Method
Same time of day
Repeat readings taken at the same time of day remove one variable you cannot otherwise account for; it costs nothing and tightens the spread.
Guides on Method: Bone-pressed length, start to finish, Stretched flaccid length, and why the clinics take it, Girth, done the way the studies do it, The measurement glossary
If you are taking more than one measurement session, take them at roughly the same time of day. It is a small rule, it costs nothing to follow, and it removes a source of spread that no amount of careful technique fixes on its own.
Why time of day matters at all
Several of the state variables that affect a reading - arousal baseline, energy, how recently you have been active, how relaxed the room is - drift across a day in ways that are hard to hold constant on demand. A session first thing in the morning and a session late at night are not drawing on the same starting state, even with identical technique in both. Metrology has a name for the ideal: the international vocabulary (JCGM VIM, entry 2.20) defines a repeatability condition as the same procedure, operator, measuring system, operating conditions and location, with replicates taken "over a short period of time". Sessions days apart can never fully meet that definition, so the job is to hold constant whatever you can. Fixing the time of day does not eliminate that drift, but it removes it as a moving part between sessions, which is the cheapest kind of control available: no equipment, no extra readings, just picking a slot and keeping it.
How this fits with repeats
This rule governs the gap between sessions on different days, not the handful of readings taken within one sitting. How many readings to take and when is a separate protocol question with its own answer; same-time-of-day is one condition that makes the readings from separate sessions worth comparing to each other in the first place, alongside things like keeping the room at a normal temperature rather than measuring straight out of a cold shower. Clinical studies control conditions the same way: Habous et al. (2015) took every measurement in a private room held at 21°C. None of this needs to be precise to the minute. Roughly the same part of the day, evening after evening or morning after morning, is enough to remove most of the effect.
What it does not fix
A fixed time of day gets you sessions that are comparable to each other and, with the rest of the method followed, comparable to the published data. It has nothing to do with how the result is perceived by another person, which is a different kind of question entirely - the sort of subjective read a human reviewer on a site like Rate Penis gives, not a state variable a protocol can control for. It is also irrelevant to a photograph: an image carries no timestamp information a viewer can use, and whatever an AI tool estimates from a picture is not drawing on session timing at all. A photo-based score, the kind of composite figure a site like Penis Rater builds, is answering a different question in the first place. And if a subjective number was what you actually wanted rather than a length in centimetres, that is what a site like Rate Cock provides - no protocol about clocks changes which of the two you are getting.