Scrupp Review 2026: What We Found

Scrupp is a data-to-email platform for teams handling LinkedIn, Sales Navigator, or Apollo records. It extracts and enriches them before running email outreach. One paid plan starts at $29 monthly for 1,000 credits, with larger credit volumes available. Its included-feature list covers email outreach and REST API access. Scrupp does not document LinkedIn invitations or message sending. Our verdict is conditional. However, a static inspection of extension v1.3.6 confirmed LinkedIn session upload. We did not check the live request origin, retention, replay, revocation, or restriction rate. The architecture therefore receives a high-risk assessment rather than a safety assurance.

Pricing refreshed September 2026; other checks August 2026 · Sources: Scrupp pricing, product scope, and the official extension package.

Verdict

Scrupp is a candidate for a controlled evaluation. The relevant workflow starts with LinkedIn, Sales Navigator, or Apollo data and ends in email. Its current product surfaces establish extraction, enrichment, exports, background jobs, email campaigns, REST API access, webhooks, and agency workspaces. Together, these surfaces form a data-to-email chain rather than a scraper alone.

Four gates block an unconditional recommendation. First, pricing uses a single volume-based plan, but still conflicts on complete-lead credit cost and email inclusion. Headline allowances therefore do not produce a dependable cost per usable lead. Second, Email campaign mechanics such as reply auto-stop, branching, A/B tests, and numeric sending limits were not established. Third, connector claims do not all resolve to operational apps. Fourth, static extension evidence confirms session upload, while cloud routing and session lifecycle remain unmeasured.

As of August 2026 · Sources: features, pricing, API documentation, and the official extension package.

The decision supports a bounded data-and-email pilot. It does not support scaling until those four gates are tested in the intended account and billing setup.

Pros

• Broad source coverage: Scrupp documents people, companies, jobs, posts, reactions, connections, Sales Navigator, Apollo, and Recruiter inputs, subject to source ceilings.
• Background export continuity: According to pricing and Help, started jobs can continue after the browser closes. This reduces open-tab dependence without proving a scheduler or uptime level.
• Built-in enrichment: Like most data tools in this category, Scrupp combines an email waterfall, SMTP verification, phone lookup, and CSV/XLSX delivery. The feature page does not establish current accuracy.
• Email features listed in the paid plan: Email outreach, AI sequences, and a unified reply inbox appear in the current included-feature list. The pricing FAQ still describes sending as separate, so confirm account access and limits (pricing).
• Documented developer routes: The paid feature list includes REST API access; the documented API is asynchronous, while webhooks publish events, signing, retries, and replay limits (API; webhooks).
• Client workspaces: Agency surfaces document isolated client workspaces, branding, intake, reports, and an Activity Log. Global cross-client suppression was not established (white label).

Cons

• Session leaves the browser: The official v1.3.6 package uploads LinkedIn cookies, user agent, and identity to Scrupp endpoints. This changes who can hold the authenticated session.
• Credit yield conflicts: Pricing implies four credits for scraping, a verified email, and a phone. The same pricing page also uses a one-credit-per-lead headline.
• Documentation is not synchronized: The main pricing selector presents one paid plan, while Help retains named tiers and the pricing FAQ conflicts with the email feature list. Confirm unclear entitlements in the account (pricing; Help).
• Connector readiness is uneven: HubSpot has a documented route, but Salesforce and Pipedrive pages conflict with Help. Zapier is labeled coming soon rather than established (webhooks).
• Safety controls lack numbers: Scrupp names throttling, limits, batching, and delays. Pricing publishes no daily budget, delay range, working hours, or adaptive-stop rule.
• Failure recovery remains untested: A small review sample contains support and reliability complaints. Those reports identify test scenarios, not current universal behavior.

As of August 2026 · Sources: pricing, Help, webhooks, and the official extension package.

Scrupp Pricing: One Plan, Credit Volumes, and Extra Costs

Scrupp now presents one paid plan with a selectable credit volume rather than a ladder of feature-gated named tiers. The entry option is $29 per month for 1,000 credits. Monthly volumes range from 1,000 to 100,000 credits; the entry option lists up to three team members and 10,000 leads per Finder search. See current pricing for the selected volume.

Pricing refreshed September 2026 · Source: Scrupp pricing. Email outreach and REST API are listed in the common paid feature set; the page does not publish the old tier-specific inbox, contact, or campaign caps.

Yearly billing is advertised as saving up to 20%, with credits delivered monthly. Check the selected volume's annual checkout total before committing; do not reuse annual totals from earlier named-tier tables. A remaining footnote mentions Growth and above for managed domains, inboxes, warmup, ads intelligence, and lead-capture funnels, without mapping them to the new volume selector. Confirm those extras separately; priority support is listed from 10,000 credits per month.

Credit consumption still needs clarification. The pricing page's detailed formula charges one credit per exported lead, one additional credit for a verified email, and two additional credits for a phone. Yet its headline says one credit equals one lead and unsuccessful contact lookups are not charged. Treat the detailed formula as budgeting scenarios until a sample transaction confirms the actual deductions.

Pricing refreshed September 2026 · Source: Scrupp pricing: headline, credit formula, and FAQ. These are arithmetic scenarios, not measured delivery yields.

One-time credit purchases work separately from the subscription:

• The entry pack costs $35 for 1,000 credits;
• The selector also offers 2,000, 5,000, 10,000, 20,000, 50,000, and 100,000 credits;
• The selected pack's current price must be checked before purchase;
• Payment is one-off, with no recurring subscription charge; and
• Purchased credits do not expire and can stack with subscription credits.

The published $35 entry pack equals $0.035 per credit, not necessarily per enriched lead. For the $29 monthly entry option, the two- and four-credit scenarios above imply $0.058 and $0.116 per lead respectively, assuming all 1,000 credits are used successfully. Larger-pack prices are not inferred from earlier pricing snapshots.

Current pricing says unused subscription credits roll over while the subscription remains active; one-time credits do not expire. Help retains older rollover restrictions, so confirm what happens to subscription credits after cancellation. Written terms allow period-end cancellation and generally exclude refunds for unused credits or partial periods. That policy does not establish what every account UI currently shows.

White-label and wholesale arrangements require a separate partner quote. The current partner page bases pricing on volume and the chosen delivery model; neither the standard credit selector nor older Agency app pricing establishes a fixed 20-account package.

Hybrid Architecture and LinkedIn Session Custody

Scrupp uses a hybrid architecture rather than operating only as “an extension” or “a cloud tool.” The extension performs authenticated LinkedIn reads locally, then sends session material to Scrupp. Started export jobs, enrichment, storage, API tasks, and webhooks can continue in cloud infrastructure. Browser-side collection therefore does not keep every part of the authenticated workflow inside the browser.

What we verified first-hand: We inspected the official extension v1.3.6 package. It reads li_at, li_a, and JSESSIONID, along with the browser user agent and account identity. li_at is the LinkedIn login in cookie form, while JSESSIONID carries the anti-forgery token. Whoever holds that bundle may be able to act as the authenticated account without the password. Uploading it therefore transfers session custody to another environment.

The shipped source also performs local Voyager and Sales API reads. Voyager is LinkedIn's internal web API used by the browser interface. In practice, “local reads” and “session upload” coexist. Neither description alone captures the architecture.

As of August 2026 · Sources: the official extension package, Chrome Web Store listing, and Scrupp Help.

First-party descriptions do not resolve the mode split. Pricing describes extension-plus-server-side scraping. Help calls the extension optional, while feature copy describes browser or user-IP execution. Those statements may refer to different workflows, but mode-by-mode behavior was not dynamically traced.

LinkedIn, Sales Navigator, and Apollo Extraction

Scrupp's documented collection surface extends beyond a single Sales Navigator scraper. The reviewed material documents these sources:

• LinkedIn people, companies, jobs, posts, commenters, reactions, and connections;
• Sales Navigator lead searches, account searches, and saved lists;
• Apollo searches and lists; and
• Recruiter or Recruiter Lite; current purchase entitlement needs confirmation.

The LinkedIn scraper documentation supports the LinkedIn surface. Scrupp can also export person and company records with roles, titles, firmographic fields, and contact data. A project can move from a source search to a file or enrichment job without changing products. This workflow does not measure field accuracy or freshness.

As of August 2026 · Sources: LinkedIn extraction, product features, and Scrupp Help.

A 2,500-result source ceiling and a 10,000-row paid export ceiling describe different layers. Neither proves unlimited daily extraction, especially because Scrupp also publishes both “no daily limit” and references to daily limits and batching. This Sales Navigator export guide explains the underlying list-export workflow without resolving Scrupp's conflicting daily language.

Scrupp's confirmed scope ends at extraction for LinkedIn actions. The reviewed product material documents no built-in connection requests, LinkedIn messages, likes, comments, or endorsements. Scrupp's listed outreach channel is email, so export capacity must not be treated as LinkedIn engagement capacity.

Enrichment, Verification, Credits, and Dedupe

Scrupp combines discovery, checking, and export in one data flow. A vendor-verified address is not a measured inbox-delivery result.

Data enrichment means appending contact fields, usually a work email or phone, to an existing profile. Scrupp uses a multi-provider email waterfall, then applies real-time SMTP checks before labeling the result. Scrupp's feature page also documents phone lookup and CSV/XLSX delivery.

As of August 2026 · Sources: features, pricing, and the standalone verifier.

The standalone email verifier provides ten no-signup checks per day. The current pricing FAQ includes re-checking an existing email for free, while finding a new verified email adds one credit to the export charge. The standalone utility's daily limit is not an in-account bulk allowance. Likewise, the word “verified” describes Scrupp's checking process. It does not guarantee inbox placement, current employment, or a zero-bounce outcome.

This email-finder overview gives category context for finding and checking work emails. Scrupp's own hit rate, accuracy, and deliverability still require a controlled sample because review praise is not a benchmark.

Dedupe has a similar boundary. Match Contacts and cleaning can reduce duplicates at list or database level. No global suppression control across isolated agency clients was confirmed. The evidence therefore does not establish that two client workspaces cannot contact the same person.

Scrupp also does not document full LinkedIn messaging-history export. Its confirmed exports are lead and company data.

Email Outreach, Reply Inbox, and AI

Like most data-to-email platforms, Scrupp offers sequences, sending inboxes, a reply inbox, and AI-assisted copy. The current paid feature list includes email outreach, AI sequences, and a unified reply inbox, but does not publish inbox, active-contact, or campaign caps by credit volume. Earlier app material shows editable sequence timing and copy; it is not proof of current plan limits. The visible AI surfaces are AI Campaign Builder, drafted replies, and AI Search. This evidence does not establish LinkedIn engagement actions.

Pricing refreshed September 2026 · Source: Scrupp pricing. Its common paid feature list includes email outreach, but its FAQ still describes sending as a separate product. The old named-tier limits are not treated as current entitlements.

Sequence presence is not workflow depth. The reviewed public material did not establish maximum steps or follow-ups. It also did not establish conditional branching, automatic stop on reply, A/B testing, or a numeric daily sending range. These are evidence gaps rather than proof the controls are absent. They matter because a reply-aware campaign and a timed series can behave differently under live traffic.

AI Search turns a natural-language Ideal Customer Profile description into target criteria or a list. It is not established as automatic profile-by-profile ICP scoring, and its model, provider, standard-plan charges, and output accuracy were not documented. AI-drafted replies and campaign copy are visible surfaces, but moderation, edit gates, and autonomous quality were not tested.

The email and LinkedIn channel guide helps separate email follow-up from LinkedIn actions. Here that distinction is decisive: Scrupp lists email campaigns in its paid feature set, while the reviewed Scrupp scope does not execute connection requests or LinkedIn messages.

API, Webhooks, CRM Integrations, and Connector Drift

Scrupp's developer surface is more specific than its connector catalog. The REST API is included in the paid feature list and is asynchronous: a request returns a process_id, and the result is retrieved later. That model supports long-running enrichment and extraction work, but published routes do not establish latency, idempotency, or partial-failure recovery under production load.

Webhooks are documented on all plans. The webhook page names:

• completion and enrichment events;
• HMAC-SHA256 signing;
• three retries after server errors;
• replay;
• ten webhooks per account; and
• 100 requests per second per endpoint.

These numbers make the interface auditable. They do not replace a live replay and duplicate-delivery test.

As of August 2026 · Sources: HubSpot, Salesforce, Pipedrive, and webhooks.

That leaves one currently confirmed direct CRM route: HubSpot. Salesforce and Pipedrive remain advertised but conflicted. Zapier is not operationally established, and Make is a generic webhook or API path. This CRM integration overview provides category context for checking data direction and field ownership rather than counting every landing page equally.

Teams, Agencies, and White-Label Workspaces

Standard team accounts document three roles: Admin, Member, and Viewer. Teams can share accounts, exports, folders, data, and credits through Scrupp Help. Current pricing lists up to three team members at the 1,000-credit monthly option. Caps for other volumes and every permission boundary still need confirmation.

Agency surfaces add a different layer. Scrupp's white-label page describes isolated client workspaces, separate campaigns, senders, lists, and inboxes. It also documents branded portals, intake, reports, and an Activity Log. These client-operation controls are separate from a standard team folder.

As of August 2026 · Sources: Scrupp white label, pricing, and Help.

Agency economics should come from a current written quote covering the chosen model, credit volume, and client-seat allocation. The partner page offers either a rebranded app/dashboard or API-based delivery, with wholesale pricing agreed on a partnership call. An older Agency app amount is not a current quote, and dividing a credit-package price by 20 would invent an account entitlement.

Client isolation creates a governance question. Match Contacts is documented at list or database level, while no global suppression layer across all client workspaces was confirmed. That does not prove collisions occur. It means an agency should test whether one client's do-not-contact record suppresses the same lead in another workspace before onboarding several clients.

Is Scrupp Safe?

Scrupp's safety verdict is conditional and architecture-based. The extension uploads session material that could support authenticated replay, while the live LinkedIn-facing route was not tested. That combination supports a high-risk architecture assessment. It does not establish that replay occurred, identify a cloud IP, or measure a restriction rate.

What I verified first-hand: I read the shipped source in the official v1.3.6 package. It reads li_at, li_a, and JSESSIONID, then posts the bundle with user agent and identity to Scrupp account endpoints. It also performs local Voyager/Sales reads and has LinkedIn page access.

Active Extension Detection, or AED, is the label visible in LinkedIn's production code for probes of specific extension IDs. If an ID appears in that list, installation can be observable before an action runs. Scrupp's extension ID was not listed in the August 2026 audit snapshot. That negative result is one dated signal—not proof the extension is invisible or safe. This extension-detection explainer provides category context for why code injection and extension identity remain separate signals.

As of August 2026 · Sources: the official extension package and Chrome Web Store listing.

Scrupp claims “smart throttling,” rate limits, batching, and delays on pricing, but it publishes no numeric action budget or delay range. The same public material mentions both “no daily limit” and daily limits for high-volume work. Source ceilings of 1,000 or 2,500 are extraction bounds, not safety budgets. The controls may exist, but their effectiveness and enforcement were not tested.

On data security, Scrupp's DPA names Hetzner Germany, AWS, TLS 1.2+, storage encryption, role controls, MFA, VPN, and audit controls. These are vendor contractual statements, not an independent certification. Public evidence for a SOC 2 or ISO certificate was not found in the reviewed material. Component-level residency remains incomplete, as do LinkedIn-cookie retention, deletion, and key management. The Terms also put third-party-platform use at the customer's risk (Terms).

No substantive LinkedIn restriction pattern appeared in the 55-experience deduplicated review sample. This is a limited negative finding, not a runtime safety test.

Manual activity patterns can also trigger LinkedIn warnings, so architecture is only one part of the risk model. This restriction overview explains the platform-level distinction.

No tool eliminates ban risk.

What Reviews and Social Mentions Actually Show

The review evidence is positive overall but too clustered to represent continuous market consensus. We analyzed 56 dated reviews: 10 from G2, 46 from Trustpilot, and none from Capterra. Removing one G2/Trustpilot cross-post left 55 independent experiences: 9 G2 and 46 Trustpilot. Publication dates span the week beginning October 30, 2023 through February 28, 2026.

Each aggregate needs its own denominator. In the G2 local sample as of August 3, 2026, 10 rated reviews average 4.3 stars. In the Trustpilot local sample as of August 3, 2026, 46 rated reviews average 4.22 stars. Across the 55-experience, cross-platform deduplicated G2-and-Trustpilot view spanning the week beginning October 30, 2023 through February 28, 2026, the mean is 4.309 and the median is 5. In that same 55-experience G2-and-Trustpilot view and publication window, 40 of 55 are five-star records and 7 of 55 are one-star records. These are local-sample statistics, not live profile ratings. Only one rated review falls in the February 3-August 3, 2026 trailing window, so no trailing mean is printed.

Temporal concentration matters. In the 55-experience deduplicated view, 42 records fall in March-April 2024 or September-December 2025. The separate 56-review view found two weekly volume spikes. They were 10 of 56 in the week beginning April 1, 2024 and 7 of 56 in the week beginning December 8, 2025. It found no negative cluster. These are patterns, not explanations. They do not establish solicitation or manipulation.

Topic-level tails are smaller. Support appears in 10 of 55 experiences: 7 negative, 2 positive, and 1 mixed. Reliability appears in 9 of 55: 7 negative and 2 positive. Pricing transparency appears in 3 of 55, all negative. Positive topic incidences were:

• data quality in 29 of 55;
• ease in 24 of 55;
• features in 19 of 55; and
• performance in 11 of 55.

Those topic incidences do not prove current product behavior.

The billing reports remain individual allegations. One separate edited dispute finished at four stars after all credits were restored. Presenting only its initial complaint would distort the record. The single visible Reddit post grouped Scrupp with five providers and several possible issues; none can be assigned specifically to Scrupp. No migration narrative was found in the supplied review evidence.

Who Scrupp Is For — and Who Should Look Elsewhere

Scrupp is reasonable to evaluate when the workflow is data-first and email-second. Fit depends on the exact source, channel, session policy, and governance requirement rather than company size alone.

Consider Scrupp when:

• A solo SDR, founder, recruiter, or small team needs LinkedIn, Sales Navigator, Apollo, or Recruiter data prepared for a file, CRM, API, or email campaign.
• The entry subscription's 1,000-credit allowance can feed an external sender or CRM, and the credit formula will be verified with a pilot.
• The paid plan's listed email features match the intended workflow, and the team will confirm account access, sending-inbox, active-contact, and campaign caps before testing reply-stop and send controls.
• An agency needs isolated client workspaces, white-label portals, and shared credits, and will obtain written account and suppression terms.

Look elsewhere when:

• The required workflow sends LinkedIn connection requests, messages, likes, comments, or endorsements rather than collecting data.
• Procurement requires the LinkedIn session to remain entirely on user-controlled hardware, or requires a measured cloud route before approval.
• A mature native Zapier, Salesforce, or Pipedrive connector must work without an installation pilot.
• The budget requires a fixed 20-account quote, stable complete-lead yield, or proven global do-not-contact suppression across clients.

As of August 2026 · Sources: features, pricing, white label, and the official extension package.

How Scrupp Compares to Alternatives

These tools occupy different parts of the prospecting chain, so their entry prices do not buy equivalent units. Scrupp starts at $29 per month for 1,000 credits, with email outreach and REST API listed in the same paid feature set. The table preserves each alternative's supplied category and architecture boundary rather than treating every product as a substitute.

Other-tool snapshot captured April 22-May 27, 2026. The supplied alternatives evidence included no direct vendor URLs, so current checkout prices and limits should be verified before purchase.

Salesforge is email-centered, PhantomBuster spans multiple automation surfaces, and Evaboot is data-enrichment-centered. Each product carries an unresolved boundary in the supplied snapshot. Scrupp combines extraction and email features in one paid plan, but its campaign controls and session route remain qualified. Their inclusion reflects mention frequency, not product quality.

See all LinkedIn automation alternatives →

Where Linked Helper Fits

If Scrupp's confirmed session upload or lack of documented LinkedIn engagement actions conflicts with the operating model, a different execution layer is worth evaluating.

Linked Helper is a desktop application for Windows, macOS, and Ubuntu with its own built-in browser. It can also run on a user-controlled VPS. Web Version provides login-based browser access to that VPS instance, so campaigns can operate around the clock while the LinkedIn session stays in the environment the user controls.

• LinkedIn action workflow: Linked Helper runs connection invitations, messages, profile collection, and ordered campaign steps. This addresses Scrupp's confirmed extraction-without-LinkedIn-engagement boundary.
• Session custody: LinkedIn actions run on the user's machine or VPS, using its IP or a dedicated per-account proxy. The session token is not stored on Linked Helper's vendor servers, which differs from Scrupp's confirmed upload path.
• Message-level conditions: Templates support nested if/then/else logic and spintax. Campaign workflows remain linear rather than branching, so this is not a claim of campaign-level if/then/else.
• Account controls: Rolling limits, randomized delays, working hours, mouse and typing emulation, isolated cookies, and a built-in proxy checker expose pacing and network choices to the operator. No automation setup removes restriction risk.
• CRM handoff: Linked Helper has 11 direct CRM connectors, plus Zapier and Make through webhooks. That supplies named connector routes where Scrupp's Salesforce, Pipedrive, and Zapier status remains unresolved.

As of August 2026 · Linked Helper facts follow the publisher's current product and pricing canon.

The paid entry comparison is $29 monthly for Scrupp's entry 1,000-credit subscription versus $15 for a one-month Linked Helper Standard license. The products meter different work: Scrupp centers data and email, while Linked Helper centers LinkedIn actions. Linked Helper is LinkedIn-first and has no native non-LinkedIn email sequence builder, so teams that need both channels may still pair it with a dedicated email platform. Round-the-clock operation can also add third-party VPS and proxy costs beyond the license.

Teams prioritizing LinkedIn invitations, messages, user-controlled session location, and explicit account controls may want to evaluate Linked Helper too.

These answers summarize the pricing, product-scope, architecture, integration, and safety findings established above. Scrupp's public pages sometimes conflict, and some live execution checks were not performed. Each answer preserves that uncertainty rather than resolving it by assumption.