How to keep candidate data up to date
Candidate records are true on the day the CV is parsed and drift from then on. How to keep them current with a routine: the LinkedIn tab when you open a record, job change alerts for the people who matter, and the Data Agent across the whole database.
You keep candidate data up to date by checking it against the candidate on a routine, instead of waiting for someone to notice it is wrong. In practice that means three habits: every recruiter refreshes the records they open from the candidate's live LinkedIn profile, the people who matter most to your desk are watched for job moves, and the whole database is cleaned and refreshed on a schedule. This guide covers how each one works, what it costs, and the part no software can do for you.
Why does candidate data go out of date so quickly?
Candidate data goes out of date because the candidate keeps living and nobody tells your CRM. A record is accurate on the day the CV is parsed. After that, the person is promoted, changes employer, moves city, gets a new phone and stops using an old email, and every one of those changes happens on LinkedIn or in their inbox, never in your database.
Recruiters do update records, but only the ones they are working on this week. A database of 20,000 candidates has a few hundred in active play at any time. The rest are only as current as the last time someone happened to open them, which for many is the day they were added.
The fields do not all age at the same speed. Contact details and job titles go first. Skills and work history drift more slowly. Availability, salary and notice period can change in a week and only a conversation will tell you.
What does out-of-date candidate data cost a recruitment agency?
Out-of-date data costs you placements you never knew were possible, and time spent on ones that are not. Each kind of stale field fails in its own way:
- Wrong job titles make search miss people. A candidate saved as a Senior Developer three years ago may be an Engineering Manager now. Search for engineering managers and they do not appear, so the best fit in your own database never reaches the shortlist.
- Old employers send you to the wrong conversation. You pitch a role that is a step down from where they already are, or you ring a switchboard that has never heard of them.
- Dead email addresses damage every email you send. Mail providers watch how many of your messages bounce, and a list full of abandoned addresses pushes your campaigns, and eventually your one-to-one emails, towards the spam folder.
- Missed moves are missed business. A candidate you placed who has just joined a new company is a friendly contact inside a possible new client, and the seat they left is a vacancy. You only win either if you hear about the move first.
- Accuracy is a legal duty. Under UK and EU GDPR, Article 5(1)(d) says personal data must be accurate and, where necessary, kept up to date. A candidate who asks what you hold about them will see every out-of-date field.
The quiet cost is trust. When recruiters learn that the CRM is often wrong, they stop searching it and go back to LinkedIn for every role, and the database you paid to build stops paying you back.
Which candidate fields go stale first, and how do you refresh each one?
Job title, employer and email go stale first, because they change every time someone moves job. The table shows how each field changes, how you find out, and where it is refreshed in Recruitly.
| Field | Why it changes | How you notice | How to refresh it in Recruitly |
|---|---|---|---|
| Job title | Promotions, new roles, LinkedIn headline edits | Their LinkedIn profile, a call | LinkedIn tab › Update record, or the Data Agent's LinkedIn refresh |
| Current employer | People move every few years | LinkedIn, a bounced work email | Same as job title, plus job change alerts for the people you placed |
| Work history | Every move adds a line | The Data Agent replaces it from the live profile; the Chrome extension refreshes it on an existing record | |
| Location | Relocation, hybrid work | LinkedIn, a call | Data Agent location fill: blanks only, from the CV, LinkedIn or the company's office |
| Work emails die with the job; personal ones get abandoned | Bounces | Email list clean-up in a Loop Channel, Enrich with Recruitly for a new one | |
| Phone | New numbers, numbers saved without a country code | A wrong number | Enrich with Recruitly, and the phone fixer on Data Health |
| Skills | Learned on the job, rarely added to a CV you hold | LinkedIn, the conversation | LinkedIn refresh adds new skills and keeps the ones you have |
| Availability, salary, notice | Change with every life event | Only by talking to them | A recruiter's call, logged on the record |
How do I find out which candidate records are out of date?
Open Data Health for candidates. It is a dashboard for super users that scores the whole database and shows, in one view, how many records have not been touched for how long and which important fields are missing.
The stale view groups every record by the date of its last activity: the last 30 days, 31 to 90 days, 91 to 180 days, 180 to 365 days, and over a year. Activity means anything your team did with the record, such as a call, an email or a note, so the over-a-year group is the part of your database nobody has looked at since that date. Records that have never had any activity are counted too.
Beside it, Missing Data Analysis counts records without an email, a phone or a LinkedIn profile, and the quality score rates the database from Poor through Fair (40 and above) and Good (60 and above) to Excellent (80 and above). Each tile opens the Data Agent already aimed at those records, so the record count you are looking at becomes the job you run.
Read the stale figure for what it is. An untouched record is not always a wrong one, but it is one nobody has checked, and the older the group, the more of it will have changed. That is the group to sweep first.
How do I update a single candidate record from LinkedIn?
Open the record's LinkedIn tab and press Update record. Recruitly shows the candidate's current LinkedIn profile inside the record, highlights each section where LinkedIn differs from what you hold, and opens a side-by-side comparison of "On record" and "From LinkedIn". You tick what to copy across, Recruitly merges it and logs a note saying what changed.
- Open the candidate and choose the LinkedIn tab. Recruitly uses the LinkedIn link on the record, or finds one in the CV.
- If there is no link, the LinkedIn Profile Finder suggests likely profiles, marked "Looks right", "Could be them" or "Probably not". One click saves the right one. If you hold an email, Find LinkedIn from email can look the profile up instead.
- Read the profile. Sections that differ from the record have their Update button highlighted.
- Press Update record, tick the fields to copy, and confirm.
Nothing is copied without your tick, so a recruiter's own notes and corrections are never replaced by surprise. A profile costs half a cent the first time anyone in the agency reads it, colleagues see the stored copy free after that, each user has 100 free every month, and profiles in your own network are free when you have connected your account through LinkedIn Connect. Refetch profile refreshes it, at most once a week.
If you spend your day on LinkedIn rather than in the CRM, the Recruitly Chrome extension does the same job from the other side. On any profile it says whether the person is already in Recruitly, and for an existing record it refreshes work history and education in one click.
How do I keep thousands of candidate records up to date without doing it by hand?
Use the Data Agent. It works through a saved search or a selection of records, refreshes each one from the candidate's live LinkedIn profile, cleans what is there and fills what is missing, and shows you what it will change before it changes anything.
- Start it from AI Hub › AI Data Agent, from a grid (tick records, then Actions › Enrich with Recruitly), or from a Data Health tile.
- Pick the records: a saved search such as "Candidates with no activity for 180 days", or everything.
- Tick the work. For keeping data current, the useful ones are Enrich from LinkedIn, the profile clean-up, and location fill.
- Press Preview. It runs on up to five records, writes nothing and charges nothing, and shows each proposed change with where it came from and how sure it is.
- Run it. A super user's own job starts at once; anyone else's waits for a super user to approve it in AI Hub. You get an email when it finishes, and any credit not used goes back to your balance.
The Data Agent treats each source the way a careful recruiter would. The LinkedIn refresh is the one step that replaces values, because the live profile is newer than your record: where LinkedIn shows something different, such as a new headline or a new job in the work history, the record takes LinkedIn's, and new skills and languages are added to the ones you already hold. Filling, from the CV or for the location, writes only to empty fields and never overwrites a value already on the record, while the tidy-ups you tick, such as casing and emoji removal, reformat existing values without changing what they mean. Contact details are never blanked by a clean-up, and the tags your team added are kept.
It needs a LinkedIn profile on the record to refresh it, and records without one are not charged for the LinkedIn step. If many of your candidates are missing a profile, run the Profile Finder on them first. There is more on how the agent decides what to change in Can you trust AI to update your CRM records?
Can the Data Agent keep candidate data up to date on a schedule?
Yes. A super user can set a Data Agent job on a saved search to repeat every day, every week or every month. The first run covers everything in the search up to that moment, and each later run, at 02:00 UTC the next day, on Monday or on the 1st, takes the records added since the run before.
A schedule therefore guarantees that every new candidate arrives clean and linked to their current profile, whoever added them and however. It does not re-check your whole database every week, which would mean paying to refresh the same people over and over. Re-checking older records is the quarterly sweep in the routine below: a job on a saved search of records with no activity for six months.
A schedule always runs as the super user who started it, emails only when a run fails, and stops when you stop it or cancel its first run.
How do I know when a candidate changes jobs?
Turn on job change alerts in a Loop Channel. Put the candidates you placed or shortlisted in a channel, press Put Simi to work, choose Job changes and pick monthly, weekly or daily. Simi checks each person's LinkedIn profile and posts a card in the channel when someone moves, showing where they were, where they are now and where you placed them, with a button to add an action, add the new company as a lead, or open a note Simi has drafted for you.
The alerts are free and each check is one LinkedIn lookup at half a cent, so 240 people checked monthly costs about $1.20 a month. The full walk-through is in how to get job change alerts for the candidates you placed.
How do I stop candidate email addresses going bad?
Turn on Email List Clean-up in the Loop Channel you send campaigns to. Simi checks every address in the channel monthly or weekly and moves the ones with no mail server, a domain that no longer exists, a disposable address or an invalid address to your Email Blacklist, so campaigns and drips skip them while your one-to-one emails still send. Each address checked costs $0.0015. See how to clean your email list automatically before every campaign.
When a candidate does need a new address or number, Enrich with Recruitly on the record finds a work email for $0.15, a personal email for $0.25 or a phone for $0.25, charged only when one is found. If your agency already pays for Apollo, Lusha or FullEnrich, connect it in the Marketplace and their buttons appear on the record too, spending your own subscription with them.
Do duplicate records make candidate data go out of date?
Yes, because two records for one person means a recruiter updates one and the other stays wrong. Recruitly checks every new candidate for the same email, mobile or name and keeps the Create button locked while a match stands. For duplicates already in the database, the Duplicate candidates system filter lists them and Merge combines them into the record you choose. It takes a few clicks, shown in how to find and merge duplicate candidates.
What is a good routine for keeping candidate data current?
A good routine splits the work by who is best placed to do it. Recruiters keep the records they touch current, automation watches the people who matter most and every new record, and a super user sweeps the rest on a calendar.
| When | What to do | Who | In Recruitly |
|---|---|---|---|
| Every record you open | Glance at the LinkedIn tab and copy across what changed | Every recruiter | LinkedIn tab › Update record |
| After every call | Log availability, salary, notice and anything they told you | The recruiter on the call | A note or the record's fields |
| Weekly | Check the people you placed and shortlisted for job moves | Automatic, a recruiter acts on the card | Job change alerts in a Loop Channel |
| Monthly | Check campaign lists for dead email addresses | Automatic | Email list clean-up in a Loop Channel |
| Monthly | Read the stale and missing-data figures and pick the next sweep | A super user | Data Health |
| Every day, week or month | Clean and refresh every record added since the last run | Automatic, started once by a super user | Data Agent schedule |
| Quarterly | Sweep records nobody has touched in six months or more | A super user approves it | Data Agent on a saved search |
Start with the bottom of the table if your database is already old. One sweep of everything untouched for a year, previewed first, resets the baseline. From then on the schedule keeps new records clean, alerts catch the moves that matter, and recruiters keep doing what they already do when they open a record, with Update record making it one click.
How much does it cost to keep candidate data up to date in Recruitly?
For most agencies it costs tens of dollars a month. Every part is priced per record or per check, shown before anything starts, and capped by a monthly limit your agency sets.
| What | Price | Free allowance |
|---|---|---|
| Live LinkedIn profile on the LinkedIn tab | $0.005 a person, once for the whole agency until someone refreshes it | 100 per user every month, and free through your own LinkedIn Connect |
| LinkedIn Profile Finder | $0.001 a lookup | 1,000 per user every month |
| Data Agent clean-up | $0.005 a record | $10 of Data Agent credit on first use, about 2,000 records |
| Data Agent LinkedIn refresh | $0.005 a record with a LinkedIn profile | |
| Job change alerts | $0.005 a check; the alerts are free | |
| Email list clean-up | $0.0015 an address | Addresses already blacklisted are skipped |
| Work email / personal email / phone found | $0.15 / $0.25 / $0.25, only when found | |
| Data Health | Free | Super users |
Data Health itself is free. Current prices for everything are on the pricing page.
What can software not keep up to date for you?
Software cannot tell you whether a candidate is ready to move, what they would move for, or what they told you in confidence. Those facts change with a new manager, a new baby or a bad quarter, they never appear on LinkedIn, and they are the ones that decide whether a placement happens.
That is why the routine above hands the machine every fact that can be checked against a public profile or a mail server. When titles, employers, emails and phones look after themselves, a recruiter's call is spent on the questions only a person can ask, and the answers go on the record while they are still true. The agencies that do this well are not the ones with the most records. They are the ones whose recruiters believe what the CRM tells them, and search it first.
How often should candidate data be updated?
Update it whenever a recruiter opens a record, weekly for the people you placed and shortlisted, and at least once a quarter for everything else. New records should be cleaned on the day they arrive, which a Data Agent schedule does for you.
Is it allowed to update candidate records from LinkedIn?
Most agencies do it under legitimate interest, and keeping data accurate is itself a GDPR duty. Say in your privacy notice that you refresh records from public professional profiles, and keep the note Recruitly logs each time a record is updated, so you can show where every change came from. Take your own legal advice on your agency's notice.
Will Recruitly overwrite what my recruiters typed?
Not without someone choosing it. On the LinkedIn tab nothing is copied until a recruiter ticks it. In the Data Agent, filling writes only to empty fields, the tidy-ups you tick, such as casing and emoji removal, reformat existing values without changing what they mean, the LinkedIn refresh changes a field only where the live profile shows something different, and Preview shows every change before the job runs.
Do I need LinkedIn Recruiter to keep candidate records current?
No. The LinkedIn tab, the Profile Finder and the Data Agent all work with no LinkedIn licence at all. A Recruiter or Sales Navigator seat only matters if you connect it through LinkedIn Connect, which also makes profiles in your own network free to read.
Who can see Data Health and run the Data Agent?
Data Health and the Data Agent in AI Hub are for super users. Any recruiter can send selected records to the Data Agent from a grid, and the job waits for a super user to approve it before anything runs or is charged.



