a personal message.

On identity the record is clean. Two passes located no persona, no rented account, no synthetic sender, no cloned voice and no undetectability claim.

What is absent is disclosure of authorship. Generated sequences and individual messages go out under a named person's address and nothing indicates to the recipient that a model wrote them.

Call recording is a listed capability and no recording notification behaviour or consent handling was located, which matters in jurisdictions requiring all parties to agree.

Ask what a recipient hears when a recorded call begins.

Three integrations for a data platform, and the browser extension does most of the work.

What is confirmed: integrations with a major record system and with a professional network's sales product, recorded as three integrations in total on the review platform, with users additionally reporting a working connection to a second record system. Reviewers consistently describe record system synchronisation and export as functioning well, and one describes transferring engaged prospects into the record system automatically at the end of a sequence.

The browser extension is the surface users praise most and treat as the product's daily interface, described variously as a daily must have and the reason the tool fits into existing workflow. Data import and export tools are listed capabilities and heavily used.

What limits the grade is the count and the depth. Three integrations is thin for a platform whose output has to reach a customer's system of record, and one reviewer explicitly asks for a tighter integration with the record system they use. Two passes located no interface documentation, no authentication scheme, no webhooks and no model context protocol server, though a directory feature listing does reference interfaces generally.

The vendor's own site could not be loaded and a documented interface may exist.

Ask whether a documented interface exists and what it exposes.

Nothing about where any of this runs could be established.

Two passes across the vendor supplied description, directory listings and 151 reviews located no hosting provider, no region, no data centre, no residency commitment, no tenancy or isolation model, no encryption statement, no backup position, no continuity plan, no uptime commitment and no status page. The vendor's own website did not load across repeated attempts, so material may exist that could not be reached.

The only geographic signals available are indirect and concern coverage rather than infrastructure. Multiple reviewers report the database performing materially better in North America than in the Middle East and Asia Pacific, and one describes support staff working from the other side of the world, which speaks to staffing rather than to where data sits.

The holdings give the question weight. A database of 220 million contact profiles including 75 million mobile numbers, plus call recordings, plus customer uploaded lists, is exactly the material a buyer with a residency requirement has to locate before signing, and European buyers in particular would need an answer before processing any records of people in that region.

A buyer will have to obtain this in a sales conversation rather than from published material.

Ask where the database, customer lists and call recordings are stored.

No credential of any kind could be located.

Two passes across the vendor supplied product description, directory profiles and 151 independent reviews located no service organisation control report, no international standard certification, no penetration test, no trust portal, no security page, no encryption or access control statement, no vulnerability disclosure route, no named security contact and no subprocessor list. Not a badge, not a claim, not a mention. The vendor's own website could not be loaded, so a trust page may exist and be unreachable by this route, and the grade should be read with that limitation in view.

What makes the absence consequential is the customer base and the holdings together. Reviewers include enterprises above a thousand employees, financial services firms and manufacturers, all of whom run supplier security reviews as a matter of course. The platform holds a database of 220 million people, 75 million mobile numbers, customer uploaded prospect lists and call recordings.

The economics of the positioning may explain it. This vendor competes on being a quarter to a tenth the price of the established alternatives, and independent audit programmes are a real recurring cost that the incumbents carry and price in.

Ask whether any independent security audit has been completed and what it covered.

The positioning is entirely about price and the metering unit is where buyers report being caught.

What is known: four plans exist, the entry point is a free browser extension for a single user, and a free trial is available. Reviewers supply the real anchors, consistently and independently. One reports an established competitor quoting over ten thousand dollars annually against this vendor at a quarter of that. Another puts it at a tenth of the same competitor. A third notes approvingly that the vendor has not let pricing spiral upward as it has grown. Being cheap is the proposition and customers confirm it holds.

The finding sits in the meter. Contacts are exported against credits, and one reviewer states that their contract said unlimited contact credits while a cap existed in the fine print, that they ran out unexpectedly, that they advise others to ask specifically about the export cap, and that they left for a competitor over it, calling it a red flag. The review platform's aggregated dislike themes separately record contact credit misunderstandings across seven reviews, so this is a pattern rather than one bad experience.

A contract using the word unlimited alongside an undisclosed cap is the sharpest commercial finding available on this vendor.

No published rate card was retrievable and the vendor's own site did not load.

Ask what the export and contact credit caps are, in writing, before signing.

Getting data out is a core daily function and it is capped.

Export is not an afterthought here, it is how the product is used. Reviewers describe exporting searches quickly and efficiently, bulk exporting prospect data, pushing records into their record system automatically at the end of a sequence, and using import and export tooling as listed capabilities. For a data platform that is the portability question largely answered in the ordinary course of operation, and it puts this ahead of records where export exists only as a support request.

Two constraints qualify it. Export is metered against credits, and one reviewer reports being capped without having understood the cap, while another reports the platform limiting how many contacts per company can be downloaded. So the route out exists and has a ceiling the buyer needs to establish in advance.

What is not addressed anywhere retrievable is the rest of it: no account closure process, no deletion commitment, no retention statement after cancellation, and nothing covering whether sequences, campaign history, call recordings and engagement analytics travel with the contact records or stay behind.

The accumulated asset for a long standing customer is the campaign history rather than the contact list, since the list can be rebuilt and the history cannot.

Ask what the export ceiling is and whether campaign history and recordings export.

Verification before send is the central mechanism, users connect it directly to domain health, and its efficacy is contested.

The mechanism is named and prominent: real time verification of addresses before export and send, marketed by the vendor as improving deliverability rather than only as a data quality feature. A reviewer states the connection explicitly, that emails often bounce and ruin domain health and that this does not happen here because of the verification. Multiple mailbox integration supports rotation across sending accounts. Suppression and exclusion lists prevent re contact. One reviewer describes coming to the platform specifically because warming an inbox had been difficult elsewhere.

That is a real discipline stack for a data vendor, and more than several dedicated sending tools in this index describe.

The contest is over whether it works. Several reviewers report bounces persisting after verification, including one stating that even after data verification there are still several bounces, and one reporting a bounce rate of 40 to 50 percent. The mechanism is credited on this axis and the accuracy dispute is recorded under provenance, where it belongs, because the failure mode described is stale data rather than absent discipline.

No authentication guidance, warmup schedule, sending limits or measured deliverability data was located.

Ask what the measured bounce rate is across customers over a stated period.

The buyer is well established from independent evidence and the geographic limit is reported by customers rather than stated by the vendor.

The review platform publishes market segmentation at roughly 48.6 percent mid market, with small business heavily represented and a smaller enterprise contingent above a thousand employees. That is a credible distribution for a product whose entire positioning is being a quarter to a tenth the price of the established alternatives, and reviewers confirm the motive directly, repeatedly describing arriving after a renewal quote from a larger vendor.

Vertical spread is wide by design. One reviewer values the platform precisely because it reaches decision makers whether the target is an oil change business or an insurance company, and reviews span logistics, electrical manufacturing, financial services, telecommunications, oil and energy, staffing, marketing agencies and software.

The limitation customers report consistently is geographic. Coverage is reported as materially weaker in the Middle East and Asia Pacific, particularly for mid to senior level profiles, and regional discrepancies in contact information appear across several aggregated dislike themes. A buyer selling outside North America should treat the database as thinner than the headline profile count suggests.

The vendor publishes no coverage breakdown by region, which is the disclosure that would let a buyer check this before purchase rather than after.

Ask for profile counts and verified rates broken down by region.