i think that it would be rather unusual to update timestamps in whois (which i guessed the parent poster was referring to) based on updates to in-zone data. A transfer would be an example of something that would change information in whois and thus update the timestamp noted there for the latest update. it is sometime possible to infer the date of the latest change of in-zone data because the serial of the zone is often constructed by using a date and a counter. But that is actually just convention and not reliable. Its also unlikely the parent poster was referring to this.