Data Freshness

Formula

Freshness, currency, timeliness, latency, cadence

• Currency (currentness): whether a value is of the right age for its use.
• Timeliness: whether data reflects the period it represents and is available when needed.
• Latency: the delay between an event and its appearance in data. Filing rules build latency in: a Form 13F is filed after the quarter it reports has ended.
• Refresh cadence: how often a source is re-checked. A process setting, not a measure of freshness; a frequent refresh of a source that has not changed confirms only that the source still says the same thing.
• Last verified: the date of the most recent deliberate recheck of a claim, one of the evidence dates.

US analytic standards (ICD 203) list the age and continued currency of information among the factors that affect source quality.

Why record-level and page dates mislead

A record or page timestamp changes for many reasons unrelated to the facts: a new field, a template change, a typo fix. Freshness computed from it overstates how current the facts are. Measured per claim, from the last observation, the same record shows which fields are recent and which are old. A summary across many claims is better shown as a distribution (how many claims fall in each age band) than as an average, which hides the old ones.

How fast private-markets facts change

• Fast: a person's current role and contact details, fundraising status, current allocation targets and mandates.
• Slow: legal name, domicile, registered address, stated strategy, entity hierarchy.
• Static: vintage year, founding date, the date and terms of a past transaction. Dated facts like these do not go stale.

Some sources refresh only on a statutory filing cycle, which caps how fresh their content can be; the cycles for Form ADV, Form 13F and Form D are set out on those forms' pages. Fund NAVs arrive after their reference date (see NAV reference date); a value carried forward unchanged is a stale mark.

Keeping data fresh

Two approaches are combined. Scheduled rechecks revisit sources at a cadence set per attribute: fast-changing fields more often. Triggered rechecks follow change detection: when successive captures of a source differ, the affected claims are rechecked. Change detection brackets when a change happened (after the earlier capture, on or before the later one); it does not give the effective date. A recheck that finds the source unavailable moves no dates. Refreshing without verifying adds noise: a re-ingested value that was wrong is still wrong, now with a recent timestamp.

How Altss applies this (Altss methodology)

In Altss methodology, freshness is a property of a claim, never of a page or a record. The age of evidence is the assessment date minus the last observed date, and a claim read as current is fresh if that age is within the review horizon of its attribute's volatility class (STATIC, SLOW or FAST; Altss-defined). Past the horizon, the claim's confidence for the current reading is lowered one step and it is flagged for recheck, a loss of support that Altss calls Evidence Decay, an Altss-defined concept; the dated claim itself ("X held role Y as of D") keeps its support. Freshness is not a validation status: a claim can be fresh and UNVERIFIED, or RESEARCH_VALIDATED long ago and now past its horizon, and passing the horizon changes neither the claim's evidence origin, its derivation status nor its validation status and date. A content edit never implies that a fact was verified, and no population-wide refresh interval is implied. See the Temporal Data & Freshness Methodology and the Source Reliability & Confidence Methodology.

Worked example

Examples are illustrative; figures are not market data.

Not the same as

• Data Quality Dimensions: Data quality dimensions include accuracy, whether a value matches reality; freshness is how old the evidence for a value is. Stale values are more likely to be wrong, but not necessarily wrong.
• Evidence Dates: Evidence dates include the last-verified date of the latest deliberate recheck; freshness is the judgment made from the age of the latest evidence.

Common mistakes

• Treating freshness as accuracy.
• Measuring freshness from a record's or page's last-modified date.
• Applying one refresh cadence to every field.
• Treating an old, dated claim as wrong; it may still be a correct statement about its date.
• Averaging ages across fields, which hides the stale ones.
• Counting a re-ingestion of unchanged data as a fresh observation of the fact.

Edge cases

• A source re-captured today that has not itself changed since 2023: the observation is recent, but the source's own as-of date is old. Both ages matter.
• Static, dated facts have no review horizon and never become stale.
• A scheduled fact (an appointment effective next quarter) is not current before its effective date, however fresh the evidence.
• If a recheck fails because the source is unavailable, the last observed date does not move.

Questions

Sources

1. The Government Data Quality Framework. Government Data Quality Hub, UK Government (GOV.UK), Published 3 December 2020. Status: Current (checked 2026-10-01). Data quality dimensions - how to measure your data quality (Timeliness) — supports: Timeliness as the degree to which data reflects its period and is up to date
2. DAMA-DMBOK: Data Management Body of Knowledge, 2nd Edition Revised. DAMA International, Technics Publications, 2nd edition revised (maintenance release), 2024. Status: Current; DMBOK 3.0 project under way since 2025 (checked 2026-10-01). Data Quality chapter (2024 revision notes) — supports: Currency added as a recognised data quality dimension
3. Intelligence Community Directive 203: Analytic Standards. Office of the Director of National Intelligence, ODNI, Signed 2 January 2015; technical amendments incl. 21 January 2022. Status: In force (as amended) (checked 2026-10-01). Section D.6.e(1) — supports: Age and continued currency of information as a source-quality factor
4. 17 CFR 240.13f-1 - Reporting by institutional investment managers of information with respect to accounts over which they exercise investment discretion (Form 13F). U.S. Securities and Exchange Commission (CFR text via eCFR; LII mirror), eCFR current as of 2026-09-29; no substantive amendment since eCFR baseline. Status: in force (checked 2026-10-01). 17 CFR 240.13f-1(a)(1) — supports: Form 13F reports holdings as of each calendar quarter end and is filed after that quarter ends
5. Form ADV (Uniform Application for Investment Adviser Registration and Report by Exempt Reporting Advisers) - SEC form cover and index. U.S. Securities and Exchange Commission, OMB No. 3235-0049; expires 2027-07-31. Status: in force (checked 2026-10-01). Form index page, 'Federal Information Law and Requirements' — supports: Form ADV is updated on a statutory filing cycle
6. 17 CFR 239.500 - Form D, notice of sales of securities under Regulation D and section 4(a)(5) of the Securities Act of 1933. U.S. Securities and Exchange Commission (CFR text via eCFR; LII mirror), eCFR current as of 2026-09-29; last amended 2017-05-23. Status: in force (checked 2026-10-01). 17 CFR 239.500(a)(3)(iii) — supports: Form D amendments follow a statutory filing cycle

• Evidence Dates
• Data Quality Dimensions
• Temporal Validity
• Confidence Score
• Point-in-Time Data

• Bitemporal Data