Leadsforge

The vendor's own metering settles the removal test, the same way its sibling product's pricing architecture did. The pricing page states that searching and refining a list costs nothing and that credits are drawn only when results are enriched or exported, with a rate card pricing each verified record by type.

So the conversational layer, which is the whole marketed differentiator, is given away, and what the customer actually buys is verified contact data from a cascade of third party providers. Take the model out and a five hundred million record database with waterfall enrichment remains, which is a complete and saleable product and is what every filter driven competitor sells. The model here is the retrieval interface and the qualification step rather than the substance.

That places it in the middle of the band on the removal test rather than at the top on vintage. Ask what the conversational layer returns that a filter query on the same underlying providers would not.

Nothing sends and nothing runs on a schedule, and the one route that could act without a person is undocumented on this point. A person describes a buyer, reviews a list and chooses what to enrich, so each spend is a deliberate act, and qualification prompts run across the returned set rather than initiating anything.

The published containment is financial and it is a good one: search costs nothing, so an exploratory session cannot run up a bill, and only enrichment and export draw credits against a balance the customer has already bought. The gap is the machine interface.

An external assistant tool connected to it can drive lookups that spend credits, and nothing published states whether an administrator can restrict that route by role, cap what a connected client may draw in a period, or revoke it independently of the account. Ask what limits apply to spend initiated through the machine interface and who can turn it off.

The interface for machines is documented and the model behind the product is not. On the disclosed side, the vendor publishes that it operates a machine interface for external assistant tools and names two specific clients it connects to, which is a precise and checkable statement about how the product can be driven.

On the undisclosed side, no provider, model family or version is named anywhere for the layer that interprets a plain language buyer description into a query, nor for the qualification prompts the vendor states are run against every contact to detect buying intent. Nothing describes how a description becomes a filter set, what the qualification step actually reads, or how a customer would know the interpretation had drifted from what they asked for.

The trust portal lists control titles covering artificial intelligence governance with no document published behind any of them. Ask what model interprets the search and what the qualification prompts read.

Comparative claims without a measurement, on a product whose whole value proposition is data quality. The vendor states that lists come out cleaner than single source tools manage and that its cascade produces higher match rates, and publishes no match rate, no accuracy rate, no bounce rate and no verification sample behind either claim.

For a product billing per verified record, the verification hit rate is the number that determines what a customer actually pays per usable contact, and it is absent. A case study section exists on this product's own site and the case studies reachable through the group's channels name the sibling sending platform and the infrastructure products rather than this one.

One modest positive is procedural: the privacy policy commits to obtaining a customer's authorisation before publishing their testimonial, and gives a removal route, which is a consent practice most vendors leave unstated. Ask for the match rate and the bounce rate on a sample list built to your own buyer definition.

The product supplies the material for outreach and conducts none of it, and the obligation transfers in writing. The published policy places the customer as controller of the contact lists they build and the vendor as processor, and states that where a person contacted through the group's sending platform asks for modification or deletion the vendor will honour it and inform the customer. That is a clear allocation as far as it goes. What is missing is specific to what this product sells.

Mobile telephone numbers are sold at ten times the credit cost of an email address, and telephone numbers are the category carrying calling restrictions in most markets, yet nothing published describes screening against any do not call register, suppression of previously objecting individuals, or jurisdictional restriction on which numbers are supplied. Ask whether any number supplied has been screened against a do not call register in your calling territories, and if not, who is expected to do it.

A substantive instrument for the customer, published under another product's name and predating this one. The substance is real and better than most: legal bases are stated purpose by purpose rather than asserted in general, retention is given as actual periods including three years after account closure and five years for evidentiary material, the full rights set is enumerated including portability in machine readable form and instructions for data after death, requests are answered within a stated thirty days, and cookie consent choices are held for a stated six months.

Against that, the document names the sibling platform throughout as the controller, gives that platform's website as the site it covers, was last updated more than a year before this product came to market, and describes the handling of contact lists a customer uploads for sending rather than anything this product does. A buyer cannot establish from it which entity processes their data here. Ask for a privacy notice that names this product and its database.

The sourcing mechanism is described precisely and every source is anonymous. What the vendor does explain is unusually clear for this category: each lookup cascades through several third party providers in sequence until one returns a verified match, addresses are validated in real time, profile addresses are confirmed against live pages and telephone numbers pass format and validity checks, so a buyer understands how a record reaches them.

What is never stated is who any of those providers are, on what licence their data is received, what permits onward resale, or what lawful basis supports holding five hundred million contact records. One route is named and it is the most exposed of them: the followers of a competitor's company page, extracted and sold at one credit each, which is collection rather than licensing and carries no stated basis at all. Ask for the provider list, the licence terms permitting resale, and the lawful basis for the follower extraction route.

A product feature systematically harvests a professional network and the vendor states a conformance position for a different platform entirely. The feature is explicit and priced: point the product at a competitor's company page on that network and it returns the followers, filterable by job title, seniority and geography, charged at one credit per follower, with the vendor's own material presenting this as a cleaner intent proxy than industry filtering.

Profile addresses are separately confirmed against live pages. Nothing published states how any of this access is obtained, whether it is sanctioned, or what happens to a customer whose account is restricted. The asymmetry is the finding: the privacy policy carries an explicit undertaking to adhere to one large platform's interface policy including its limited use requirements, so the vendor plainly knows how to state a conformance position and has not stated one for the network its product reads hardest. Ask how follower extraction is performed and under what permission.

Governance titles are published and the two questions this product actually raises are unanswered. The trust portal lists a set of controls addressing artificial intelligence governance by name, covering awareness, change planning, communication, competence, concern reporting and continual improvement, monitored alongside more than a hundred others. That is more than most vendors of this size publish and it is worth recording.

But the portal reports no documents at all behind any control, so the titles are the disclosure. Neither of the questions specific to this product is addressed anywhere: whether customer searches, buyer descriptions or enriched results are used to train or tune the interpretation layer, and what an external assistant tool connected through the machine interface may retain once contact data leaves through it. The second matters because contact records for identifiable people can be pulled into a third party client the vendor does not control. Ask for both commitments in writing.

Five hundred million people are held and sold, and no published instrument addresses any of them. The privacy policy at this product's own address covers the customer, the customer's account data and the contact lists the customer uploads for sending. There is no notification practice, no opt out route, no removal portal, no suppression register and no statement of lawful basis for the people whose verified work addresses and personal mobile numbers are the inventory.

The one deletion route in the policy is narrower than it first appears: it covers a person who has received email through the group's sending platform and asks for removal, which is the customer's sending activity rather than the database, so a person who has never been contacted has no described route at all. Sharpest at the follower extraction feature, where individuals are harvested for following a company page and sold as intent. Ask what notice, if any, a person in the database receives, and for the removal route.

A shipped machine interface for external assistant tools, which very few records in this index have at all. The vendor publishes that assistant clients can query the product directly through it and names two specific ones, so this is a working route rather than a roadmap item, and it is the deepest form of integration a data product can offer because it lets another system do the asking.

Around it sit conventional routes that are named and checkable: one click transfer of a finished list into the group's sending platform with no export step, file export at any time, a browser extension published on the official extension store under the sibling brand, a public feature request roadmap and a help desk. It stops below the top band because the machine interface is the only programmatic route documented. No conventional interface reference, no webhooks and no developer documentation for this product were reached on either pass, and no marketplace listing exists under its own name.

A residency commitment stated as an exclusive rather than a preference, with the onward transfer paragraph actually completed. The published policy states that the host servers on which the company processes and stores its databases are located exclusively within the European Union, which is a stronger form than the usual assurance that data may be processed in various regions.

The transfer question that most privacy policies skip is answered directly: where a processor sits outside the Union, the transfer rests either on an adequacy decision or on a specific contract incorporating the standard contractual clauses. Behind that, the trust portal publishes a subprocessor list running to fifteen named third parties including the cloud host, so a buyer can see who touches the data rather than being told that safeguards exist. It sits below the top band because no region within the Union is named, no residency election is offered, and nothing states which products in the group the subprocessor list covers.

A live trust portal with real certifications and a request route, claimed on this product's own pages rather than inferred from its parent. That distinction decides the grade: the badges sit in this product's own footer and link out to the portal, so the vendor is publishing the claim here rather than an assessor reaching across to a corporate relative. The portal itself is substantive.

Two certifications are current, a service organization control attestation of the first type and a cloud application security assessment at the second tier, held across three frameworks, with twenty five policies named by domain covering access control and least privilege, change and release management, risk management, remote access, physical security and data subject rights, and more than a hundred and thirty controls described as continuously monitored.

A security questionnaire is available through a named request process. Held below the top band because the portal reports no documents published at all, no auditor is named, and the attestation is the point in time type rather than the period type. Ask for the report and its auditor, and for the scope statement identifying which products it covers.

A buyer can compute their own bill precisely, and the headline saving on the page does not survive arithmetic. Everything needed is published: one base plan at a stated monthly rate with a stated credit allowance, a rate card giving the credit cost of each record type with mobile numbers at ten times an address, unused credits carried forward and stated never to expire, cancellation at any time with no contract, invoicing confirmed, and a hundred credits on signup without a card.

Search is explicitly free with credits drawn only on enrichment and export. The defect is the annual claim. The plan card advertises two months free, but the published annual figure is exactly twelve times the published monthly figure, so the buyer pays the same either way and the discount is delivered in credit volume rather than in money.

The vendor's own question section states the saving as over fourteen percent, which is the credit uplift correctly expressed and does not match the headline. The add on credit slider did not render to a machine, so those rates were not read and are not asserted absent. Ask which figure governs the annual order.

Leaving is cheap, documented and stated as a commitment rather than left to be discovered. Lists export to file in one click or move straight into the group's sending platform, and the vendor states in its own words that there is no extra charge for taking data out and no lock in on what has already been enriched, which is a direct answer to the usual practice of metering export separately.

The policy adds that contact lists can be recovered from the account at any time through an export control and that contacts can be modified or deleted from the account directly, and it enumerates a right to receive personal data in a standard machine readable format. Subscription mechanics support the same position: no contract, cancellation at any time, and credits that carry forward indefinitely. It stops below the top band on the question that follows from those same mechanics. Nothing published states what happens to a credit balance that never expires once a subscription is cancelled. Ask whether unused credits survive cancellation or are forfeited.

No sending apparatus here, and the upstream contribution to deliverability is real but unmeasured. The product operates no mailbox, no warmup, no rotation and no throttling, because the list moves into whatever tool the customer sends from, and the group sells that sending platform and its infrastructure separately. Graded in the middle of the band with that scope stated rather than at the floor, since the floor asserts an undisciplined sending product and this does not send.

The genuine contribution is validation before supply: addresses are described as validated in real time at the point of enrichment, profile addresses confirmed against live pages, and telephone numbers checked for format and validity, and a list that bounces less protects the sending reputation of whatever tool receives it. What is absent is any figure. No bounce rate, no validity rate and no accuracy rate is published for any record type. Ask for the measured bounce rate on exported addresses and how recently each was validated.

The buyer is clear and narrow, and the coverage behind the headline number is not described. The commercial shape states the segment better than any positioning page could: one plan at a single low monthly rate, credits bought in the application, cancellation at any time and a free allowance without a card, which is a product built for an individual representative or a small outbound team rather than for procurement.

Independent accounts place it with sales representatives, development representatives and small outbound teams, and there is no enterprise tier, no seat structure and no negotiated path. Against the stated five hundred million contacts, no country list, no regional breakdown, no coverage rate by market and no industry or seniority distribution is published anywhere reached, so a buyer selling into a specific territory cannot establish what proportion of it is held before paying. Ask for coverage counts by country and seniority for the markets you sell into.