The same search next month, priced as what changed
Save the filters and Cuvy runs the same search again on a schedule. Anyone you have already paid for is skipped, so a monthly sweep charges only for people who are new to it. Someone the search missed is tried again after a cooling-off window — thirty days by default — and that retry is free unless it works.
- People already paid for
- Skipped
- Retry window
- 30 days by default
- A retry that misses
- Free
- A retry that works
- One credit
50 addresses free every month · no card
The second run is priced as the difference
Say a search returned 400 people in January and 448 in February. Those 448 fall into three groups, and only one of them can cost anything.
- 296
- already paid for, and skipped
- 104
- missed in January, retried for nothing
- 48
- new to the filters since last month
The 74% found in January. A person costs one credit once, so February does not look at them again — not in this search, and not in anybody else's on the same workspace.
Their cooling-off window has passed, so they are tried again. If nothing comes back a second time, nothing is charged a second time.
A new job title, a new employer, a profile that did not exist. These are first lookups and are priced like any other.
So February bills for roughly thirty-six people rather than 448 — three quarters of the forty-eight new ones — plus whichever of the retries came good this time. The search did not get cheaper. It stopped charging you for the part that had not changed.
Four things a person can be to a saved search
Which one they are decides what the next run does about them.
- Settled
- A confirmed address was handed over and one credit was charged. Later runs list the person and do not look them up again. A single-source row is charged the same once, and is tried again after the cooling-off window in case a second source has appeared — that retry is free either way, because a person cannot be charged for twice.
- Cooling off
- A person the search missed, or found on one source, waits out a window before anyone tries again — thirty days unless the workspace has set its own. Runs inside the window pass over them without spending queue time.
- Retried, and missed again
- Costs nothing, appears on no invoice, and starts the window over. The only way a retry becomes a credit is by finding an address that was not there before.
- New to the filters
- Somebody who has changed title, changed employer or only just appeared on LinkedIn. They have no history with this workspace, so they are priced as a first lookup.
Three decisions, made once
- Filters, not people
Save the filters after a run
When a search has finished, save it. What is stored is the filter set, not the people it happened to return that day — which is why the next run can find anybody new.
- Or run it by hand
Choose how often it should run itself
Pick the cadence when you save it and change it whenever you like. A run started by hand does not disturb the schedule.
- Skipped, retried, new
Read the summary when it finishes
Every run says how many people it skipped, how many it retried, how many were new to the search and how many credits that came to. The new people are its own list, ready to export.
What lands is the delta, not the search again
The whole point is that February's file is February's people.
Each run leaves its own list. Export it and you get the people that run found, which is what you wanted to send to — rather than the standing list again, with last month's rows in it, to be diffed by hand in a spreadsheet before anybody can be contacted.
The summary sits above it and reconciles to the invoice: what was on the page, what was skipped, what was retried, what was new, what was found and what it cost. Six numbers, and the last one is the only one that reaches billing.
What a saved search will not do
It re-runs filters, so it inherits whatever LinkedIn decides those filters mean that week. If the result set moves, the sweep moves with it, and a person who has quietly dropped out of the filters is not reported as missing — they are just not on the page any more.
It also does not watch individuals. Somebody who leaves the company you found them at will simply stop matching, and finding that out is a different job: job-change monitoring follows the people on your lists rather than the search that produced them.
What people ask about re-running a search
How often can a saved search run?
You pick the cadence when you save it, and most searches end up on a weekly or monthly sweep. You can also run one by hand at any point without disturbing the schedule. A run that turns up nothing new costs nothing at all.
Will a scheduled run spend credits while I am not watching it?
Only on people who are new to the search, and only as far as the balance goes. There is no overage, so a sweep that reaches the end of your credits stops there and says so rather than billing your card. The run summary lists what it used, and you get it whether the run cost anything or not.
Why would you retry somebody the search has already missed?
Because a miss is a statement about a particular day. Sources appear: somebody unfindable in January turns up in March when their employer publishes a team page or their address is seen somewhere new. Retrying costs nothing unless it works, so the only thing a failed retry uses is a place in the queue.
Can I change the thirty-day retry window?
Yes, and it is set per workspace rather than per search. Shorten it and misses come round again sooner, which is free but puts them back in the queue more often. Lengthen it and a stubborn list stops being re-read every month. Thirty days is a default, not a rule.
What happens if I edit the filters on a saved search?
The saved search is the filters, so editing them changes what the next run reads. People who match now and did not before are new to that search and are priced as first lookups; people who have dropped out of the filters are simply not read. Nobody you have already paid for is charged again either way.
Is a saved search the same thing as job-change monitoring?
No. A saved search watches a set of filters and tells you who is newly inside them. Job-change monitoring watches people already on your lists and tells you when one of them moves. Teams that run both use the first to grow a list and the second to keep it true.
50 addresses free every month · no card