I've been doing this long enough to remember when upgrading was obvious. You'd hold a phone from two years ago next to the new one and the difference practically slapped you in the face. Faster, sharper, a camera that could actually see in the dark. Those days are mostly gone, and a new piece over on AOL makes the case that the big tech companies' every-two-years persuasion routine is wearing thin for regular people.

Can't say I'm surprised. I remember when 1GB of storage was unthinkable, when we all marveled that an entire music collection could fit in a pocket. Now a mid-range phone has more computing power than the desktop I paid a small fortune for in 1998, and most of us use it to scroll and check the weather.

So when should you actually upgrade? Let me give you my honest Aunt Elizabeth guide, built from decades of being the family tech support line.

Upgrade when something is broken or abandoned. If your phone's battery dies by lunch, or the manufacturer has stopped shipping security updates, that's not vanity, that's self-defense. My rule of thumb: once a device stops getting security patches, it's time to start planning your exit. Everything else is negotiable.

Upgrade when a new device solves a problem you actually have. Notice I said a problem you have, not a problem a keynote told you about. If your knees are complaining and a wearable finally tracks the thing your doctor keeps asking about, wonderful. If your smart lock means your granddaughter Jess can let herself in when I'm stuck on a call, that's technology working for me instead of the other way around, and it's the only test that matters.

Upgrade when the math genuinely works. Trade-in values, carrier credits, preorder gift cards, sometimes the stars align and the new device costs almost nothing out of pocket. I told folks back in July that if they wanted a Galaxy Flip, the preorder window was the moment. I ran the numbers, and the numbers said go.

And when should you not bother? More often than you think.

Don't upgrade for a camera that's 8 percent better in lab conditions. Don't upgrade because the new one comes in a color you like. And please, don't upgrade because a phone is suddenly feeling "old." Slow can often be fixed with a battery replacement and a storage cleanup, which costs a fraction of a new device. My daughter's phone was "dying" last spring; one new battery later it runs like a champ. She felt a little sheepish. I didn't gloat. Much.

The phone makers, understandably, aren't thrilled about any of this. Their business model grew up in an era when each generation leapfrogged the last, and now the leaps have become shuffles. So they lean on financing plans and trade-in programs to keep the conveyor belt moving. There's nothing sinister about it, it's just a company doing what companies do. But you don't have to march to that drum just because it's playing.

Here's a question worth sitting with: does your current device do the things you need it to do, reliably, every day? If the answer is yes, you have your answer. If it's no, then figure out exactly which need is going unmet and buy for that reason, not for the sparkle.

My colleague Marcus could explain to you exactly why year-over-year chip improvements have flattened out, and if you're curious about the under-the-hood story, his deep-dives are worth your time. I just look at the shelf result: the device you bought three years ago is still, in most cases, a perfectly good device.

Keep your money in your wallet until something earns it. Useful wins every single time, and the newest thing is not always the useful thing.