A dataset can say “latest” and still mean 2023

Appreciate the honesty here, most people only share the wins.

Solid lesson. Which channel has worked best for you so far?

Two dates is the right default and almost nobody does it. In SEO data the same confusion shows up constantly: a rank tracker shows "current position 4" but the crawl ran three days ago and the page moved to position 8 yesterday. Google Search Console itself reports with a 2-3 day lag, and people treat the latest available date as live.

We built UtilitySEO to scan sites and one design decision we got right early was showing "scanned at [timestamp]" next to every finding, not "current issues." A finding from a scan that ran before the fix deployed is not current — it is historical. Same principle as your period-date vs filing-date split.

The interpolation point bites hardest. A smooth line between two known data points feels like signal and reads like confidence, but it is a guess carrying a chart's authority.

I deal with a milder version of this with SEC 13F filings: fund holdings as of the end of a quarter, filed up to 45 days later, and people read them as "what fund X owns now". What helped:

1. Two dates, never one. "Positions as of June 30, filed Aug 14" instead of "updated Aug 14". The filing date is when you got the data; the period date is what the data describes. Readers conflate them unless you show both.
2. Say the lag in words next to the number the first time someone sees it ("this is 3 months old by design"), not in a footnote.
3. Never use "current" or "live" anywhere in the copy, even in headings. One stray "current holdings" undoes all the careful date labelling.

Your "leave missing years blank rather than interpolate" call is the right one too. A gap reads as honest; a smooth line reads as live.