If you run the same LinkedIn search every week, you’ll usually see the same profiles near the top. The goal isn’t to “start over,” it’s to keep a cumulative master list and pull a weekly view of net-new leads.
PhantomBuster Automations are cumulative by default. They append results to the same destination and use that history to avoid duplicates. The reliable way to get a fresh weekly list is to keep your master list intact, then filter for what was added since the last run.
This guide shows a repeatable workflow to build a weekly net-new lead system that keeps data clean and reduces avoidable LinkedIn risk.
Why starting fresh each week fails: What to do instead
The common mistake: Deleting history to “clean up”
Many BDRs assume the cleanest approach is to delete last week’s data, rerun the same search, and export a new file.
That breaks the deduplication logic. PhantomBuster solves that and relies on your existing output history to recognize profiles it already processed. If you delete the master list, clear your output, or switch to a new output destination, the next run can treat previously captured profiles as new.
You’ll get duplicate entries, wasted execution capacity, and a higher chance you contact the same person twice.
The cumulative approach: How PhantomBuster keeps a net-new output
PhantomBuster appends results to a single output destination, such as a CSV file, a Google Sheet, or the LinkedIn Leads page in your dashboard. On the next run, it checks what’s already been processed and skips duplicates.
Your “master list” grows over time. Your “weekly list” is a filtered view of what was added since a given date.
How to set up your Phantom for weekly fresh leads
Step 1: Choose the right Automation and lock in your search URL
Use a search export Automation. These are built for repeatable, cumulative lead extraction.
Paste your LinkedIn search URL into the Automation’s configuration. Make sure the search type is People, not Jobs or Posts, so the output stays consistent.
Your search URL is the foundation of the workflow. Too broad, and you create noise. Too narrow, and you’ll run out of new profiles quickly. Aim for a search that can produce new results over time, even after deduplication removes repeats.
Step 2: Turn on duplicate removal and use Watcher mode when it fits
In your Automation settings, enable Remove duplicate profiles. This helps prevent reprocessing profiles that show up again in LinkedIn’s ranked search results.
If your Automation offers Watcher mode, enable it when your input list changes over time. Watcher mode is a good fit for workflows where new profiles get added continuously, like job changes, new followers, or new commenters.
Enable duplicate removal by default, and use Watcher mode when you want the Automation to focus on new additions since the last run.
Step 3: Schedule weekly runs and keep activity patterns steady
Set the Automation to run repeatedly on a weekly schedule, for example every Monday morning. This gives you predictable lead flow without manual exports.
If you run multiple LinkedIn Automations, avoid launching them all at the same time. Spread them across the day so your activity looks like a normal work pattern, not a single burst.
You can also set a batch size. If you leave the “rows to process per launch” setting blank, you can create long runs, timeouts, or unusual activity spikes that are avoidable.
Note: When you scale, set a steady batch size, such as 100 to 500 rows per run, depending on the Automation and your overall LinkedIn activity.
“Avoid slide and spike patterns. Gradual ramps outperform sudden jumps.”
– PhantomBuster Product Expert, Brian Moran
Three ways to extract only this week’s new leads
Method A: Google Sheets filtering: recommended for most BDRs
Connect a Google Sheet as your output destination. PhantomBuster will append new rows to the same sheet each week, which becomes your cumulative master list.
Create a second tab called “This week’s leads.” Filter the master tab by the capture timestamp. If Column A contains a timestamp, you can use:
=FILTER(‘Sheet1′!A:Z,’Sheet1’!A:A>= TODAY()-7)
This second tab is filtered for weekly outreach, while the master tab stays intact for deduplication.
This approach is useful for teams because it’s visible, shareable, and easy to extend with columns like owner, status, segment, and last touch date.
Method B: JSON export: Useful for ad hoc snapshots
Open your Automation, go to the Console, then the Files or Results tab, and download the JSON output.
In many cases, JSON reflects the latest execution output, while a cumulative CSV reflects everything appended over time. If you need the latest run only, JSON is often the fastest path.
If you want to work in a spreadsheet tool, convert JSON to CSV with a converter you trust, then import it.
Method C: Zapier or Make routing: For hands-off delivery
If your output goes to Google Sheets, you can use Zapier or Make to trigger on a new row and push leads into the next step of your system.
Common routes include:
- Send a Slack message with name, company, and LinkedIn URL.
- Create a CRM task for research and personalization.
- Send the lead into an enrichment step before outreach.
This is useful when you want leads delivered into your team’s workflow without team members having to check spreadsheets.
| Method | Best for | Pros | Cons |
| Google Sheets filtering | Most BDR teams | Clear weekly view, easy to share and enrich | Requires a simple formula setup |
| JSON export | Ad hoc or manual workflows | Fast snapshot of a recent run | Manual steps and conversion work |
| Zapier or Make routing | Hands-off delivery | Leads flow directly into your process | Requires a Zapier or Make account |
What to avoid: Mistakes that break a weekly net-new workflow
Deleting the master list or clearing rows
If you remove a profile from the master list, you remove the history your workflow uses for deduplication. On a future run, that profile can reappear as “new,” which creates duplicates.
Instead of deleting rows, keep the master list intact and add a status column, for example, “Disqualified” or “Already contacted.” Then filter those statuses out of your weekly view.
If you need to archive data, export a snapshot. Keep the active master destination unchanged so your Automation keeps continuity.
Renaming or switching the output destination between runs
Many Automations track “already processed” logic based on a specific output destination. If you rename a results file, switch to a new file, or change where results are saved, you can reset that continuity.
Instead, if you want an archive, duplicate the file or export a copy. Keep the original output destination consistent so deduplication keeps working.
Launching too many LinkedIn Automations at once
Running several LinkedIn automations simultaneously, or pushing unusually large batches in one launch, creates a pattern that doesn’t look like normal usage.
Schedule launches across the day and keep volumes steady. A staggered schedule is also easier to monitor, because you can spot which step caused a failure when something changes in the UI.
Safety note: The goal is not to “hide automation.” The goal is to run a workflow that stays consistent, moderate, and easy to operate.
“Stability beats speed when you’re building automation that lasts.”
– PhantomBuster Product Expert, Brian Moran
How this workflow protects your account and your pipeline
Why cumulative workflows support responsible automation
When you keep a master list and only work net-new leads each week, you reduce accidental repeat outreach. That protects your credibility and keeps your outreach history clean.
It also helps you run steady operations. Predictable schedules and stable batch sizes are easier to maintain than stop-start campaigns with sudden spikes.
Responsible automation is mostly workflow design. The more repeatable and controlled your process is, the fewer edge cases you create.
Why this compounds over time
Over a few months, your master list becomes a deduplicated view of your market. That makes it easier to segment, enrich, and coordinate outreach across the team.
You also get cleaner reporting. When duplicates are controlled at the source, meeting attribution and reply tracking become more accurate.
Conclusion
If you want a fresh lead list every week, don’t rebuild the dataset. Keep a cumulative master list, then filter for what’s new.
With PhantomBuster, that means keeping the same output destination, using deduplication, and creating a weekly “view layer” in Google Sheets or your downstream routing.
If you want to build this workflow in PhantomBuster, start with an Automation to export your search results, connect a Google Sheet as the output, and schedule a weekly run with a steady batch size. Try it by starting your 14-day free trial.
Frequently asked questions
What if you need to change search criteria mid-campaign?
You can change the search URL and keep the same output destination. The Automation can still dedupe against existing history, and it will add profiles that were not already captured.
If you need separate reporting or ownership by segment, create a separate Automation and a separate output destination for that campaign.
Can you use this weekly workflow with other Automations, like commenters or followers?
Yes. This approach works anywhere you have a list that changes over time, for example followers, commenters, event attendees, or group members.
When available, Watcher mode helps focus on new additions. Duplicate removal helps prevent reprocessing when the platform shows the same results again.
What if you deleted the master list by mistake?
If the master output is gone, you lose the history that supported deduplication. In practice, that often means rebuilding the master list from scratch.
Going forward, treat the master list as an append-only dataset. Use status columns and filtered views for weekly outreach, and keep a periodic backup export.
Why does LinkedIn keep showing the same people in search?
LinkedIn search is ranked and often capped. When you rerun the same URL, LinkedIn may surface the same profiles again near the top.
That’s normal ranking behavior. Your job is to let your extraction workflow dedupe repeats and segment your searches over time if you need broader coverage.
My Automation ran but no new rows appeared: What should you check first?
Start with the simple checks:
- No net-new profiles: Your search may not have new additions since last week.
- Manual parity: Run the same action manually to confirm LinkedIn behaves normally on your account.
- Execution issues: UI changes, session issues, or prompts can interrupt runs. Check the Automation logs for where it stopped.
When you troubleshoot this way, you fix the right problem faster, and you avoid reacting by increasing volume.