PhantomBuster LinkedIn: Full Guide to Scraping & Automation

What PhantomBuster Actually Is

PhantomBuster is a cloud-based automation platform built around pre-built workflow scripts called "Phantoms." Each Phantom handles one specific task - scraping LinkedIn profiles, exporting Sales Navigator searches, auto-connecting with prospects, pulling company data. You chain them together to build a prospecting workflow, no code required.

The LinkedIn use case is the main draw for most sales teams. You can export LinkedIn Sales Navigator search results, scrape company pages, auto-connect with prospects, and pull contact data from profiles without writing a single line of code. That's genuinely useful - and it's why the tool has a real following among outbound teams.

But there's more to know before you put your credit card in. Let me break down exactly how the LinkedIn-specific Phantoms work, how to get the thing set up correctly, what the pricing reality looks like, what the safety limits actually are, and how to build a full prospecting stack around it.

How to Connect LinkedIn to PhantomBuster (Step-by-Step Setup)

Before you run a single Phantom, you need to connect your LinkedIn account. This is the part most tutorials skip over, and it's also where most people make their first mistake. Here's exactly how it works.

PhantomBuster needs access to your active LinkedIn session to run automations on your behalf. It does not ask for your password. Instead, it uses a session cookie - a small token your browser stores when you're logged in that tells LinkedIn "this person is already authenticated." You're essentially giving PhantomBuster a temporary visitor pass to act as your logged-in account.

The fastest way to set this up is through the PhantomBuster Chrome or Firefox browser extension. Install it from the Chrome Web Store, log into LinkedIn in the same browser, then open the Phantom you want to run. When you see the "Connect to LinkedIn" field, click the button - the extension detects your active session automatically and populates the cookie field without you having to copy or paste anything manually.

If you'd rather not use the extension, there's a manual method: open Chrome's Developer Tools (right-click anywhere, select Inspect), navigate to the Application tab, find the Cookies section, locate your LinkedIn session cookie (called li_at), copy its value, and paste it into the PhantomBuster setup field. PhantomBuster also needs your user agent string for LinkedIn specifically - you can find yours by searching "what is my user agent" in Google and copying the result. The extension handles this automatically; the manual method means grabbing it yourself.

One important note: LinkedIn also requires your user agent string - a short piece of text that identifies which browser and device you're using. This helps PhantomBuster match your real browsing environment so LinkedIn doesn't flag a session mismatch. For other platforms like Instagram or Twitter, you don't need the user agent. LinkedIn is the exception.

Once connected, your session stays active until it expires or you manually log out of LinkedIn in your browser. Most sessions last days to weeks, but LinkedIn can force re-authentication at any time, especially if it detects unusual activity. If your session drops mid-run, you'll need to reconnect and relaunch. That's the operational overhead this tool carries - and it's worth planning for before you build workflows that depend on consistent uptime.

For agencies running Phantoms on behalf of clients, you can connect multiple LinkedIn accounts to a single PhantomBuster workspace. You can do this by using the extension in a separate browser window for each account, or by having clients send their session cookie values securely so you can paste them in manually. Just make sure whoever owns the account does not log out of LinkedIn before closing their browser after sharing the cookie - logging out immediately invalidates it.

The Core LinkedIn Phantoms Worth Knowing

PhantomBuster has over 100 Phantoms total, but for LinkedIn prospecting, you're really going to use a handful of them repeatedly. Here's what matters:

LinkedIn Profile Scraper

This is the workhorse. Feed it a list of LinkedIn profile URLs and it extracts names, headlines, roles, companies, locations, and recent activity. The Phantom pulls up to 44 data points per profile, including company industry, recent job experience, and profile image URLs. You can enable email discovery to populate a professional email column alongside the profile data using integrations with Dropcontact, Hunter, or Snov.io. It pushes directly to Google Sheets or HubSpot, so your CRM stays clean without manual copy-paste.

One thing worth knowing: the LinkedIn Profile Scraper does not extract profile pictures directly. For full work history, skills and endorsements, and profile images, PhantomBuster recommends the LinkedIn Profile Visitor instead. Know the difference before you build your workflow around the wrong Phantom.

The Profile Scraper also supports running multiple LinkedIn accounts simultaneously, which means if you have three accounts feeding into it, you can extract up to three times the daily volume without duplicating your automation setup.

LinkedIn Sales Navigator Search Export

This one takes your Sales Nav search URL and exports every profile in the results to a spreadsheet. If you're running a filtered search - say, VPs of Marketing at SaaS companies with 50-500 employees in the US - this Phantom turns that filtered list into a downloadable CSV. It's the fastest way to pull a targeted prospect list out of Sales Nav without manual clicking. This should be your first Phantom if you're new to PhantomBuster - get comfortable with a small export of 10-20 profiles before you scale up.

LinkedIn Company Scraper

Point it at a list of company LinkedIn pages and it pulls company-level data - headcount, industry, location, description. You can chain this with the Profile Scraper: first scrape the company pages of your target accounts, then pull the individual employees you want to reach. That's a clean two-step workflow for account-based prospecting.

LinkedIn Auto-Connect and Outreach Phantoms

Beyond scraping, PhantomBuster has action-based Phantoms that send connection requests and messages automatically. This is where most people get nervous - and rightfully so. LinkedIn actively flags accounts running connection requests at volume. Keep auto-connect runs conservative. Their own documentation flags that running too many actions in a single launch can shorten your session cookie's life and trigger account restrictions.

If you're going to use the outreach Phantoms, run them slow. LinkedIn's algorithm isn't stupid, and I'll cover exactly how to stay safe in the section below.

LinkedIn Safety Limits: What Actually Gets Accounts Banned

This section matters more than most guides admit. The simple "stay under X connection requests per day" advice you see everywhere is an oversimplification that gets accounts restricted.

LinkedIn does not enforce a fixed action counter. It evaluates behavioral patterns over time - how your activity changes relative to your own baseline, whether multiple signals appear simultaneously (session instability, sudden spikes, IP inconsistencies), and whether your overall account behavior looks human or scripted. A sudden jump from minimal activity to 20-30 connection requests per day is riskier than consistently sending 25 per day for months, even though the absolute number is the same.

Here's what the community consensus and observed data actually shows about safe daily limits:

• Connection requests: 15-25 per day for most accounts. Keep pending unaccepted invitations low - ideally a few hundred or fewer. Withdraw requests older than 30 days to maintain account health.
• Profile views: Around 80 per day on a warmed, established account. Start lower on newer accounts and ramp gradually.
• Messages to 1st-degree connections: Up to 50-100 per day, but space follow-ups 3-5 days apart rather than blasting everyone at once.
• Weekly connection cap: Roughly 100 requests per week for newer accounts. Higher-SSI (Social Selling Index) accounts with a long activity history can push higher, but not dramatically so.

LinkedIn evaluates activity at the account level - not per Phantom. If you're running multiple Phantoms on the same LinkedIn account simultaneously, LinkedIn sees the combined total of all actions. Three Phantoms sending 25 connection requests each is 75 requests, not 25. Factor that in before you stack automations.

The other thing that matters: acceptance rate. If your connection requests are getting ignored at a high rate, LinkedIn interprets that as spam behavior. Aim for at least 30-40% acceptance on your requests. That requires targeting precisely and personalizing connection notes - generic "Let's connect" messages tank your acceptance rate and your account health simultaneously.

If you see LinkedIn forcing re-authentication, unexpected session disconnects, or warning emails, stop immediately. Those are early signals. Ignoring them and pushing through is what turns a temporary restriction into a permanent ban.

One practical safeguard: treat any LinkedIn account you use with PhantomBuster as a dedicated prospecting account separate from your main professional profile. If the account gets restricted, you lose the automation capacity - not your entire professional network and years of connections.

How the Pricing Actually Works (and Where It Gets Complicated)

PhantomBuster's current paid plans run from roughly $56 to $352 per month on annual billing, or $69 to $439 month-to-month across three tiers. The Starter plan gets you into the tool, but the tier most serious users end up on is the middle Pro plan - and that price jump is significant because execution hours quadruple.

The thing most people don't realize until they're already in: PhantomBuster bills on three separate caps simultaneously - execution hours, slots (how many Phantoms can run at once), and credits (AI, email, URL finder). You can hit any one of those three walls before the month is up. Run a LinkedIn profile scraper on a list of 200 profiles and you've burned 30-60 minutes of execution time in one session. On the Starter plan's 20 hours per month, that math leaves you running roughly one meaningful Phantom per day maximum. Most teams discover this in week two and upgrade.

Credits add another wrinkle: email discovery is counted per attempt, not per result. So if PhantomBuster tries to find an email for a prospect and comes back empty, you've still spent a credit. URL finding and AI usage have their own separate pools. Run out of execution hours but have email credits left - you're locked. Have hours but no email credits - locked on that function. Three separate quotas, each capping independently.

No rollover on unused credits or hours. Whatever you don't use, you lose at month reset. For teams running consistent, high-volume prospecting, that math gets frustrating fast.

There's also the cookie management overhead. PhantomBuster runs automation through your browser session cookie. If your cookie expires mid-run, the Phantom stops. For solo users it's manageable. For agencies running multiple client accounts, it's a recurring maintenance task that adds up over time. Budget for that operational overhead before you commit to it as a core workflow tool.

How to Build a Real LinkedIn Prospecting Workflow with PhantomBuster

Here's the actual workflow I'd run if I were using PhantomBuster today to fill a pipeline:

1. Start with a Sales Nav search. Filter by title, seniority, geography, company size, and industry. Get your search tight before you export - a wide search just creates a messy list you'll have to clean later. Download our Sales Navigator Guide for the filter combinations that generate the highest-quality lists before you ever touch an export tool.
2. Run the Sales Navigator Search Export Phantom. Start with a small batch - 10-20 profiles - to confirm the setup is working correctly before you scale. Export the full list to a Google Sheet. You'll get LinkedIn profile URLs, names, titles, and companies.
3. Run the LinkedIn Profile Scraper on that list. Enable email discovery. Now you have enriched profiles with up to 44 data points per contact - not just LinkedIn URLs.
4. Validate the emails before you send anything. PhantomBuster's email discovery is useful but not perfect. On a list of 500 profiles, coverage might land somewhere between 60-70%. Run the list through an email validation tool to strip bounces before they damage your sending domain. Bounce rates above 3-4% will hurt your deliverability fast.
5. Load verified leads into your cold email tool. I use Smartlead or Instantly for sending at scale. Both handle inbox rotation and warm-up, which matters when you're sending volume.
6. Run LinkedIn connection requests in parallel, slowly. Keep the auto-connect Phantom at 15-25 per day. Touch the same prospects on email and LinkedIn - the double-channel approach lifts reply rates significantly in my experience.

That six-step flow is repeatable. Once it's set up, you can run it weekly with new searches and keep a fresh list flowing into your sequences.

Enriching PhantomBuster Data: Filling the Gaps It Leaves

PhantomBuster's data output is a starting point, not a finished lead list. Here's what you'll typically need to layer on top:

Email gaps

Even with email discovery enabled, PhantomBuster won't return a verified email for every profile in your list. The coverage depends on which enrichment service you've connected (Dropcontact, Hunter, or Snov.io) and how well-indexed those prospects are. For prospects PhantomBuster can't find, run the remaining URLs through a dedicated email finding tool before you write those contacts off. A second enrichment pass on the gaps often recovers another 15-20% of the list.

And regardless of source - PhantomBuster, a standalone finder, or a database - validate every email before you load it into a sending sequence. Unvalidated lists cause bounce rates that damage your domain reputation over time, and that damage compounds.

Phone number gaps

PhantomBuster does not provide phone numbers. If cold calling is part of your outbound, you need to layer in a separate tool. For direct dials and mobile numbers on the prospects your PhantomBuster list surfaces, a mobile number finder gives you direct dials without a separate prospecting workflow.

Pre-built database as an alternative starting point

If you want to skip the scraping step entirely for your initial list and go straight to filtered, verified contact data, a tool like ScraperCity's B2B email database lets you filter by title, seniority, industry, location, and company size and pull lists directly - without setting up Phantoms, managing cookies, or burning execution hours. For some workflows, it makes more sense to start here and use PhantomBuster only for LinkedIn-specific data points you can't get from a database.

CRM and tech stack enrichment

If you're targeting companies based on the tools they use - say, finding everyone running a specific CRM or marketing platform - a BuiltWith scraper gives you technographic filters that PhantomBuster doesn't offer natively. Layer technographic data onto your PhantomBuster-sourced list and you can narrow down which companies are worth prioritizing before you spend credits on enrichment.

Integrating PhantomBuster with the Rest of Your Stack

PhantomBuster is a data extraction and light automation layer - it's not a complete outbound system. You're going to stack it with other tools regardless. Here's how the pieces connect:

CRM: PhantomBuster pushes data directly to Google Sheets, HubSpot, Salesforce, and Pipedrive. If you're using Close as your CRM, export from PhantomBuster to a CSV and import - there's no native direct integration, but the handoff takes two minutes. Keep your CRM as the system of record so you're not chasing duplicate contacts across spreadsheets.

Cold email: PhantomBuster extracts and automates LinkedIn actions. It does not send cold email sequences. Once your list is enriched and validated, load it into a dedicated sending platform. Smartlead and Instantly both handle inbox rotation and warm-up. Lemlist is worth considering if you want dynamic personalization variables at the template level.

LinkedIn sequencing: If you want a full LinkedIn sequence - connection request, then follow-up message, then InMail - in one managed campaign flow, tools like Expandi are purpose-built for that and handle the pacing and safety limits more gracefully than PhantomBuster's action Phantoms do. Think of PhantomBuster for data extraction, and a dedicated LinkedIn sequence tool for the follow-up cadence.

Workflow automation: PhantomBuster connects to Zapier and Make, which opens up routing possibilities - push new PhantomBuster leads to a Slack notification, trigger a Clay enrichment workflow, or auto-add to a specific sequence. If you're using Clay for enrichment and waterfall logic, you can pipe PhantomBuster CSVs straight into a Clay table and run additional enrichment passes without rebuilding your list from scratch.

Where PhantomBuster Falls Short for LinkedIn

PhantomBuster is genuinely versatile - it works across LinkedIn, Instagram, Twitter, Google Maps, and more. That breadth is its strength, but it also means it's not purpose-built for LinkedIn sales workflows. If all you need is LinkedIn prospecting, you're paying for infrastructure you'll never use.

A few specific gaps matter for sales teams:

• No native phone number finding. PhantomBuster doesn't give you direct dials or mobile numbers. If cold calling is part of your outbound, you need to layer in a separate mobile finder tool.
• Email discovery is hit-or-miss at volume. It works, but coverage isn't 100%. On a list of 500 profiles, you might get verified emails for 60-70% of them. The rest need another enrichment pass.
• No built-in email sequencing. PhantomBuster extracts and automates LinkedIn actions. It doesn't send cold email sequences. You need a separate sending platform.
• Cookie management is a real operational burden. For solo users it's manageable. For agencies running multiple client accounts, it's a recurring headache. Session cookies can expire at any time, especially if LinkedIn forces re-authentication.
• Three separate billing caps create forecasting complexity. Most other tools use a single credit pool. PhantomBuster's three independent quotas - execution hours, slots, and credits - add planning overhead that solo operators and small teams often underestimate.
• Learning curve is real. Users consistently report that the platform requires significant time reading documentation before you can run reliable workflows. It's genuinely no-code, but "no-code" doesn't mean "no setup."

When to Use PhantomBuster vs. When to Use Something Else

PhantomBuster makes the most sense if you're running automation across multiple platforms simultaneously - LinkedIn, Twitter, Instagram, and Google - and you need one tool to handle all of it. For growth teams and agencies managing multi-channel data extraction, the Phantom library is hard to beat on flexibility.

For pure B2B sales prospecting, though, there are more focused options worth knowing:

• If you want a pre-built B2B lead database - rather than scraping LinkedIn yourself - this B2B lead database lets you filter by title, seniority, industry, location, and company size and pull lists directly, without needing to set up Phantoms, manage cookies, or burn execution hours.
• If you're using Apollo.io already and want to export your search results without the platform's built-in limitations, the Apollo scraper is a simpler path than rebuilding the same workflow in PhantomBuster.
• If local business prospecting is your focus - contractors, restaurants, service businesses - PhantomBuster's Google Maps Phantom exists but it's not the most efficient option. A dedicated Google Maps scraper purpose-built for local lead gen will give you faster, cleaner results.
• If LinkedIn outreach sequencing is what you actually need - where you want connection request, follow-up message, and InMail in one managed flow - Expandi is purpose-built for that and handles the safety limits more gracefully than PhantomBuster's action Phantoms.
• If you're a solo user on a tight budget who just needs basic LinkedIn sequences, Waalaxy is worth a look. It's a LinkedIn prospecting and cold emailing tool designed for simplicity, with pre-built campaign templates and a lower price point for entry-level volume. The tradeoff: it has basic data extraction capabilities and you can't build full-fledged prospect databases with it.
• If you need to find someone's email from a partial name and company, run that separately through an email finder rather than burning PhantomBuster credits on every profile.

PhantomBuster vs. the Main Alternatives: A Direct Comparison

I get asked this a lot. Here's how the tools stack up honestly for LinkedIn-focused B2B sales use cases:

PhantomBuster vs. Expandi: PhantomBuster wins on data extraction breadth and multi-platform coverage. Expandi wins on LinkedIn sequencing - it's a campaign management tool, not a job runner, which means reps can focus on replying rather than wiring together CSV exports and Phantom chains. If LinkedIn is your primary outreach channel and you want a campaign flow with safety controls baked in, Expandi is the more purpose-built option. PhantomBuster is better when data extraction is the primary goal and sequencing happens in a separate tool.

PhantomBuster vs. Waalaxy: Waalaxy is the fast-launch option for teams new to LinkedIn automation. You install the Chrome extension, pick a campaign template, and you're sending connection requests within minutes. The free tier (roughly 80 LinkedIn invites per month) lets you test before committing. The ceiling is also lower - Waalaxy is not built for complex data extraction workflows, doesn't export to detailed prospect databases, and is limited to LinkedIn plus email. PhantomBuster handles more complex workflows across more platforms. For large-scale scraping or multi-platform needs, PhantomBuster is the stronger tool.

PhantomBuster vs. a B2B database: This is a fundamentally different question. PhantomBuster scrapes live data - it's pulling current profile information every time you run it. A pre-built database like ScraperCity's B2B email database gives you filtered, ready-to-export lists without any setup, cookie management, or execution hour math. If your goal is building a prospect list fast rather than building a custom scraping workflow, the database approach saves significant operational overhead. Many teams use both: database for initial list-building at volume, PhantomBuster for LinkedIn-specific signals and real-time profile data.

The Right Mental Model for PhantomBuster

Think of PhantomBuster as a data extraction and light automation layer, not a complete outbound system. It's good at getting data out of LinkedIn and triggering simple actions at scale. It's not a sending platform, not a CRM, and not a phone finder. You're going to stack it with other tools regardless.

The teams that get the most out of PhantomBuster treat it as step one of a multi-tool workflow: scrape and enrich with PhantomBuster, validate emails with a dedicated verifier, find phones separately, send sequences through a proper cold email platform, and close deals in a CRM like Close.

If you want to skip some of the assembly and just get clean, verified B2B contact data, check out our LinkedIn Playbook - it covers the full outbound workflow including which data sources produce the cleanest lists and how to structure your sequences for maximum reply rates.

A Few Things to Never Do with PhantomBuster on LinkedIn

Since I've seen these mistakes tank accounts:

• Don't run connection requests above 20-25 per day on most accounts. Even that is aggressive on a profile with limited activity history. Ramp up slowly over weeks, not days.
• Don't spike activity after a period of inactivity. Two quiet weeks followed by 20-30 connection requests per day is a common trigger pattern. The change from your baseline matters more than the absolute number.
• Don't scrape 1,500 profiles in one launch. Spread it across multiple runs during business hours. Running large scrapes in a single launch at off-hours looks nothing like human behavior.
• Don't ignore cookie expiry warnings. A dead session cookie mid-run doesn't just pause your Phantom; it can leave incomplete data in your spreadsheet that takes time to reconcile. Check your session status before kicking off large runs.
• Don't skip email validation before loading lists into your sender. Unvalidated lists from any scraping tool - PhantomBuster included - will contain bad addresses. Even a 5% bounce rate will damage your sending reputation.
• Don't run multiple action Phantoms simultaneously on the same LinkedIn account. LinkedIn evaluates combined activity across all Phantoms. Three Phantoms sending connection requests at the same time adds up to three times the individual volume from LinkedIn's perspective.
• Don't send generic connection notes at volume. Low acceptance rates signal spam behavior. Personalized notes to precisely targeted prospects protect your account health and lift your reply rate.

If you want to see how I personally structure LinkedIn voice note outreach as a separate channel alongside scraping-based cold email, grab the LinkedIn Voice Note Script - it's one of the highest-reply tactics I've seen in the last couple of years, and it pairs well with a PhantomBuster-built list.

Quick Reference: PhantomBuster LinkedIn Phantom Cheat Sheet

For practitioners who want the summary without re-reading the whole guide:

Bottom Line

PhantomBuster is a legitimate tool for LinkedIn scraping and automation. If you're technical enough to set up Phantoms, chain them together, and manage session cookies without it becoming a weekly maintenance project, you can build a solid prospecting workflow with it. The Sales Navigator Search Export and Profile Scraper Phantoms in particular are genuinely useful - they do what they claim to do, and the Google Sheets integration means your data lands somewhere usable without manual cleanup.

Just go in knowing what it is: an extraction and light automation platform, not a complete outbound system. The all-in cost once you factor in execution limits, complementary tools, and the operational overhead of cookie management is higher than the headline plan price suggests. The safety risk on LinkedIn is real and requires active management - not just a set-it-and-forget-it configuration. And the learning curve is steeper than the marketing copy implies.

Build your stack accordingly, keep your LinkedIn actions conservative and consistent, and validate everything before you send. That's how you actually get meetings out of it.

If you want hands-on help building and optimizing your full outbound stack - not just the PhantomBuster piece - I cover this inside Galadon Gold with practitioners who are running these workflows right now.