Fetcher Stopped Getting Replies Oct 16. Re-find the Warm Ones by Nov 19

Fetcher shut its platform on Oct 16, and whatever is still in your account gets deleted after Nov 19. That is also the day Juicebox's migration offer ends. If your team requested an export, you now hold a CSV of past sequences and replies, and it gets less accurate every week. Treat it as a list of people who may have moved, not as a pipeline you can paste into the next tool.

What happened in the Fetcher shutdown, and what does Nov 19 mean?

Juicebox bought Fetcher's intellectual property and other assets, not the operating business or your account. After Nov 19, Fetcher says remaining account data will be securely deleted and not transferred outside Fetcher. Your data only reaches Juicebox if you sign up and upload it yourself.

The timeline, in order:

• Oct 2, 2026: Fetcher co-founder and CEO Chris Calmeyn emails customers with next steps and the Juicebox offer.
• Oct 6: Public announcement. Juicebox (CEO David Paffenholz) agrees to acquire Fetcher's assets.
• Oct 13 to Oct 16: The export-request cutoff. Some sources give Oct 13 as the safe date. HeroHunt reports Fetcher would fulfill requests received by Oct 16.
• Oct 16: The platform shuts.
• Nov 19: Fetcher email support and the Juicebox offer both end. Remaining data is deleted.

Fetcher's FAQ is plain about the Juicebox Fetcher migration: candidate data moves only "if you become a Juicebox customer and choose to upload it yourself." Juicebox has the money to keep buying. It raised an $80 million Series B at an $850 million valuation in March 2026, and Axios reports it will stay acquisitive in active recruiting. None of that brings your history across for you.

Which Fetcher data export do you actually have?

Fetcher offered three export routes, and only the full account export, requested by email, is a complete copy. Before you sort anything, work out which file you are holding, because each one is missing something different.

If you only pulled per-position CSVs, you have a stack of separate files that almost certainly overlap. If you pulled the dashboard, you have performance statistics and no candidates. Open your actual file and check which columns are there. I couldn't verify whether the full export includes reply text, so don't assume it does. If it doesn't, your inbox and your ATS notes are the record of who said yes.

Why did the warmest candidates go dark first?

Fetcher says replies to messages sent through its own email service will not be received after the Oct 16 shutdown. So anyone who answered a Fetcher sequence after that date replied into nothing. Their "interested" may be sitting in their own sent folder and nowhere in your export.

Think about who those people are. A prospect who was still mid-sequence when the platform shut, and who finally answered a later step, did it after the pipe was cut. Those are high-intent rows, and they are the ones most likely to be missing.

Work the file in this order:

1. Replies already in the export. Anyone marked interested, or anyone who replied at all. Write to them from your own mailbox this week.
2. Prospects who were mid-sequence on Oct 16. They may have answered and gotten silence. Send a short, human note: the tool you used was retired, you may have missed their reply, is the conversation still worth having.
3. Everyone else. Cold rows. They can wait until after deduplication, because nobody is waiting on you.

The second group matters most. A candidate who said yes and heard nothing has a worse impression of you than one you never contacted. A two-line apology from a real person fixes that. A new automated sequence from a new tool makes it worse.

How many people in your export have changed jobs?

Assume roughly one in five has moved within a year. The 2025 BLS JOLTS quits rate averaged 2.0% a month, which comes to about 24% a year from voluntary exits alone, and total separations ran 3.3% a month.

24%

Median private-sector tenure is 3.9 years, but that median hides the problem. The people you sourced were not a random sample. You picked them because they looked movable, and movable people move. So in a file of sequences from 12 months ago, a real share of the title and company fields are wrong.

The mechanism that hurts most is email. When someone changes jobs, their old work email bounces, or worse, it keeps delivering to a mailbox they no longer read. Your new tool will log it as "sent" and you'll read the silence as a no.

So re-find people by name plus their old employer, not by email. A name, a title and a past company is a search you can resolve. A bounced address is not. Fetcher's plans capped volume at 500 to 1,000 human-sourced candidates a year, so a typical team export runs from a few hundred to a few thousand rows. That's small enough to re-verify one person at a time. Done by hand, though, it's a week of tabs.

That is the gap Refolk closes. You describe the person the way your export does - name, old employer and role - and get back where they work now, with a profile URL you can keep.

How do you stop old sequences from re-firing in a new tool?

Merge and deduplicate before you upload anything, and load your suppression list first. Revoking Fetcher's access stops Fetcher from sending. It does nothing to stop your next tool from emailing the same people again.

Fetcher disconnected integrations on its side at shutdown, but it asks customers to revoke access themselves. Do that. Then deal with the bigger risk, which is the per-position export structure. If you load ten position files into a new tool one by one, a backend engineer who sat on three of your searches gets enrolled in three new sequences in the same week.

The order of operations:

1. Merge every per-position CSV into one file. Keep a column for which positions each person appeared in.
2. Key each row on a stable profile URL, not on email. Emails change when people change jobs. Profile URLs mostly don't.
3. Build a suppression list. Anyone who said "not interested," unsubscribed, or opted out. Load this into the new tool before any active sequence exists.
4. Revoke Fetcher's mailbox access and any integration it still holds, including your ATS.
5. Only then enroll, starting with the warm and mid-sequence groups from your triage, written by hand.

Step 3 is where teams get burned. Someone who opted out of a Fetcher sequence earlier this year doesn't care that you switched vendors. To them it's the same company ignoring the same request.

Revoking Fetcher's access stops Fetcher. It does not stop the next tool from re-contacting everyone who already said no.

Should you plan around Fetcher's 5,000 recruiters?

Don't plan around Fetcher's 5,000-recruiter figure: it is a lifetime total, so size the migration from your own row count instead. Fetcher's announcement cites more than 1,500 customers and over 5,000 recruiters since 2016, while Juicebox's own blog describes the customer base as "collectively representing hundreds of recruiters."

Both can be true. The active base at shutdown was far smaller than the lifetime count. Vendor headline numbers count everyone who ever logged in, so they tell you nothing about your own migration. Here's how those figures sit next to the size of the profession in Refolk's index of professional profiles:

Fetcher's lifetime recruiter count equals about 22% of the US recruiter and sourcer pool. That's a rough sense of scale only, since Fetcher's users were global and not all technical recruiters. The US-to-UK ratio is about 27x, but treat the UK number as low: British teams often use titles like "Talent Acquisition Partner" or "Resourcer." In the US, top current employers of these titles include K2 Partnering Solutions, Experis, Snowflake, Google and Blue Origin.

The number that matters is in your merged file. The harder problem is the open roles those people were sourced for. If a search was live on Oct 16, its pipeline is now part stale and part missing replies, and rebuilding a shortlist from scratch while hiring managers wait is the slowest part of the whole migration.

How do you keep candidate history out of a single vendor's hands?

Make an ATS or CRM your system of record, and have every sourcing tool feed into it. Fetcher's help center said sourced candidates were "yours to use in perpetuity." In practice that only held for teams that had already pushed candidates into their ATS.

That's the real lesson of this recruiting tool shutdown. Sequences, replies and notes all lived where the outreach happened, because that was convenient. When the vendor went away, the history went with it, and the "perpetuity" promise covered only what had already left the building.

It also hits harder than you'd expect, because so few people own sourcing as a job. In Refolk's index, only 2,101 people in the US hold sourcer-specific titles (Technical Sourcer, Talent Sourcer, Sourcer, Sourcing Recruiter), at companies like Rippling, MongoDB, Verkada and American Express.

9.3%

When nobody's job is to curate the pipeline, the tool becomes the memory. And tools get acquired. Prime Intellect's purchase of Moondream came with a hard cloud shutdown date of Nov 13. Dayforce agreed to buy Ribbon AI. Expect more of these.

What to demand from Fetcher alternatives

Juicebox is the obvious landing spot, with reference customers like Cursor, Lovable, Ramp and Notion. HeroHunt and Pin publish their own comparisons. Whichever tool you pick, judge it on exit as much as entry:

• One-file export of every candidate, sequence and reply, self-serve, any day.
• Stable identifiers (profile URLs) on every row, so dedup survives a migration.
• Two-way ATS sync, so suppression lists and replies land in your system of record, not just the vendor's.
• Suppression that travels, meaning opt-outs export as their own list.

FAQ

Does my Fetcher data transfer to Juicebox automatically?

No. Juicebox bought Fetcher's IP and other assets, not customer accounts. Fetcher's FAQ says candidate data moves to Juicebox only if you become a Juicebox customer and upload it yourself. Anything left in Fetcher after Nov 19 is deleted and not transferred outside Fetcher.

What if I never requested a full account export?

Use what you have. Per-position CSVs still hold your candidate lists, and your own mailbox and ATS hold whatever replies reached you. Sources disagree on the request cutoff (Oct 13 versus Oct 16). Fetcher email support runs through Nov 19, so ask now rather than assume.

Can I just load my export into a new tool and re-send?

Not safely. Per-position files overlap, so the same person can end up in several new sequences at once. Around one in five people may have changed jobs within a year, so many work emails will bounce or go unread. Merge, dedupe on profile URL, load opt-outs as a suppression list, and contact warm replies by hand first.

What is the best Fetcher alternative?

The one that makes leaving easy. Look for full self-serve export, stable profile identifiers, two-way ATS sync and suppression lists that export on their own. Then keep your ATS or CRM as the system of record so the next shutdown costs you a tool, not your history.