LinkSprig

Generation runs through targeting analysis, written intros, and an auto reply assistant, and all of it is marketed as the reason to buy. The removal test still leaves a working product: automated connection requests, follow up sequences, a unified inbox with tagging and reminders, meeting booking and lead export survive intact. The vendor's own packaging makes the point better than any analysis could.

Automation and campaigns are both sold as unlimited while the generated messages are rationed at 10 a day on the trial and 25 a day on the paid tier, so the model layer is the scarce input and the sending is the free one.

Two vendor statements about control sit on the same site and cannot both hold. The disclosure page states the buyer stays in full control of what gets used or sent, and the frequently asked questions say the buyer defines audience and strategy while the platform executes.

The features page then sells an auto reply assistant that answers incoming messages in a natural way, keeping momentum going and guiding prospects closer to a deal, in the vendor's own words without the buyer having to lift a finger. That is unsupervised conversation with a prospect. An approval queue, an escalation threshold and an audit record of what the assistant said are all absent, and the contradiction itself is the finding a buyer needs.

A dedicated disclosure page, linked from the footer, names the provider and the model family outright as Google Gemini, states which inputs flow to it, prompts and outreach content, and describes what those inputs produce. Naming the model at all puts this vendor ahead of most of the surrounding cohort, several of whom market model driven personalisation and name nothing.

Off the top band on five specifics: no model version, no hosting region, no processing terms with the provider, no retention statement for the prompts and outreach content sent across, and no answer to whether those inputs train anything. The claim that data is handled securely carries no evidence on that page, though the privacy policy does document a control set elsewhere.

Three testimonials appear and each is attributed to a role and a company type rather than a person or a company, a sales head at a business software firm, a founder at a marketing agency, a growth lead at a consulting firm. They sit in a section alongside five photographs whose filenames identify them as commercial stock library images.

The quantified claims are three phrases with no basis attached: three times faster follow ups, 40 percent higher reply rates, zero missed conversations, above a line reading trusted by hundreds of professionals. Case Studies appears in the footer resources list as plain text rather than a link. The launch announcement carries one statistic and it describes the market rather than the product.

A full legal surface exists, which separates this vendor from most of the surrounding block: a privacy policy, terms and conditions, an end user licence agreement and a disclosure page, all linked from the footer. The privacy policy states the vendor will not send unsolicited marketing without explicit consent, which governs the vendor's email to its own customer rather than the customer's outreach to a prospect.

Across every marketing page and the privacy policy, no electronic marketing statute is named, no consent position on the outreach itself appears, and no unsubscribe, suppression or recipient removal route is described. The terms and the licence agreement were not read and could carry an acceptable use position, so this row is flagged for re verification rather than graded as an absence.

The privacy policy is substantially better than the size of this vendor would predict and better than several much larger vendors in this index publish. It enumerates the data categories collected, states a purpose for each, publishes a sub processor table naming four third parties with what is shared and why, states adherence to the Google limited use requirements with the specific scopes requested, describes cookie practice down to the attributes used, sets retention rules per data category, commits to deletion within 30 days of a verified request, and enumerates six data subject rights including portability and consent withdrawal.

Three things hold it off the top band. No statute is named anywhere, and the rights are framed as depending on the reader's jurisdiction without naming one. The international transfer clause is a consent grab rather than a residency posture. And every data subject in the document is the customer: the prospects whose profiles are analysed and who receive the messages appear nowhere in it.

There is no data supply chain here to interrogate, which is itself the finding. Prospects are located through the buyer's own authenticated session by a desktop client running searches, and the privacy policy records keyword search history and campaign task data as the artefacts of that. No purchased database, no waterfall of unnamed providers, and no enrichment credits appear anywhere, which is a cleaner provenance story than most of the surrounding cohort.

What is missing is any position on the profile data the system analyses to personalise a message: what is retained about a person who is not a customer, for how long, and on what basis, are all unaddressed.

The architecture is the most conservative in this alphabetical block by a wide margin. Automation runs from a desktop client on the buyer's own machine, on the buyer's own account and address, and the privacy policy confirms the mechanism plainly by disclosing that credentials are stored so the client can act on the user's behalf.

Proxies, address rotation, antidetect browsers, sender multiplication and account rental are all absent, and no marketing copy anywhere celebrates evading detection. Held at the middle band because automating a platform with stored credentials is squarely what that platform's agreement addresses, and the vendor states no conformance position of its own. The one reference to the platform's programmatic interface and terms sits in an agency blog post as advice about choosing a partner, not as a claim about this product.

One sentence in the privacy policy addresses this and it is precisely limited in a way worth reading closely: the personalisation data a buyer supplies is processed in real time and is not sold or shared with third parties for their own marketing purposes. That rules out onward sale for marketing and leaves training untouched, both at the vendor and at the model provider it names.

The disclosure page states that prompts and outreach content may be processed through a third party model and stops there. The same wording pattern has been recorded elsewhere in this index, where a vendor ruled out training public models and left its own unaddressed. A clear statement about training would move this row two bands.

The concealment goal is stated as a selling point in three separate places. Generated intros are described as human like and as sounding authentic rather than like a cold sales pitch, the auto reply assistant is described as responding in a natural, human like way, and the frequently asked questions answer the question of whether messages will feel automated with a flat no. A prospect can therefore hold an entire exchange with software while being told, by design, that it feels human.

Held at the middle band rather than lower on two grounds. There is no identity substitution: the buyer's own account and real name carry the message, unlike the rented profiles graded twice in the preceding builds. And the vendor does publish a disclosure page, though that page faces the buyer rather than the person receiving the message, and no position on the European disclosure obligation appears anywhere.

Three third party systems are named with real detail rather than as logos: a customer relationship platform connected by delegated authorisation and gated to the paid tier, calendar and mail integration with the specific permission scopes disclosed and limited use adherence stated, and a payment processor. A desktop client ships for Windows and macOS with rotatable keys. That is genuine integration documentation.

Depth is the limit: one customer relationship platform and no second, no developer documentation, endpoint reference or marketplace located, no other connector named, and the footer's integrations link resolves to an unrelated path. Four of the six items in the footer resources list are plain text rather than links.

The hosting platform is named, which is more than most vendors of this size disclose, and the sub processor table records that all account and usage data sits there. Region is where it stops. The policy states servers and database infrastructure sit on cloud platforms that may store data in various geographic regions, and asks the buyer to consent to transfer outside their country of residence, which is a consent clause rather than a residency posture.

No region is named, no choice is offered and no transfer mechanism is described. A vendor graded earlier in this index earned a higher band for stating plainly that data is stored in one named country: a residency posture with no options in it, disclosed clearly, beats a clause that names nowhere.

A documented control set exists and is specific: encryption at rest and in transit on a named database platform, the account password hashed with a named algorithm, session cookies configured with the protective attributes named individually, client keys rotatable from the dashboard, and transport encryption throughout. Certification, an audit, a penetration test and a trust centre are all absent.

The finding a buyer should carry away is one the vendor disclosed itself and deserves credit for disclosing. The policy states the account password is hashed and never stored in plain text, and separately lists the buyer's professional network password as stored so the desktop client can act on their behalf, with no equivalent protection claimed. A credential that has to be replayed into another service cannot be hashed and still work, so it is held in recoverable form.

Several silent competitors in this category almost certainly do the same and simply do not say so, which is why this is graded as an ordinary absence of certification rather than punished as a disclosure.

Two tiers carry real numbers, a 30 day trial at no cost with a stated allowance and a paid tier at 29 dollars or 2,400 rupees a month with its feature list and its own allowance, and the buyer can see the usage economics that matter most, 10 generated messages a day on the trial and 25 on the paid plan. Dual currency pricing is published rather than geo switched, which is unusual and useful. Three gaps keep it off the top band.

The unit of pricing is never stated, so a team cannot tell whether the figure is per user or per account and cannot budget beyond one person. The paid tier advertises unlimited automation and unlimited campaigns in the same box as a hard daily cap on the generated messages, which is the feature being bought. And the third tier is priced as custom with the words coming soon beneath it, spelled with a typographical error, while its contact button routes to the registration form rather than to sales.

Portability is an enumerated right in the privacy policy, stated as a request for data in a structured, commonly used, machine readable format, alongside deletion of the account and associated data within 30 days of a verified request, retention rules set out per data category, and third party tokens held only until the integration is disconnected. Lead export and sync to a connected customer relationship platform are separately claimed as product features.

That combination puts this vendor above several far larger ones in this index on this axis, one of which publishes no deletion timeline at all. Off the top band because the portability right is a policy commitment rather than a described function: no self service export, no file format and no scope are named, the fate of campaign history on cancellation as distinct from deletion goes unstated, and the subscription terms were not read.

The discipline language is present and entirely qualitative: scale without sounding like spam, natural conversation flow, smart follow ups. The only published numbers govern the generated messages rather than the sending, 10 a day and 25 a day by tier. What is missing is the whole substance of this axis: no daily action limit, no warmup guidance for a new or cold account, no throttling policy, no account health monitoring and no stated response to a platform warning.

The gap is sharpened by the vendor's own headline claim. Unlimited automation, sold against a platform that throttles a single account hard, is a promise the vendor cannot keep and a risk it never names. Two vendors graded immediately before this one publish explicit caps, one of them contractually at 20 requests a day.

Buyers are named consistently across the launch announcement, the site and the frequently asked questions as founders, agencies, sales teams, recruiters and growing businesses, and the blog reaches for specific shapes such as agencies running client campaigns under their own brand and multi location franchise operations. Coverage detail is where it thins.

Size band, region availability, industry fit and any stated ceiling are all absent, the enterprise tier is marked as not yet existing, and the pricing unit is unstated, so the point at which this product stops fitting a growing team cannot be read from anything published.