Switching systems without losing a season

You renew the software you have now every year around the same week, mutter the same complaint while you sign, and open the case the next morning anyway. Not because it’s good. Because the store you know how to run today is worth more than the store you might be running in three months if the changeover goes badly. That trade-off is rational. Most owners who are still on something they’ve outgrown didn’t fail to notice — they did the math on switching mid-stream and it came back negative. The real question isn’t whether to move. It’s how to move without three months on the calendar quietly turning into your worst quarter.

Never cut over in season

Pick the slowest stretch on your calendar and do it there, on purpose, with a date circled. Not the week before a big trade show. Not bridal season. Not the run-up to the holidays, when every register needs to be fast and every ticket needs to be right and you have no patience left over for a staff member squinting at a new screen while a line forms. A changeover in your slow month costs you a slow month. The same changeover in your busy quarter can cost you the quarter, because a new system’s rough edges and a full case at the same time is how a good decision turns into a bad memory.

What has to move exactly, and what doesn’t

Not everything in your old system deserves the trip. Years of superseded price changes, notes on a customer who hasn’t walked in since a piece was still in style, a vendor you stopped buying from — none of that needs to arrive intact. Summarize it, archive it somewhere you can still find it if a question ever comes up, and let the new system start clean.

What does need to arrive exactly, to the piece and the dollar, is anything still open: inventory on your case today, a stone still out on memo, a repair on a bench jeweler’s bench, a layaway three payments deep, a special order that hasn’t come in yet. An open record isn’t history. It’s a promise you already made to somebody, and it has to survive the move in one piece or you’re the one who broke it.

Inventory needs its real cost, not an average

Here’s where a changeover quietly costs you money for years afterward instead of just for one bad week. If your inventory arrives in the new system priced at an average — one blended number smoothed across everything you happen to own — instead of the real cost of each piece, every margin report you run from that day forward is wrong by a small amount that never corrects itself. This is orientation, not accounting advice — your own accountant should confirm how your store carries inventory cost — but the operational risk holds regardless: a diamond you paid $4,000 for and a similar-looking one you paid $5,500 for are not interchangeable once they’re sitting in the same case, and an average erases the difference the moment it’s created.

Say your case carries two hundred loose stones and mountings at a real total cost of $340,000. Averaged out, the system might tell you each piece cost $1,700. Sell the $4,000 stone and the report says you made money against $1,700 of cost instead of $4,000 — a healthy-looking margin that’s actually a loss you won’t see until you’ve sold enough of the underpriced side of the average to notice the case is worth less than the books say. That gap doesn’t announce itself. It shows up two or three years later as a shortfall nobody can trace back to a single bad afternoon, because it was never one afternoon — it was every sale since the day the average replaced the real number.

The records that cost you the most if they’re dropped

Four kinds of open record deserve more care than the rest, because none of them are only yours. A stone out on memo belongs to the dealer who trusted you with it — if the new system can’t tell you what’s out, with whom, and since when, you’re one phone call away from an argument you can’t win with a shrug. A repair on the bench is somebody’s ring, already promised back by a date you gave them out loud. A deposit or a layaway is money you already took, against a piece you haven’t handed over yet — if that balance doesn’t move with the record, you can end up asking a customer to pay twice for something she’s already partly paid for, which is the fastest way to lose a customer who was already yours.

None of these are edge cases. On any given week, they’re a meaningful share of what’s actually happening in your store, even though none of them will show up if you only look at what’s sitting in the case. Count them before the move, and count them again after. If the numbers don’t match, you don’t flip the switch.

Run both systems until the numbers agree

The safest way to move is to run the old system and the new one side by side for a real stretch — long enough to close a few weeks the normal way twice. At the end of each week, the two have to agree: the same inventory count, the same open memo count and dollar value, the same open repairs, the same deposit and layaway balances, the same customer balances owed. Agreement isn’t a feeling. It’s a number you can put next to another number and see that they match, piece for piece.

Pick who on your staff has to be fluent in the new system before that date, not during the first week after it — the person everyone leans on at five o’clock on a Saturday when something looks wrong needs to already know the answer, not be learning it live in front of a customer. That’s usually not the owner. It’s whoever’s been doing the closing count for years and already knows what “off” looks like.

None of this needs to be dramatic. It needs to be boring, on a schedule, with two sets of numbers that either match or don’t.

One more question is worth asking before you ever sign with anyone again, this time or the next time: how do you get your data back out, in what format, and on what notice. A vendor who can answer that in one sentence is telling you something useful. A vendor who changes the subject is telling you something too.

The standard that actually protects a season is simple. The new system has to earn the cutover — matched counts, fluent staff, every open promise accounted for — before you touch anything. Until it has, the old one stays switched on, complaints and all.

Gemercel is, among other things, a system a store would be switching to — which means the whole argument above applies to us as well, including the exit question. Founding stores join a waitlist and are brought on by hand, one at a time, in the order they joined, in part so no changeover ever lands in anyone’s busy quarter. If your own calendar is the thing holding you back, get on the list and pick your own season.