How the search criteria work

Job titles, boolean keywords, locations and ranges: what each filter on step 2 does and how to sharpen it.

Updated

The search criteria are what separates a list of 100 useful candidates from a list of 100 profiles you'll discard one by one. This is step 2 of the list wizard, and the same screen comes back when you hit “Add candidates” inside an existing list.

You arrive with the filters already filled in from the role description you wrote in step 1. Here's what each one does.

AI-prefilled job titles and keywords on step 2

Current Job Titles

Matches against the profile headline. Type Data Engineer and it pulls candidates whose headline is that or contains it: Senior Data Engineer, Data Engineer at Acme.

Titles show up as tags. To add another, type it and press Enter (or the Add button); to drop one, the × on the tag.

Below there's “Generate more”, which asks the AI for more variants of the title. It's capped at 3 uses per list.

Think about how people describe themselves, not what the role is called in your ATS. For a data role, Data Engineer, Analytics Engineer and Ingeniero de Datos bring in three different populations, which is why the AI usually proposes local-language variants too.

Keywords

Keywords are searched anywhere in the profile (About, experience, skills), not just the headline.

There are two views:

  • Builder: each word is a tag, and between tags you pick the operator from a dropdown (AND, OR).
  • Preview: the same search written out as text, so you can read the whole thing at once.

It takes AND, OR, NOT, parentheses () and quotes "" for exact phrases, up to 300 characters. The “Add NOT at the beginning of the search” checkbox negates the search from the start, and there's another “Generate more” here, also capped at 3 uses.

For example, for people who know ETL, SQL, Python and dbt:

"dbt" AND "SQL" AND "Python" AND "ETL"

Current Job Titles and Keywords are required one or the other: the screen says so under each field. You can search by title only, by keywords only, or by both.

Locations

Up to 20 locations. Type and pick cities from the search box. “Expand” (up to 5 uses) asks the AI to widen the area for you, handy when one city runs short of candidates.

It's the filter that trims the candidate universe the most, so it's the first one to loosen when a search returns too little.

Optional filters

Below that there are five more filters, all optional:

  • Schools (up to 10): specific universities or schools.
  • Industries (up to 20): the current company's sector.
  • Years of Experience: total years across all roles.
  • Years at Current Company: how long they've been where they are.
  • Company Headcount: size of their current company.

The last three are range-based: tick every range that works for you, not just one. Looking for 4 to 10 years of experience? Select every range inside that span.

Maximum Candidates and cost

A slider sets the cap on how many candidates the search returns (25 by default). It's a ceiling: if fewer candidates match you get fewer; if more match, you can extend the list later.

Right below, the Cost Estimation box spells out the price: each candidate extracted and AI-enriched costs 3 credits, so 25 candidates is up to 75 credits.

The maximum candidates control and the cost notice

More in how credits work.

How to combine them

  1. Review the titles the AI proposed and drop the ones that don't describe the role.
  2. Set the location: it's what trims the most.
  3. Keep only the must-haves in Keywords. Every AND narrows a lot.
  4. Touch the ranges only if the volume is still unmanageable.
  5. Ask for few candidates the first time, look at what comes back, then scale.

If the search returns nobody, see what to do when a search returns no results.

Related articles