LinkedRadar Review 2026: What We Found

LinkedRadar is a Chrome and Edge extension for solo operators running separate invitation, first-degree message, and InMail campaigns. It also includes search imports, a built-in sales CRM (SCRM), AI-assisted copy, and a Prospeo-backed email finder (product scope).

Start costs $19.90 month to month or $13.90 per month equivalent when $166.80 is prepaid yearly (pricing). Our workflow verdict is conditional, while overall safety is not established.

Public evidence did not establish whether raw li_at—the LinkedIn login in cookie form—moves from the browser to the vendor SCRM. Nor did it establish the proposed direct-write, injection, synthetic-event, or extension-detection surfaces. A live cloud/IP test was not run.

Verdict

LinkedRadar is a conditional option for a buyer who accepts browser-bound, linear LinkedIn automation. Its published month-to-month entry price is below $20. Current vendor pages show three LinkedIn campaign types, regular-search and Sales Navigator imports, CSV movement, AI text generation, and per-action pacing controls. They also show a five-account Team allowance (pricing).

The safety evidence limits that recommendation. The Store listing identifies the v1.8.23 package, but no direct public URL was found for the static technical cluster. Public evidence therefore does not establish the reported raw-session path, direct writes, content injection, synthetic events, telemetry behavior, or extension-detection surface. Those gaps are part of the verdict rationale, not proof that any mechanism is absent. The required live cloud/IP test was not run, so overall safety is not established. Because the applicable live cloud/IP test was not run, this draft remains blocked from publication. The static teardown does not replace that required test.

Public materials also did not establish multistep branching, a reply-triggered stop, A/B variants, or a unified inbox as of September 2026. No current $0 plan card or timed trial is published, and the available 45 raw reviews end in 2023. They cannot establish present product behavior or current consensus.

Pros

• Three distinct LinkedIn campaign surfaces: Invitation, first-degree message, and InMail flows are documented separately (tutorial hub). Each action type stays explicit rather than sitting behind a generic automation label.
• Published entry pricing: Start is $19.90 month to month or $13.90 monthly equivalent on annual prepayment (pricing). The commitment trade-off is calculable.
• Search-to-list workflow: Regular LinkedIn search, Sales Navigator search, CSV/XLSX import, lists, tags, and contact export cover common list-building tasks (extension page).
• AI text allowances are quantified: Every paid tier lists a monthly AI Optimization allowance, making generation volume visible before checkout (pricing).
• Pacing controls are visible: Invitation and message settings expose separate Daily Limit and Interval Seconds fields (settings tutorial). The operator controls queue timing directly.
• Five-account commercial option: Team bundles five LinkedIn accounts at one flat plan price (pricing). That gives a small multi-account operator a known subscription floor.

Cons

• Overall safety is not established: The required live cloud/IP test was not run. A static teardown cannot replace that blocking test or establish a live restriction outcome.
• The technical cluster lacks public support: The Store listing identifies the package, but no direct public URL was found for the reported li_at path, direct writes, injection, synthetic events, telemetry behavior, or extension-detection status. Those mechanisms remain not established for publication.
• Browser-runtime dependence needs validation: Chrome and Edge distribution is public, but current human-readable documentation does not establish the complete runtime dependency or whether normal LinkedIn use can continue while a campaign runs.
• Workflow depth is unresolved beyond linear actions: Public materials did not establish multistep branching or a reply-triggered stop. They also did not establish A/B variants or a unified inbox as of September 2026 (message tutorial).
• Email production detail is unresolved: A “Send Now” UI is visible (InMail tutorial). Mailbox authentication, sender identity, allowance, and deliverability were not established.
• Commercial documents conflict: The Terms use categorical no-refund language. The Refund Policy describes a narrow 24-hour exception for a first-payment system error.
• Current market evidence is thin: The 45-review raw sample ends in 2023 and contains only one strict negative record. It provides no current six-month ratings, so it cannot support a 2026 consensus.

LinkedRadar Pricing: Monthly Reality Behind the Annual Default

LinkedRadar’s pricing page defaults to annual-equivalent figures. Start first appears at $13.90 per month, but that requires a $166.80 yearly prepayment. The cancellable month-to-month state is $19.90. The displayed equivalent is about 30% below monthly billing, but it is not a monthly commitment.

The upgrade pattern is quota-led: Biz doubles the published Start invitation and InMail allowances. It increases AI Optimizations fivefold for $10 more month to month. Pro keeps invitations and InMails at 800 each but raises AI Optimizations and campaign count. That means a buyer should select a tier by the limiting quota, not infer more workflow depth from the higher price.

“Unlimited” remains scoped: Team names unlimited invitations, unlimited InMails, and unlimited campaigns as plan allowances. It does not name ordinary first-degree messages, remove hidden daily counters, or establish unlimited safe activity. For neutral platform context, see how LinkedIn invitation limits differ from a vendor’s monthly allowance.

Free access is unresolved, not disproved: the 2026 pricing page has no $0 card and publishes no timed trial. Older reviews describe capped free access, including one report of 30 messages per day. Those historical accounts do not establish the current post-signup state. The first registration step shows no card field; that does not establish whether paid checkout later requires one (registration).

Exit terms require reading two pages: the Refund Policy describes Profile → My Plan → cancel and a narrow 24-hour first-payment/system-bug refund exception. The Terms say completed orders are non-refundable and describe contacting the company to cancel. Post-cancellation data access is not documented, so procurement should preserve that uncertainty rather than infer deletion or continued access.

The pricing page does not disclose setup fees or tax/VAT treatment. It also omits explicit currency text and a custom tier. These are unresolved invoice inputs, not evidence that the charges do not exist. Confirm the checkout total before annual prepayment.

We found one dated current tier snapshot, not two comparable price points for the same plan. A repricing trend is therefore not established.

LinkedRadar Features: Seven Evidence-Supported Axes

LinkedRadar’s visible interface, paid allowance, architecture, and outcome evidence answer different questions. A button establishes a path; it does not prove delivery, restriction avoidance, or conversion.

• Capability means a public UI or code path exists.
• Entitlement means a plan publishes an allowance.
• Architecture tracks where the session and data move.
• Outcome requires live execution.

Sources checked in September 2026: pricing, extension distribution, email finder, and safety claims.

v1.8.23 Extension: Local Scrape and Cloud Session Handoff

The Chrome Web Store listing establishes that LinkedRadar distributes an extension and identifies its package. It does not establish the source-to-sink technical cluster recorded in the static inspection. No direct, human-readable public URL was found for that cluster, so the path descriptions below preserve the research question and both reported destinations without presenting either path as publicly verified product behavior.

Voyager is the internal web interface used by LinkedIn’s own pages. JSESSIONID carries a CSRF token, while li_at is the LinkedIn login in cookie form; whoever holds the session cookie may be able to act without asking for the password again. Those definitions explain why the proposed paths matter. They do not establish that LinkedRadar used either path.

Evidence status checked August 29, 2026

The diagram is therefore an uncertainty map, not a finding that vendor session custody or backend replay occurred. Public evidence also did not establish the proposed permission scope, content injection, synthetic-event generation, telemetry behavior, or extension-detection surface. This overview explains generally why browser-visible extension surfaces matter, but it is not evidence about LinkedRadar.

A live cloud/IP test was not run. That missing blocking test means no cloud egress IP, subnet, IP reputation, fingerprint, live session path, or overall safety outcome was observed. The static teardown does not replace it.

Invitations, Messages, InMail, and Campaign Limits

Like most tools in this category, LinkedRadar documents three common campaign types (tutorial hub):

• connection invitations
• first-degree messages
• InMail

That is the closed action list established for campaign sending in the dated public/package review as of September 2026. Follow, like, endorse, post-commenting, and post-boost automation were not found on that surface. This remains a September 2026 evidence boundary, not proof about private or future versions.

The pricing table publishes monthly invitations and InMails, AI Optimizations, campaign counts, and LinkedIn-account allowances. It does not publish a monthly allowance for ordinary first-degree messages or numeric daily package caps. The monthly plan allowances therefore cannot be converted into a supported daily sending number.

The difference between InMail and a first-degree message matters here because the public plan table names one but omits the other. The documented message-campaign surface did not establish a multistep sequence, branching condition, or reply-triggered stop. It also did not establish a unified inbox or A/B variant. Campaigns therefore read as separate linear action queues in the evidence, not reply-aware sequences.

Search Harvesting, Lists, SCRM, and CSV

The source-to-list workflow has six parts. The inputs are regular LinkedIn People search, Sales Navigator search, and CSV/XLSX import. Prospect lists, tags, and contact CSV/Excel export complete the flow. The regular-search tutorial and Sales Navigator tutorial establish the two search sources. The extension page also names compatibility with LinkedIn Free, Premium, Sales Navigator, and Recruiter Lite.

Vendor pages checked August 18, 2026

This supports list movement, not every data operation. The dated review did not establish:

• import/export row caps
• full message-history export
• post, group, or event collection
• duplicate handling
• company-data enrichment
• native ICP scoring

A buyer can plan a search-or-file-to-campaign workflow. Broader audience sources and conversation portability remain unresolved. This guide gives neutral context on moving Sales Navigator leads into a spreadsheet.

The SCRM is built into LinkedRadar; it is not evidence of external CRM sync. A native-integration count was not established because no named external native connector was found. The vendor’s generic CRM article uses unnamed “System A/B/C” examples. The live /api response establishes an auth-gated backend route, not a documented customer API. Public Zapier and Make apps were not found as of September 2026. Neither were customer-webhook or named CRM routes. The same dated review did not establish late-reply CRM resync or identifier-cascade/fill-empty merge rules. It also did not establish a named Snov.io campaign handoff. Those are evidence gaps around external sync, not proof about an unseen implementation. For conceptual background, see how LinkedIn-to-CRM data transfer differs from an actual connector.

AI Personalization and Template Boundaries

LinkedRadar’s homepage documents vendor-managed ChatGPT/GPT-4 copy, AI Connect, and profile-based icebreaker text. The product evidence therefore supports AI message generation, AI personalization, and AI icebreakers. Like much of the category, this is text generation. It does not prove originality, reply awareness, conversion lift, or campaign-level decision-making.

The quota implication is narrow: moving from 500 to 12,000 monthly AI Optimizations buys more text-generation allowance, not more sequence depth. The dated public sweep did not find a user-supplied model key or provider choice beyond vendor-managed ChatGPT/GPT-4. It also did not establish:

• native AI ICP scoring
• AI post commenting
• campaign-level agentic logic

Those gaps remain evidence boundaries rather than claims about an unseen implementation.

The complete template-variable catalog was also not established. That matters because AI-generated wording and field substitution are separate layers. Generation volume cannot show whether every merged field will render cleanly for every prospect.

For practical context, this guide shows how AI-assisted LinkedIn message personalization differs from workflow logic. A personalized sentence can improve copy relevance. It cannot decide whether a lead replied or stop a later action unless the campaign engine exposes that control.

Email Finder and Direct Email: Third-Party and Sender Gaps

LinkedRadar’s branded email-finder page names Prospeo’s Email Finder as the underlying lookup service. It says the finder returns contact information and uses “Only pay for valid email addresses” language. That is a vendor claim about the finder; no independent accuracy, catch-all, freshness, or deliverability test was run for this review.

The sending path is separate. An official InMail tutorial shows a “Write An Email by AI” modal. It includes recipient, subject, body, and “Send Now.” That establishes a visible one-recipient send path. It does not establish the sending domain or whether a user mailbox is authenticated through OAuth or SMTP. The allowance and inbox delivery also remain unresolved.

Vendor pages checked August 18, 2026

The dated public review did not establish:

• a connected-mailbox screen
• a named OAuth provider
• SMTP setup
• a published email allowance
• a native email sequence
• a LinkedIn-plus-email campaign

No company-data enrichment or freshness-date field was found beyond the Prospeo-backed email/phone lookup. That remains a bounded August finding.

The main pricing grid contains no email-credit or mailbox line, while the finder page uses pay-for-valid language without a public pack price. The plan allowance, finder total cost, rollover rule, and sender cost therefore cannot be derived from the current public pages. Test those inputs before treating the visible composer as a production email channel.

That uncertainty matters because lookup, one-off sending, and a reply-aware multichannel sequence are three different capabilities. This overview of combining LinkedIn and cold-email workflows provides neutral context for the distinction.

Safety Controls Versus Detection Surfaces

LinkedRadar documents separate Daily Limit and Interval Seconds fields for invitation and message activity in its settings tutorial. The vendor’s safety page says the queue adjusts to the user’s sending limit. In practice, those controls can reduce bursts and stop the queue at the configured threshold.

The homepage promises random delays and gradual increases. The documented settings screen shows one Interval Seconds value rather than a min-max range. Current public evidence therefore establishes user-configured pacing, not independently verified randomization. Its “up to 100 invites and 100 messages in one day” language is marketing guidance. It is not a supported safe volume or a substitute for LinkedIn-side limits.

These controls also do not establish package detectability. The Store listing identifies the extension, but no direct public URL was found for the reported probeable-resource, injection, synthetic-event, direct-write, or telemetry cluster. Those mechanisms remain not established for publication. An evidence gap is not proof that the surfaces are absent; it means pacing claims cannot support an overall safety conclusion.

The dated public review did not find:

• working hours
• a proxy layer
• a proxy-quality checker
• randomized min-max delays
• automatic risk-triggered pausing

That is not proof these controls can never exist. Do not include them in a buying decision without current product evidence. Team’s unlimited invitation and InMail allowances also remain commercial entitlements, not safety controls.

Team Accounts, Roles, and Agency Ceiling

The Team plan bundles five LinkedIn accounts. It costs $99.90 month to month or $69.90 per month equivalent with $838.80 prepaid yearly. This is a flat plan allowance, not a published per-account price, and the page does not define five human seats.

The implication is a ceiling, not an allegation. Five account slots do not establish five isolated client tenants or five independent human seats. Public materials did not document:

• roles or permissions
• client workspaces or white labeling
• sender rotation or lead distribution
• cross-account duplicate protection
• an official path above five accounts

A small operator can evaluate the entitlement. An agency should validate tenancy and governance rather than infer them from the plan name.

There is no custom, Enterprise, or Agency card on the dated pricing grid, and no per-account add-on was documented. Creating multiple separate subscriptions is therefore not an official bulk offer and should not be used as a vendor price.

The public contact page lists support email and a VIP WhatsApp route. No tier-specific SLA or dedicated success manager is documented. Paying for Team therefore establishes account capacity, not a higher support outcome.

Is LinkedRadar Safe?

Short answer: overall safety is not established. The vendor’s safety page says data is stored “in your local, not cloud-base” and claims LinkedIn cannot determine that the extension is installed. Those are vendor claims, not verification of session custody, detectability, or live outcomes.

What I verified first-hand. I inspected a v1.8.23 MV3 package record in September 2026. The Chrome Web Store page publicly identifies the extension and ID. No direct human-readable public URL was found for the accompanying static technical cluster, so I do not present the reported source-to-sink paths or detection mechanisms as publicly established findings.

Active Extension Detection (AED) is a label used for LinkedIn extension-scanning events, not an official LinkedIn feature name. In general, a known extension resource can make installation visible before an automation action runs. LinkedRadar’s AED-list status was not established by a direct public URL, so it cannot support this verdict. This overview gives general context on browser-extension detection, not evidence about LinkedRadar.

Honest read: A live cloud/IP test was not run. The required live cloud/IP test was not run, so overall safety is not established. The static teardown does not replace this required test. It also cannot support a claim that vendor session custody, backend replay, direct writes, injection, synthetic events, telemetry behavior, or an AED-visible detection surface occurred when the direct public source for that cluster was not found. That uncertainty is not proof that any mechanism is absent.

The Privacy Policy permits international transfers to the United States, Russia, Japan, China, and India. It does not identify exact storage residency or a public subprocessor list. A general GDPR section is present; a separate public DPA was not established. Public contact and Terms materials use conflicting company and location references, so procurement should verify the contracting entity rather than infer one.

Manual behavior also matters: opening many profiles directly by URL or processing actions at unrealistic speed can trigger warnings without automation. LinkedRadar’s pacing controls may reduce bursts, but vendor allowances remain separate from LinkedIn-side limits. This restriction overview explains the broader behavioral layer.

What LinkedRadar Review Research Actually Shows

The available review evidence is historical. Our review sample contains 45 records through 2023. Within that sample, G2 contributes 30 reviews averaging 4.63/5 through 2023. Capterra contributes 13 reviews averaging 4.85/5 in the same sample and time window. Trustpilot contributes 2 reviews through 2023—too few to print a mean. These are not live August 2026 platform totals.

Cross-platform author deduplication reduces the same review sample to 41 voices: G2 30, Capterra 9, and Trustpilot 2. That denominator is separate from the 45 raw records, so no percentage derived from one should be read as using the other. The trailing six-month window ending August 3, 2026 contains 0 rated reviews and is too thin to calculate a trend.

We found one weekly volume spike. In the week beginning June 5, 2023, 5 of the 45 raw reviews appeared, or 11.1% of that raw sample. We found no cluster of negative reviews. This is a date pattern, not evidence of why the reviews appeared. It is not proof of solicitation, manipulation, purchase, fabrication, or fraud.

We found no LinkedRadar mention in the Reddit posts we reviewed. The 45 reviews also contained no switched-from or switched-to story. These are bounded sample findings, not proof of market-wide absence. These reviews can illustrate historical perceptions. They cannot validate v1.8.23, current pricing, present cancellation behavior, or current safety.

The Trustpilot card is one dated reviewer allegation, not a finding of vendor conduct or a recurring billing outcome. The other cautionary cards come from mixed or positive ratings. They show specific friction—screening, browser dependence, and missing follow-up depth—rather than broad negative consensus.

Who LinkedRadar Is For — and Who Should Look Elsewhere

LinkedRadar’s documented scope covers separate, linear invitation, first-degree message, and InMail campaigns. A solo operator or very small team can compare that scope with its requirements. The complete browser-runtime and session-custody paths are not publicly established, and the required live cloud/IP test was not run. Buyers should also validate quotas, free access, and cancellation before committing annually.

Security-sensitive procurement still faces unresolved session, detection-surface, residency, and live-egress questions. Public evidence also did not establish reply-triggered stopping, branched multistep sequences, email deliverability, role-based access, or isolated client workspaces. Buyers requiring those capabilities should look elsewhere or validate them directly. These are evidence, workflow, and governance boundaries, not generic company-size labels.

A low entry price does not resolve those boundaries. It only lowers the cost of testing them. A Start or Biz buyer should confirm current free access, ordinary-message limits, reply behavior, and what happens after cancellation. A Team buyer should add seat, role, client-isolation, and sixth-account questions.

For a prospective user, the decision turns on three questions:

• Can the vendor document the complete browser-runtime path?
• Can the vendor supply public evidence for session custody and detection-surface claims, plus a completed live cloud/IP test?
• Are separate linear campaigns enough?

If any answer is no, the price advantage does not repair the workflow mismatch. If all three are acceptable, begin with a reversible monthly test. Do not treat older reviews or annual-equivalent copy as current outcome evidence.

How LinkedRadar Compares to Alternatives

LinkedRadar’s public position is a $19.90 month-to-month extension with separate linear campaign surfaces. Its complete runtime, session-custody, and detection-surface mechanics are not publicly established, and the required live cloud/IP test was not run. The table below gives a narrow snapshot of the five named alternatives supplied for this review. Architecture and price are dated fields, not safety or outcome rankings. The “best known for” field states a bounded capability with its limitation.

Two slug-specific pages were confirmed for deeper reading: Dripify comparison → and Dux-Soup comparison →.

See all LinkedIn automation alternatives →

This table does not establish that one architecture is universally safer. It shows differences in session location, extension presence, and native email sequencing. It also shows inbox evidence and whether current public evidence supports each cell.

Where Linked Helper Fits

If LinkedRadar’s missing live cloud/IP test, not-publicly-established session and detection mechanics, and missing reply-triggered stop matter to your evaluation, a documented user-controlled execution model is worth examining.

Linked Helper is a standalone desktop app with its own browser engine. It runs on the user’s PC or VPS, with Web Version providing browser access to that user-controlled instance.

• Different browser boundary: It does not inject a Chrome extension into LinkedIn or send the LinkedIn session token to a vendor cloud.
• Reply-aware linear campaigns: Reply checks can stop later messaging. Templates support IF/THEN/ELSE logic and spintax within messages, not branching campaign workflows.
• Named integration routes: It documents 11 direct CRM connectors, plus Zapier and Make through webhooks.
• Visible network check: Its proxy checker shows IP reputation before use, and a user can assign a dedicated proxy per account.

LinkedRadar starts at $19.90 month to month or $13.90 per month equivalent with annual prepayment. Linked Helper starts at $15 for a one-month Standard license or $8.25 per month equivalent on a 12-month license. Always-on operation still requires the user’s PC or a separately obtained VPS.

Teams prioritizing user-controlled session custody, reply checks, and named CRM routes may want to evaluate Linked Helper too. The supplied sample contains no migration story between the products, so this is an architecture and workflow comparison, not a switcher claim.

Methodology and Disclosure

See the publisher note above for authorship context. Our September 2026 research cycle covered live public pages, pricing, legal terms, tutorials, the Chrome Web Store, status pages, and dated screenshots. We reconciled those sources by recency, directness, and whether they documented a claim or merely exposed a visible interface.

First-hand work was limited to a September 2026 static inspection of LinkedRadar's publicly viewable shipped MV3 package, version 1.8.23 (public package source). This establishes package contents and inspectable code paths, not live account behavior or backend replay. A live two-account cloud/IP test was required but was not run. The static teardown does not replace that required test. We make no claims about vendor-cloud execution, cloud egress IPs, shared infrastructure, or restriction rates. No authenticated campaign, billing change, support exchange, or delivery test formed part of this review.

We used two search queries and 26 source captures to calibrate coverage and structure. We treated reviews as historical context because the 45 available records deduplicated to 41 and ended in 2023. We excluded sources that did not meet our evidence standards and separated current documentation from older review evidence. This evidence hierarchy supports bounded observations, not daily-use experience or campaign outcomes. Where a public capability was not established, the article preserves that uncertainty rather than treating an evidence gap as absence.