PhantomBuster LinkedIn Outreach: Full Setup Guide

What PhantomBuster Actually Does (and What It Doesn't)

PhantomBuster is a cloud-based automation platform that runs pre-built scripts - called Phantoms - to scrape data, enrich leads, and trigger outreach sequences on LinkedIn and other platforms, all without writing a single line of code. It launched in 2016 and has built a real reputation among growth teams for one specific reason: it's a toolbox, not a system.

That distinction matters. PhantomBuster gives you 130+ individual automation scripts. You assemble them into workflows yourself. Scrape a Sales Navigator search, enrich those profiles with emails, push to a CRM, trigger a message sequence - each step is a separate Phantom you configure and chain together. If you like building things and want complete control over your pipeline, that's genuinely powerful. If you want a plug-and-play outreach system that handles everything end-to-end, you'll need to pair it with other tools.

For LinkedIn specifically, PhantomBuster is strongest at three things: profile scraping and data extraction, automated connection requests, and follow-up message sequences after someone accepts. It does not replace a dedicated cold email platform for email outreach, and it doesn't manage replies or booked meetings on its own.

One thing worth noting upfront: PhantomBuster does not have its own email or contact database. It relies on enrichment credits and external data sources to find emails - and accuracy varies depending on the profile. For building prospect lists outside of LinkedIn, or enriching scraped data with verified contact info, you need something else in your stack. More on that below.

The Core LinkedIn Phantoms Worth Knowing

Not all Phantoms are created equal. For outreach, there are a handful that actually move the needle:

• LinkedIn Search Export - Pulls profile data from any LinkedIn search or Sales Navigator search result into a structured CSV. Job titles, company names, locations, skills. This is your list-building engine and the right place to start before you run any outreach.
• LinkedIn Auto Connect - Sends connection requests with a custom note to a list of profile URLs. LinkedIn caps you at around 100 invites per week, and PhantomBuster's own best practice is to spread those out to roughly 20 per working day. Don't try to send 100 in a day.
• LinkedIn Outreach - The more complete version of the above. Sends a connection request plus up to three follow-up messages after acceptance, giving you a basic drip sequence inside LinkedIn itself. Follow-up messages are only sent if the prospect hasn't replied to a previous message, which keeps the sequence from feeling robotic.
• Sales Navigator Search Export - Pulls lead data directly from Sales Navigator searches. If you're paying for Sales Nav, this is how you actually extract that data into something you can work with. You can also chain it to the LinkedIn Outreach Phantom using Watcher mode to continuously feed new leads into your sequence as they appear in search results.
• AI LinkedIn Message Writer - Uses GPT to generate personalized connection request messages or follow-ups based on a prospect's profile data. Feed it a profile URL and it returns a tailored message. Useful for avoiding the copy-paste problem at scale. These use AI credits, which are included in paid plans but expire monthly.
• LinkedIn Post Commenter and Liker Scraper - Extracts profiles of people who commented on or liked a specific LinkedIn post. This is one of the most underused Phantoms in the library. Commenters show higher intent than likers - they took time to write something - but both groups are warmer than a cold search export. You can target your own posts, competitor posts, or any thought leader content relevant to your niche.
• LinkedIn Group Members Export - Pulls member profiles from any LinkedIn group you belong to. Groups are a reliable source of pre-segmented, niche-specific leads who have already self-identified as interested in a topic.

For a complete outreach workflow, you chain these: Search Export or Post Engager Scraper → Auto Connect (with AI-written messages) → LinkedIn Outreach for follow-ups after acceptance. That's a functional pipeline that moves from cold list to active conversation without manual intervention at each step.

The Workflows Feature: Pre-Built Pipeline Sequences

PhantomBuster distinguishes between individual Phantoms - single-step automations - and Workflows, which chain multiple Phantoms together into a complete sequence you configure once. For LinkedIn outreach, the Workflows are where most users should start before they try to build custom Phantom chains from scratch.

The key LinkedIn Workflows to know:

• LinkedIn Search to Outreach - Converts a LinkedIn or Sales Navigator search directly into an outreach sequence. Extract leads, send connection requests, follow up after acceptance, all in one configured flow.
• LinkedIn Post Engagers to Lead Outreach - Extracts profiles of people who liked or commented on a post, filters them, and routes them into connection request and follow-up sequences. This is the highest-intent source of leads available in the platform - people who are already active and engaged on the topics you care about.
• LinkedIn Group Members to Outreach - Pulls group member profiles and sends them directly into an outreach sequence. Useful for ICP-specific targeting when the right groups exist in your niche.
• LinkedIn Company Follower Collector to Outreach - Starts conversations with people who follow your company page. These are warm inbound leads who already know your brand - an easy win that most teams ignore.

The practical difference between Workflows and manually chained Phantoms: Workflows handle the handoff between steps automatically. You don't have to export from one Phantom and import to the next. The tradeoff is less configurability - Workflows have fixed structures, and the scheduling options are limited to presets like "weekdays during working hours" or "randomly throughout the day." If you need granular control over timing, manually chained Phantoms give you more flexibility.

Setting Up Your First PhantomBuster LinkedIn Workflow

The setup process has a real learning curve. Here's how to approach it without wasting a week troubleshooting.

Step 1: Connect Your LinkedIn Session

PhantomBuster runs through a stored LinkedIn session cookie - an authentication token that lets it act as your logged-in account. Install the PhantomBuster Chrome or Firefox extension, open LinkedIn in the same browser, then click "Connect to LinkedIn" inside any Phantom's setup page. The field auto-populates. Save it. This session can expire if you log out of LinkedIn or if LinkedIn forces a re-authentication, so expect to refresh it occasionally.

One critical note: LinkedIn Phantoms run via stored session cookies, which is technically against LinkedIn's Terms of Service. Account restrictions are a real risk at higher volumes. PhantomBuster includes human-like delays and configurable schedules to reduce this risk, but the responsibility for safe usage is ultimately on you. Manage your own rate limits, use delays between actions, and stay conservative on volume especially when starting out.

Step 2: Build Your Prospect List

Start with the LinkedIn Search Export Phantom. Run a LinkedIn or Sales Navigator search with your targeting filters, copy the search URL, paste it as input, and set your first run to export 10 results. Validate the output - check that names, titles, and companies are pulling correctly - before you scale up. Don't jump straight to exporting 500 profiles on day one. LinkedIn notices behavioral changes more than raw volume, so ramp gradually.

If you want a cleaner, larger prospect list before running any LinkedIn automation, a dedicated B2B lead database is worth adding to your stack. ScraperCity's B2B email database lets you filter by job title, seniority, industry, location, and company size - useful when you want a list that's fully built before you touch LinkedIn. You can also grab emails from that list to run parallel cold email outreach while your LinkedIn sequences are warming up.

For the LinkedIn approach specifically, build your list first, verify the URLs, then move to outreach. If you're using a Google Sheet as input - which PhantomBuster supports natively - make sure the sheet is shared with "Anyone with the link" or the Phantom won't be able to read it.

Step 3: Run Connection Requests

Load your list of LinkedIn profile URLs into the Auto Connect Phantom. Write a short, specific connection note - or use the AI Message Writer Phantom to generate variations based on each person's profile data. Keep it under 200 characters. Do not pitch in the connection request. Just establish a reason to connect that's relevant to them.

Schedule the Phantom to run twice during business hours, sending 10 invites per launch. That keeps you well under LinkedIn's weekly limit with room to spare. Enable the "wait before sending" option - it mimics more human browsing behavior and reduces detection risk.

One thing to watch: keep your pending invitations low. If you have hundreds of unaccepted requests sitting in your queue, that itself becomes a signal. Withdraw requests older than 30 days regularly. PhantomBuster has a LinkedIn Auto Invitation Withdrawer Phantom that handles this automatically - run it as part of your weekly maintenance.

Step 4: Follow Up After Acceptance

The LinkedIn Outreach Phantom handles follow-ups automatically once someone accepts your connection. Set a delay of at least 24 hours between the connection acceptance and your first message, and another day or two before any subsequent message. Three messages total is a reasonable sequence - a warm opener, a value-add follow-up, and a short direct ask.

The Phantom only sends the next message if the prospect hasn't replied to the previous one. That means if someone responds after your first message, they won't get the second and third automatically - which is exactly right. The sequence pauses when it detects a reply. You take it from there manually.

Download your accepted connections and results as a CSV, import them into your CRM, or push them directly to HubSpot, Salesforce, or Pipedrive using PhantomBuster's native integrations. The Report tab inside each Workflow also shows you performance metrics: time saved, invites sent, requests accepted, and message sequence performance over time.

The High-Intent Lead Source Most People Miss: Post Engagers

Here's the PhantomBuster use case that consistently outperforms cold search-based outreach, and most people running LinkedIn automation never touch it.

The Post Commenter and Liker Scraper lets you pull profiles of people who engaged with any LinkedIn post - yours, a competitor's, or a thought leader's. You provide a post URL. PhantomBuster returns a structured list of profiles who liked or commented, ready to feed into your outreach sequence.

Why this works better than cold list outreach: these people have already demonstrated interest in the topic. They showed up, they engaged, they left a comment or a reaction. That's intent signal. When your connection request references the post or the topic they engaged with - "saw your comment on [specific post], wanted to connect" - your acceptance rate goes up because the outreach is grounded in something real.

Commenters typically show higher intent than likers because they invested time in writing a response. For maximum coverage, run both exports and merge the lists. Then filter by job title, location, and company size to match your ICP before you start the outreach sequence.

You can also run this against competitor content. Use the LinkedIn Activity Extractor to find recent posts from key competitors, run the Post Likers and Commenters Export against those posts, and build a pipeline of prospects who are already actively engaged in your space. They're evaluating solutions. They're reading about problems your product solves. That's the warmest cold outreach you can do.

The Real Limits of PhantomBuster for LinkedIn Outreach

PhantomBuster is strong for data extraction and for setting up basic outreach automation, but it has real gaps you need to know about before you build your entire outbound stack around it.

It doesn't manage replies. Once someone responds, you're back to doing that manually. There's no unified inbox, no reply detection that pauses sequences globally, no meeting booking logic. The Phantom tracks whether someone replied and stops automated follow-ups to that contact, but it doesn't give you a place to actually manage the conversation. If you're running volume, this becomes a real operational bottleneck.

The execution time caps bite at scale. Every minute an automation runs burns through your plan's execution time. Large enrichment jobs, multi-step Workflows, and high-volume scraping eat through that quota fast. There's no top-up option - if you hit the ceiling mid-month, you either pause until the next billing cycle or upgrade to a higher plan permanently. Size your plan based on actual volume, add a 20% buffer, and don't assume the Starter plan handles serious prospecting at scale.

It doesn't have its own email database. PhantomBuster can find some email addresses through its enrichment integrations, but it relies on external data sources and accuracy varies. For finding contact emails at scale, you're better served pairing it with a dedicated tool. This email finding tool fills that gap directly - useful when LinkedIn profile data doesn't include a verified work email and you need to reach someone through a parallel channel.

The learning curve is real. Setting up browser sessions, configuring Phantoms correctly, managing cookies, building multi-step workflows, and troubleshooting when things break takes time. Users report meaningful ongoing overhead for configuration, safety management, cookie refresh cycles, and workflow maintenance. It's not hard once you understand the system, but don't expect to be running a polished automated pipeline on day one.

No no-code conditional logic. PhantomBuster Workflows follow fixed sequences. If you want dynamic branching - different messages based on job title, different follow-up sequences based on whether someone viewed your profile - you're either building custom Phantom chains or integrating with Clay or Zapier to handle the logic layer.

What to Pair With PhantomBuster for a Complete Outbound Stack

PhantomBuster works best as part of a stack, not as a standalone outreach system. Here's how I'd put it together:

• List building: LinkedIn Search Export or Sales Nav Export inside PhantomBuster, or pull a pre-built list from a B2B lead database when you want to work outside LinkedIn entirely.
• LinkedIn outreach: PhantomBuster for connection requests and follow-up sequences. Alternatively, Expandi if you want built-in account safety features and a more structured campaign interface without assembling the workflow yourself.
• Email outreach (parallel): Run cold email sequences alongside your LinkedIn touches using Smartlead or Instantly. LinkedIn gets you the warm relationship; email gives you another touchpoint and keeps you in control of deliverability.
• Workflow automation and enrichment: Clay integrates directly with PhantomBuster via two-way sync and webhooks. Use it to enrich PhantomBuster-scraped data, score leads against your ICP, and route qualified prospects into different sequences automatically. It's what bridges the gap between raw PhantomBuster exports and truly personalized outreach.
• CRM: Push all your accepted connections and engaged prospects into a CRM like Close so you can track follow-ups and manage the pipeline without losing anyone. PhantomBuster has native HubSpot integration - you can sync extracted leads directly into HubSpot Contacts if that's your CRM of choice.
• Lead enrichment and data: If you need to enrich your PhantomBuster-scraped list with direct contact numbers for a multi-channel sequence that includes calls, a mobile finder tool can append direct dial numbers to your prospect records.
• Email verification: Before you run cold email in parallel with your LinkedIn sequence, clean your list. An email validator removes bad addresses before you send - protecting deliverability and keeping bounce rates low.

For a deeper breakdown of how to structure your LinkedIn prospecting strategy - including what to say in your connection requests and follow-ups - grab the LinkedIn Playbook. It covers the sequencing logic in detail.

PhantomBuster Pricing: What You're Actually Getting

PhantomBuster structures its pricing around three core resources: execution time (how many hours per month your automations can run), Phantom slots (how many automations you can keep active simultaneously), and credits for email enrichment, AI message generation, and URL lookups.

Plans start at $69/month for the Starter tier. Higher tiers unlock more execution hours, more active Phantom slots, and higher credit limits. All plans include access to Phantoms and Workflows, unlimited CSV and JSON exports, API access, and up to 100 workspace members. Pricing is per workspace, not per seat - so adding collaborators doesn't add cost. A 14-day free trial is available on paid plans.

The thing to watch: all credits expire monthly with no rollover. If your automations hit the execution ceiling mid-campaign, they stop until your next billing cycle. There's no top-up option - you upgrade to the next tier permanently or you wait. Size your plan based on actual volume. A practical way to estimate: a typical LinkedIn profile scrape takes roughly 10-20 seconds per profile. Scraping 1,000 profiles is 3-5 hours of execution time before you've touched any outreach or enrichment steps. Add follow-up sequences and enrichment on top of that, and mid-tier plans fill up faster than most users expect.

Also worth flagging: the Starter plan limits you to a small number of concurrent Phantom slots, which means if you're trying to run list export, connection requests, and follow-up sequences simultaneously, you'll hit those slot limits. Build in a 20% execution buffer and verify your slot count matches your workflow complexity before committing to a plan.

PhantomBuster vs. Dedicated LinkedIn Outreach Tools

The comparison that comes up most often is PhantomBuster versus tools like Expandi, Dripify, or other dedicated LinkedIn sequencers. The honest take: they're solving slightly different problems.

PhantomBuster is a multi-platform automation toolbox that happens to include strong LinkedIn capabilities. It gives you flexibility and raw data extraction power that purpose-built LinkedIn tools don't match. But it puts the workflow assembly and account safety responsibility entirely on you.

Expandi, by contrast, is built specifically for LinkedIn outreach campaigns. It includes built-in throttling, randomized timing to reduce restriction risk, and a more structured campaign interface. Less flexible, but faster to get running and safer for people who don't want to manage rate limits manually.

Dripify is another option in the dedicated LinkedIn sequencer category, with a more user-friendly interface for managing multiple drip campaigns and built-in analytics. It's more prescriptive than PhantomBuster but requires less technical setup.

If you're technically comfortable and want to build custom multi-step workflows across LinkedIn and other platforms, PhantomBuster is the right choice. If you just want LinkedIn outreach running safely with minimal configuration, a dedicated tool is probably a better fit. The key questions to ask yourself: Do you need data extraction across multiple platforms, or just LinkedIn? Are you comfortable maintaining sessions, cookies, and workflow logic? Do you have the time to troubleshoot when something breaks?

Also worth considering: if you're running LinkedIn and cold email together as a multi-channel sequence, tools like Lemlist or Reply.io handle both channels in a single platform - which eliminates some of the manual handoff work between PhantomBuster and your email tool. The tradeoff is less scraping power and data export flexibility.

Account Safety: The Non-Negotiable Stuff

This deserves its own section because getting your LinkedIn account restricted kills your outreach pipeline completely.

The most important thing to understand about LinkedIn's enforcement: it's pattern-based, not counter-based. LinkedIn evaluates your current behavior against your historical baseline. If you've been quiet for weeks and suddenly send 50 connection requests in a day, that change looks suspicious even if 50 is within commonly cited limits. The riskiest factor is the behavioral spike, not the absolute number. Copying someone else's volume settings doesn't mean those settings are safe for your account specifically - your baseline is different from theirs.

A few rules that actually matter:

• Stay under 20 connection requests per day, spread across working hours. LinkedIn's hard cap is around 100 per week, but operating near that ceiling consistently will trigger friction - especially on newer accounts or accounts that recently had low activity. For new accounts, start at 50-70 per week and only increase if your 7-day acceptance rate stays above 40% with no warnings.
• Keep your acceptance rate above 30%. If it drops below that, your targeting is off or your messaging is off - and LinkedIn will notice the pattern of ignored requests. Fix the targeting first. According to PhantomBuster's own data, sellers who personalize connection requests with context specific to the recipient are 4-5x more likely to hit acceptance rates above 40%.
• Withdraw unaccepted requests older than 30 days. Keeping a large backlog of pending invites is an account health signal. Run the Auto Invitation Withdrawer Phantom weekly to clean this up automatically.
• Use delays between profile visits and actions. Behavior that looks human - visiting a profile, pausing, then acting - is far less detectable than rapid-fire automated actions. Enable the wait options in your Phantoms and don't override them.
• Warm up a new LinkedIn account gradually before running any automation. Spend one to two weeks doing light manual activity - profile visits, post reactions, thoughtful comments - before sending any connection requests. This establishes a realistic activity baseline that automation can then extend, rather than creating a sudden behavioral spike from nothing.
• Don't run Phantoms outside business hours in your target market's timezone. LinkedIn notices accounts that are "active" at 3 AM. Schedule within working hours windows.
• Don't run multiple automation tools on the same LinkedIn account simultaneously. Overlapping logs and unnatural timing from two tools running on the same account can push you over LinkedIn's limits faster than either tool would alone.
• Monitor for session friction: repeated logouts, cookie expirations, verification prompts. These are early warning signals. When they increase, slow down and stabilize before they escalate into restrictions. If you see a restricted banner, stop all automations immediately, use LinkedIn manually only for 3-7 days, then restart with conservative warm-up settings.

If you want to add LinkedIn voice notes to your sequence - which consistently outperform text-only connection requests for response rates - download the LinkedIn Voice Note Script for a tested framework you can use alongside your PhantomBuster automation.

Integrations: How PhantomBuster Connects to the Rest of Your Stack

PhantomBuster's value multiplies significantly when you connect it to other tools in your pipeline. On its own it's a data collection and LinkedIn action layer. Integrated properly, it becomes part of a full outbound system.

HubSpot: The most robust native integration. You can push extracted leads directly to HubSpot Contacts, enrich existing contacts from PhantomBuster data, track job changes, and automate LinkedIn outreach tracked inside your CRM. You can also pull LinkedIn profiles from a HubSpot contact list as input for a Phantom, which closes the loop between your CRM and LinkedIn prospecting. Make sure each contact has a LinkedIn Profile URL attached - use the LinkedIn Profile URL Finder Phantom first if they don't.

Salesforce and Pipedrive: Available as push destinations for accepted connections and engagement data. Less native than the HubSpot integration but functional for teams already invested in those CRMs.

Clay: Two-way sync between PhantomBuster and Clay via webhooks. This is the combination that serious growth teams use. PhantomBuster handles the LinkedIn scraping and actions; Clay handles enrichment, ICP scoring, conditional routing, and personalization at scale. If you want to build truly dynamic sequences based on lead attributes, Clay is the layer that makes it possible without custom code.

Zapier and n8n: Launch Phantoms via Zapier triggers or HTTP requests in n8n. This is useful for building automations that respond to external events - a new contact added to a spreadsheet, a form submission, a CRM status change - and automatically kick off a PhantomBuster workflow. Zapier is easier; n8n gives you more control and costs less at scale.

Lemlist: You can send data from LinkedIn Search or LinkedIn Profile Phantoms to Lemlist for personalized email sequences. This is a common setup for multi-channel outreach: PhantomBuster handles LinkedIn data extraction and connection requests; Lemlist handles the email sequence that runs in parallel.

PhantomBuster vs. Apollo: A Practical Comparison

A question that comes up a lot: if you're already using Apollo for outbound, do you need PhantomBuster at all?

Apollo is primarily a B2B contact database with built-in email sequencing. PhantomBuster is a scraping and automation layer that works on top of LinkedIn and other platforms. They're not direct competitors - they're complementary tools with different primary use cases.

Apollo gives you access to a contact database with email addresses already attached. You build lists, you sequence emails, you track opens and replies in one platform. The limitation is data freshness and coverage - Apollo's database is only as good as its last update, and not every profile has a verified email.

PhantomBuster, by contrast, scrapes LinkedIn in real time. The data you extract is current - job titles, companies, recent posts, actual current roles. But it doesn't come with emails attached by default. You need enrichment credits or external tools to get contact information.

If you're running Apollo-heavy outbound and want to complement it with fresher LinkedIn data, a useful setup is: use PhantomBuster to export a Sales Navigator search, enrich with emails via an Apollo data export tool, then route into your sequencing tool. That way you're getting the targeting precision of Sales Navigator with the contact data coverage of Apollo.

Messaging That Actually Gets Replies

The automation infrastructure only matters if your messages are good. Here's what works and what kills response rates.

Connection requests: Keep it under 200 characters. Reference something specific - a post they wrote, an industry they're in, a challenge relevant to their role. Don't pitch. The only goal of the connection request is to get accepted. "Saw your post on [topic], wanted to connect with others thinking about this" beats "I help companies like yours increase revenue" every time. The former is about them; the latter is about you.

Use the AI LinkedIn Message Writer Phantom to generate variations rather than sending the same message to every profile. Feed it the profile URL and it returns a tailored opener based on job title, company, and recent activity. Review the outputs before you run at volume - AI-generated copy needs a human pass to remove anything that sounds generic or hollow.

First follow-up (after acceptance): Wait at least 24 hours. Don't pitch immediately. A short warm message that references the reason you connected or asks a relevant question outperforms a feature dump. "Thanks for connecting - curious if [specific problem] is something you're dealing with at [Company]?" is the model. Specific question, low friction, easy to respond to.

Second follow-up: Add value. Share something useful - a relevant resource, a specific insight, a case study that's actually relevant to their situation. This is where most automated sequences lose people: they send three variations of the same pitch instead of building any actual value between messages.

Third message: Short direct ask. If the first two didn't land, a longer pitch won't help. Keep it brief: "Worth a quick call to explore if this fits?" and stop there. Three messages with no response means move on. Don't automate beyond that.

The Bottom Line on PhantomBuster for LinkedIn Outreach

PhantomBuster is a genuinely capable tool. The Phantom library is deep, the cloud-based execution model means it runs without your computer being on, and for technically comfortable users it can meaningfully accelerate LinkedIn prospecting and multi-platform data extraction. The Post Engager workflows in particular are underused and consistently outperform cold search-based outreach when the targeting is right.

What it's not: a complete outreach system. It's the data and automation layer of your stack. You still need to pair it with good list sourcing, a way to manage replies, a CRM, and - if you're serious about outbound - a cold email platform running in parallel.

If you want to go deeper on how to structure a full LinkedIn outbound system, including targeting, messaging, and follow-up cadences, check out the Sales Navigator Guide - it covers how to build the exact kind of high-quality search exports that make PhantomBuster outreach actually land.

And if you want live coaching on putting all of this together - the list building, the automation setup, the messaging - I cover it inside Galadon Gold.