What is data freshness?
Data freshness is how recently a record was confirmed against its source. Refresh cadence is how often a data provider re-checks it. Both matter in B2B data because people change jobs, titles and numbers, so a correct field decays.
Also called: refresh cadence · data recency · data decay
Data freshness in practice
Freshness is a property of each field, not of a database. A person's education rarely changes; their employer and title change every few years; a work email dies the day they leave. So the question to ask a provider is not "how fresh is your data" but "when was this field last confirmed, and how do I know".
Refresh cadence describes the provider's process. A data feed is re-delivered on a schedule; an API serves whatever the provider holds at the moment of the call. The two are different promises, and a buyer should read the cadence in the agreement rather than in marketing copy.
How freshness is kept and measured
- Timestamp per fieldRecord when each field was last observed or confirmed. A single "updated at" for the whole record hides which part is stale.
- Re-check on a cadenceThe provider re-reads sources on a schedule and re-delivers (feeds) or serves the current value (API).
- Detect changeA new employer, title or location is a change event. Feeds can carry these as deltas so you process only what moved.
- Measure decayRe-run an old sample and count the fields that changed. That is your decay rate for that field, on your data.
Data freshness vs refresh cadence
| Data freshness | Refresh cadence | |
|---|---|---|
| What it describes | A record or field: when it was last confirmed | A provider's process: how often it re-checks and re-delivers |
| Where to find it | Timestamps on the record, change events in a feed | The licence agreement or feed documentation |
| How to test it | Re-run an old sample and count changes | Compare two deliveries and count what moved |
Limits and when not to use it
- A refresh cadence is a promise about process, not a guarantee that every field is current on the day you read it.
- Fresh is not the same as correct. A value confirmed yesterday against a bad source is fresh and wrong.
- Forager does not publish a refresh figure on this page. Refresh terms for the feeds are stated in the licence agreement.
How data freshness works with Forager
Forager data feeds are re-delivered on a schedule agreed in the licence, to S3, Google Cloud Storage or Snowflake, and the API serves the current record at the time of the call. Change-driven use cases, such as detecting a CRM contact who changed jobs, run on the people feed.
- Recipe: detect CRM contacts who changed jobs
- People data feed
- Docs: data feed overview
- Docs: data feed FAQ
- $5 free credits
- No credit card
- Run your sample
Questions about data freshness
What does data freshness mean?
How recently a record or field was confirmed against its source. It is best tracked per field, because employer and title change far more often than education or location.
What is B2B data decay?
The rate at which stored B2B fields stop being true: people change jobs, numbers and emails. Measure it by re-running an old sample and counting the fields that changed.
What is a refresh cadence?
How often a provider re-checks its sources and re-delivers the data. For a feed it is a delivery schedule; for an API it is how often the underlying record is re-confirmed.
How often does Forager refresh its data?
The refresh terms for data feeds are stated in the licence agreement rather than on this page. Ask on the call and we will walk through the schedule for the feed you need.
Can I get only the records that changed?
Change-driven workflows such as job-change detection run on the people feed, where you compare deliveries and act on the records that moved.
Written answers are not available right now. The matching pages above still help, or book a demo and ask the team.
- $5 free credits
- No credit card
- Run your sample
Keep reading
How-to guides
Related terms
Sources
- Forager docs: data feed overview (read Nov 1, 2026)
- Forager docs: data feed FAQ (read Nov 1, 2026)