I had similar questions about why so many scales were all so inconsistent, so I did some research on the issue and the following is basically what I learned about why the issue is supposedly so complicated:
I’ve had that happen with several scales (changes slightly when moved a few inches on the floor). It’s usually indicative of a floor that is not 100% flat (has some bumps, bows, etc), or is not totally level, even if you can’t tell with your own vision, just by looking at it.
Most scales will also change slightly depending on exactly where you stand on them. If you stand a little more forward or backward or to one side or the other at all, the weight will slightly alter. So often when you get on, then get off and get on again, many scales will give you a very slightly different reading. This is fairly common.
In fact, when using different analog scales, you can often see how the needle will vibrate a little, and not be 100% static, especially as the balance of your weight shifts a little from your heel to toe and left foot to right foot. So even analog scales are often giving somewhat of an approximation. If Digital scales were honest, they would also show that their measurement of your weight is fluxuating constantly while you stand there, but most companies program them to pause the weight after a few seconds and tell you that’s what it is, so it isn’t so confusing. Lots of scales will even limit their readings to something like every 0.2lbs (instead of 0.1lbs to help reduce this, or I had scales that remembered the last weight it told me, and if I got on again in a short period of time, it would still tell me the exact same weight again even if I purposely added or dropped a little weight (clothes, other items that I know weigh at least 0.2-0.5lbs), just so they seem consistent. This is not uncommon for many scales I’ve tested to do to prevent people complaining about inconsistency.
Scale reliability is a difficult subject. Each company handles it differently, and so when you compare one scale to another when both have taken different approaches to address the constant fluctuating weight issue…well, it becomes really difficult to say which is right and which is wrong. Any scale that gives you a single static measurement within a few seconds is not telling you the absolute truth of what they are reading, because you can’t possibly hold perfectly still even if you stopped breathing and stopped your heartbeat for a couple of seconds, you are constantly slightly shifting to keep balance and those slight shifts show up on scales, they just lie to you about it and take the range of readings and give you their best guess.
At least, that’s how it was all explained to me as I read up on the issues. It made sense too. Still, I don’t personally claim to be any kind of scale engineer or expert though, and anyone can claim anything on the internet, so I fully leave room to learn that things work much differently than the above.
I guess the main point is, your other scale could also potentially be the one that is feeding you results that are not completely consistent with what is really happening. I have had scales purposely use an algorithm to force the scale to give results to be more consistent when a weight is close to a recent previous measurement, just to make sure people wouldn’t complain about the inconsistency that most every honest scale deals with. Personally, I’d rather have the scales be totally honest and show me the real time fluctuation with every shift of weight, but almost no digital scales do that for more than a second or two before pausing the result and telling you it’s “guess” based on the constant variations.