This page makes two points:
1) Photos of coins do not tell the size of the coin unless you already know what you are looking at.
2) You should pay attention to the actual size of a coin because two coins supposedly of the same "type" can differ in size in an important way.
Usually, larger is better. When a coin is larger than normal, it is more impressive in hand (but, maybe not more impressive in a photo) and worth more.
Look at these two coins, both Byzantine coins identified as "Sear 512".
Measurement shows the design is the same size. but the coin on the left is larger than normal.
Sometimes you can see that a coin is larger than normal because the die-diameter is visible on an over-large flan. The reverse of the coin on the left makes it clear the flan is much larger than necessary.
35 mm. (Much larger than a US 50-cent piece.) 11.58 grams.
Byzantine emperor Maurice, 582-602 AD, struck year 8 (589/90) at Nicomedia. Sear 512.
Yes, it is identified as "Sear 512", but that is not the whole story. Look at how large it is compared to the design.
Look how large it is compared to the coin on the right with the same Sear number. They are not from the same year, but the sizes of the designs are the same; it is only the flan size that makes the big difference.
Why is it so large? Byzantine coins are very often overstruck on previous coins. At 12:00 on the obverse "ANVS" from the undertype can be seen and at 3:30 on the reverse is the top of the large off-center "M" of the undertype. With that orientation "II" is clear on the lower right of the "M" and above it "X" is partially there. The size of the flan is correct for a coin of Justinian I (527-565) from year 22 (548/9) arranged:
but the flan, at 35 mm, is significantly larger than the beading (22 mm) and the die (25 mm). Of course, if you use an old coin as a flan your new coin will be the size of the old coin which is not necessarily the intended size of the new coin.