Skip to main content

POS migration

You have had the same POS since 2014. Here is what changing it really involves.

The system in your back office has run every service for a decade. It is slow, the support call costs more each year, and nobody wants to be the one who breaks it.

That fear is rational. It is also the reason these projects go badly, because waiting until the server dies means migrating in a panic instead of on a plan.

Tell us what is happening on your floor →

The short answer

How do you move off an old POS like Aloha or Micros?

You extract while the old system is still healthy. Menu, modifiers, employees and gift card balances come off a back office server rather than a cloud dashboard, and a lot of what matters was never data at all: routing, comp rules, floor plan. Document it, rebuild it, have your team run the test-order script on the new system, then cut over early in the week. Eight to twelve weeks is realistic.


Why waiting hurts

The instinct with an old system is to leave it alone until it forces the issue. That instinct is backwards, and it is worth understanding why.

On a legacy on-premise setup, your data lives on a back office machine in a cupboard, usually a Windows server of a certain age, with terminals talking to it over the local network. Everything you would need for a migration is on that box. The menu. The modifier structure. Employee records and permissions. Gift card balances. The routing rules. When that box fails on a Friday in December, you are not doing a migration. You are doing a reconstruction, from memory, during your busiest fortnight, and the gift card liability becomes an argument with customers instead of a spreadsheet.

The whole of the advice in this section is one sentence. Export while the legacy system is healthy. Even if you are not switching this year, having a current menu export, an employee list and a gift card balance report sitting somewhere that is not the cupboard converts a catastrophe into an inconvenience. That costs you nothing and we will happily tell you how to do it without hiring anybody.

The fear is not the new system. It is the twenty years of decisions living inside the old one that nobody wrote down.

Not in the export

The item list is the easy part and it is not the part that takes the time.

A decade-old system holds an enormous amount of accumulated operational judgment that was never stored as data in the first place: which printer gets which modifier, which staff role can void after a check is closed and which cannot, how the happy hour price rule interacts with a comp, what happens to a coursed order when a table is split, which button on screen three the closing manager presses every night without knowing what it does. None of that exports. All of it matters.

So the first fortnight of this project is not technical. It is somebody sitting with your closing manager and your longest-serving server and writing down how the place runs, then getting your agreement on which of those decisions are worth carrying forward and which were workarounds for a limitation the new system does not have. That second category is usually larger than owners expect, and finding it is most of the value in doing this deliberately.

The other thing frequently trapped in a legacy setup is your processing arrangement. Older systems often came bundled with an acquiring relationship, and the rate you are on was set a long time ago. Work out your effective rate honestly, total card fees divided by total card sales across three months from statements, before you assume the new system is the expensive part.

Gift cards and loyalty

This is the one that keeps owners awake, and on a legacy system it is genuinely harder than on a modern one, because the balances often sit in a local database or with a separate gift card processor that the POS merely talks to.

Step one is finding out which. If your gift cards run through a third-party program, the balances may not be trapped at all and the program may be able to continue alongside the new system. If they live inside the POS, you need a balance report off that back office machine, exported to a file, before anything is switched off. Get that report early and get it more than once, because it is the one artifact you genuinely cannot reconstruct.

Then the honest part about where the balances land. Neither Square nor Toast will simply ingest another system’s gift cards. On Square, the route is to contact them with a CSV of previously issued cards and balances, which they load so you can look cards up by entering the number, with specific formatting rules and a clear statement that non-Square gift cards cannot be activated or reloaded. On Toast, balances can be adjusted one card at a time in Toast Web, and their documentation states that bulk updates cannot be done there at all, so a batch goes in through Toast Customer Care from a spreadsheet. Toast’s loyalty import, separately, wants a point balance plus an email or a ten-digit phone number and takes roughly ten to fifteen business days.

What this means for a place with twenty years of cards in circulation: some of them will be honored manually, from a printed report, at the counter, for a long time. That is not a failure of the migration. It is the actual shape of the problem, and the only bad version is the one where nobody warned you. Plan the script your staff will use, decide whether you are replacing old cards with new ones at the register, and tell your regulars before it happens rather than after.

Loyalty is usually simpler on a legacy system, because a lot of them never had a real program, or ran one on paper. If yours is a punch card in a wallet, this is an excellent moment to build something proper, and it is a better use of the money than trying to carry a broken scheme forward.

Choosing the target

We work with both and our fee is the same either way, which is why our answer is worth reading.

Roughly: if you run a dining room with coursing, seat-level check splitting and a kitchen coordinating fire times across stations, Toast is built around that shape of service. If you are counter service, a cafe, a bar, a bakery or a shop, Square is usually the better fit and does not tie you to a term. If you are principally retail with an online catalog, the conversation involves Shopify rather than either. Clover is a reasonable system and worth knowing that its hardware is locked to the processor that sells it, which matters more to your next decision than to this one.

The thing not to do is choose based on which salesperson called you back. The thing to do is write down the five moments in your week that go badly on the current system, then have each candidate show you those five moments. If nobody will demonstrate them on your actual menu, that is information.

And there is a real cultural change alongside the technical one. A modern cloud system means your staff learn a new screen, your reporting looks different, and your monthly cost moves from a maintenance contract to a subscription plus a processing rate. Some of that is better. Not all of it is cheaper. We will lay out both sides for your situation before you sign anything with anyone, including us.

Reasons to stay

An old system is not automatically a bad one, and we have talked owners out of this move more than once.

If your legacy setup is stable, the hardware is healthy, support is still available, and the only complaint is that it looks dated, replacing it is an expensive way to buy a nicer screen. The things that most often hurt are online ordering going through a marketplace that takes a cut and keeps the customer, a Google listing with the wrong hours, and having no way to reach the people who already come in. All three of those are fixable without touching the point of sale at all, usually faster and for far less.

What genuinely justifies the move: support for your version is ending or already gone, the hardware is failing or unreplaceable, you cannot add online ordering without a mess of bolt-ons, your processing rate is locked into an old bundle, or you have a second location coming and running two disconnected systems is not viable.

If you are unsure which list you are on, a POS health check is a remote review of your settings and workflows ending in a prioritized action list, and it costs a fraction of a migration. Some of them end with us saying the system is fine, fix these six things instead. That reflects our belief in honest value: an investment should make sense for your business.

How the weeks go

Eight to twelve weeks for a single location, longer than a modern platform-to-platform move, and most of the extra is documentation rather than build.

Weeks one to three: extract from the back office while it is healthy, and write down everything that only exists in the building. The menu and modifiers. Employees and permissions. Gift card balances, more than once. Routing, comp and discount rules, day parts. Interview the closing manager. This is the phase people want to skip and it is the phase that decides the outcome.

Weeks four to seven: build on the new platform, deliberately, deciding which legacy workarounds to carry and which to drop. Then validation against the source, and set up whatever data feeds your accountant currently depends on, because those exist and nobody remembers them until they break.

Then the test-order script: your team rings a real service on the new system from the quick-start guide while the old one is still the record. On a legacy migration this is not optional, because your staff have a decade of muscle memory pointed at a different screen and the test run is where that surfaces safely instead of at seven on a Friday.

Launch early in the week after our team’s review. The first week on this kind of move is busy, because ten years of edge cases arrive one at a time, and the support team in your portal works through them from the setup record. And leave the old system powered on and readable for a while afterward. There is no prize for unplugging it on day one, and there is a real cost to having done so when a question comes up in March.

Side by side

Scroll sideways on a phone.

Legacy system to a modern POS: what comes forward
Feature WhatComes forward?What it takes
Menu and prices Yes, if extracted earlyPulled off the back office server while it is healthy, then mapped
Modifier structure RebuiltDocumented off the old system, then designed rather than copied
Employees and permissions As source dataRoles are re-scoped deliberately. Most legacy permission sets have drifted
Kitchen routing and floor plan NoNever stored as portable data. Written down, then rebuilt
Comp, discount and pricing rules NoDocumented with your closing manager. Half of them turn out to be workarounds you can drop
Gift card balances Only as a reportBalance report off the old system, then a support-side load. Old cards are honored, not reloaded
Sales history NoExported to files. Leave the old system readable rather than unplugging it on day one
Processing arrangement RenegotiatedOld bundled acquiring rates are frequently the real cost. Price your effective rate first
Hardware NoLegacy terminals do not run cloud POS software. The back office server goes away with them

What the migration includes

Stated in numbers, so nobody has to guess where it ends.

The same bounded scope as any migration and launch, with the limits in plain numbers. Legacy extraction is more work than a modern export, so we scope it honestly rather than discovering it later. The price is on the pricing page.

01

Full configuration on Square or Toast

One location. Up to 100 items, 20 modifier groups and 100 modifier choices, categories, taxes and service settings, up to 5 staff roles, up to 2 standard integrations, and online ordering. On Toast, that includes menu hierarchy, ordering channels, prep and routing settings and modifier logic.

02

Legacy extraction and documentation

Menu, modifiers, employees and balances pulled off the back office system while it is still healthy, plus a written record of the routing, permissions and pricing rules that never existed as data.

03

Clean-source data mapping

The extracted data mapped field by field before anything is loaded. Clean source means a legible export. If the item list carries a decade of retired specials, cleaning it is separate work and we raise it before we start.

04

Migration and validation

Loaded, then checked back against the source: item counts, price spot-checks, tax by category, and modifier and routing behavior on the twenty things you sell most.

05

Launch checklist and test-order script

A written checklist you keep, and a test-order script your team runs, ringing a real service on the new system while the legacy system is still the record. On a legacy move this is where a decade of muscle memory shows up safely.

06

Launch review and portal support

Our team reviews the configuration against the checklist before it goes live, early in the week, never into a Friday night. Afterward, the support team in your portal works from the setup record, and on this kind of move that earns its keep.

07

Quick-start guides, walkthroughs and a setup record

One for the floor, one for whoever opens and closes. The configuration is documented and the document is yours, which is the thing your last system never gave you.

Anything past these limits is quoted before it starts, never after. Every published price is on one page.

Be honest with me

Do not do this now if:

  • Your legacy system is stable, supported, and the only complaint is that it looks dated. That is an expensive way to buy a nicer screen, and the money does more for you spent on ordering, listings and reaching customers you already have.
  • The thing costing you money is online ordering going through a marketplace that takes a cut and keeps the customer. That is fixable in weeks without touching the point of sale, and it usually pays back faster than a migration ever will.
  • You want it done before a holiday season. Eight to twelve weeks is honest for this, most of it documentation, and compressing it means going live with a decade of undocumented edge cases still undocumented.
  • Nobody can give us time with your closing manager and your longest-serving staff. Everything that matters and did not export lives in their heads. Without those conversations we would be rebuilding a guess.
  • You have significant gift card liability and no balance report you can produce this week. Get the report first. It is the one artifact that cannot be reconstructed, and everything else waits behind it.

Questions we get

What if the old server dies before we migrate?

Then the project becomes a reconstruction from memory during whatever week it happens, and your gift card liability becomes a conversation with customers rather than a spreadsheet. This is why we tell people to export while the system is healthy even if they are not switching this year. Take a menu export, an employee list and a gift card balance report off the box and store them elsewhere.

Will we lose twenty years of sales history?

As live reports, yes. As data, only if you leave it too late. We extract what the system will give up before anything is switched off and hand you the files, and we recommend leaving the old machine powered on and readable for a while after launch. There is no prize for unplugging it on day one and there is a real cost to having done so.

Should we go to Square or Toast?

Square, in almost every case we see: it now handles coursing, seat-level splitting and a coordinating kitchen on the Premium plan, publishes its pricing, and does not tie you to a term. Write down the five moments in your week that go badly now, and make each candidate demonstrate those five on your own menu.

How disruptive is the actual cutover?

One shift, if the test-order script was run. Your team will be slower for two or three services because a decade of muscle memory is pointed at a different screen, which is normal and worth staffing for. We launch early in the week, and the support team in your portal works from the setup record through the first week, when the accumulated edge cases arrive.

Can we keep our gift cards working?

Honored, generally. Reloaded, generally not. Square states that non-Square gift cards cannot be activated or reloaded, and Toast has no bulk gift card update in its back office at all, so balances go in through their support team from a spreadsheet. Plan the script your staff will use at the counter, decide whether you are swapping old cards for new, and tell regulars before it happens.

Is there a version of this we can do in stages?

Sometimes, and it is worth asking. Online ordering, your listing, and reaching your existing customers can all be fixed now, independently, while the old system keeps running service. If your legacy setup is stable, that sequence often delivers more in the first quarter than a migration would, and it leaves the bigger project for a calmer month.


Take the export this week, decide later.

Even if you are not switching, get a menu export, an employee list and a gift card balance report off that back office machine and store them somewhere else. Tell us what system it is and we will tell you how, at no cost, whether or not you ever hire us.

Platform pricing and features change; where a platform does not publish a number we say so rather than guessing.

Shop & Table connects with supported commerce platforms. Square, Toast and Clover are trademarks of their respective owners.

Your preferences

Privacy choices

Choose which optional services can run on this website. Everything optional stays off until you choose. You can change your settings at any time.

Advertising tags are not configured on this website. These settings apply to this browser on this website; platform and ordering accounts have their own choices.

Make yourself comfortable

Accessibility

Choose what makes this page comfortable. Preferences stay on this device.

Navigate this page

Keyboard shortcuts

On a Mac, Alt is the Option key. Shortcuts do not run while you’re typing in a field, or with Control or Command held.

Alt + Shift + A
Open accessibility tools
Alt + Shift + S
Navigate by page structure
Alt + Shift + ?
Show keyboard shortcuts
Alt + Shift + T
Text size: next size
Alt + Shift + C
Contrast: next level
Alt + Shift + F
Focus outline on or off
Alt + Shift + L
Highlight links on or off
Alt + Shift + H
Highlight headings on or off
Alt + Shift + M
Stop animation on or off
Alt + Shift + G
Reading guide on or off
Alt + Shift + R
Reset everything

Policies & Choices

Open Page →