What Werkstattsystem costs

Compare what is in the figure, not the figure

Two monthly prices are only comparable once you know what each one contains. In this field the gap is not in the workshop functions — everybody has jobs, times and invoices. It is in the two things beside them.

• Website, online booking and the customer portal are in the plan here. Of the international suppliers we are compared with, Orderry and GaragePlug do not offer them at all. Elsewhere that means a second supplier, a second project and a second bill — and for a workshop without a website it is not a feature at all, it is a web agency you no longer need.
• E-invoicing is in the plan. Selling a standard format as a paid add-on is not a feature.

So the honest comparison is not three numbers side by side. It is three numbers plus, in each column, whatever you would still have to buy to reach the same result. That sum is almost never the one on the pricing page.

A price you cannot see is a price you cannot compare

Part of this field does not publish a price at all. GaragePlug asks you to book a call; Workshop Software shows no price on its home page and leads with a fourteen-day trial. Orderry does publish its plans — in US dollars, and expressly excluding sales tax.

None of that is dishonest. It is simply the part of the comparison that has to be done by hand, and the part where a monthly figure quietly stops meaning what you think it means. Five things are worth having in writing before a trial, not after:

• Which currency will I actually be billed in?
• Is tax included in the figure shown, or added to it?
• What does a second location cost?
• What does setup cost — and is it a one-off or a project?
• What does leaving cost, and what do I get out with me?

The last one is the one nobody asks while signing and everybody remembers while leaving. It is the same question as data ownership, only priced.

Per seat or per shop — the difference grows with your team

Our model is one shop, one price. The field it sits against mostly is not, and there is a checkable example: Orderry charges $6 to $19 per additional user and $49 to $179 per additional location on top of its plan, from its own price page.

The argument against per-seat billing is not that it is unfair. Suppliers have costs per user and may pass them on. The argument is what per-seat billing does to behaviour, and it is entirely predictable: a workshop that pays per login stops creating logins.

The one that gets skipped is never the owner's. It is the part-timer's, the apprentice's, the Saturday cover's — and those are exactly the accounts that would have told you who took the car in, who released the job and who changed the position. The saving is small and visible every month. What it costs is invisible until the day somebody asks who did what.

Which is why the price model belongs in the same conversation as personal logins rather than in a separate one about cost.

Which plan — and the honest way to find out

The plans above differ in team, stock and locations. They do not differ in whether the core works: website, online booking, work orders and invoicing are in all three. So the sizing question is not “how big am I” but “which of three thresholds am I about to cross”.

• A second person who needs their own login.
• Parts and tyre storage that has become stock to manage rather than a shelf to look at.
• A second location.

These pages are written for workshops of roughly two to twenty people. Below that, the smallest plan is not a stripped-down version — it is the whole product with one login, which is the point of not billing per head.

One piece of practical advice that saves more than it costs: if you expect to cross a threshold in the next twelve months, look at the plan you will be on then, not the one you are on now. Moving between plans is cheap. Moving between suppliers is not, and that is the comparison you are really making.

The arithmetic, without a single number from us

Our own figures are in the cards above and they are not repeated down here, because a price written twice is two prices. What this block gives you instead is the calculation — and it works with any supplier's price, including the one you already pay.

Convert the monthly figure into your own shop rate. A price per month is an abstraction; an hour of your own labour is not. Divide the one by the other and you have the only number that matters: how many billable hours a month the software has to give back before it has cost nothing. Do it for every offer on your desk and the comparison stops being about money and starts being about hours, which is the currency you actually manage.

Then go and find those hours. Three places they hide, in falling order of size:

• Transfer work. The same plate, the same customer, the same positions written more than once between intake and invoice. Count it on one finished job and multiply by your jobs per month. This is almost always the largest of the three and the hardest to see, because it is spread across the whole day in one-minute pieces that never feel worth measuring.
• Labour that was worked and never billed. Times written from memory in the evening come out low, and what that costs is not admin time — it is revenue. It shows up nowhere in a time saving and everywhere in the invoice total.
• Utilisation. The slowest lever and the biggest. An hour not spent searching for a part or a key is an hour that can be sold, at no visible price increase. The labour rate calculator puts your staff costs, overheads and productive hours together and shows what the rate has to be — and how far better utilisation moves it.

And the counterweight, because a payback calculation with nothing on the other side of it is a sales pitch: if your invoices already fall out of your job sheet without anyone copying figures across, the first of those three is zero for you and the payback is correspondingly slower. Software does not pay for itself by existing. It pays for itself by removing one specific piece of work you can name before you buy.

Bought, rented, hosted — where the cost really sits

Three shapes are on offer in this field: buy a licence and run it on your own machine, rent a licence and run it on your own machine, or rent it hosted so that somebody else runs it. Only in the third does the monthly figure cover the whole of what you need, and that is the entire reason the shapes are hard to compare.

What a self-run installation adds after the licence: a machine that has to be replaced, a backup that somebody has to test rather than merely switch on, updates that somebody has to apply, and the morning it does not start being your morning. None of that appears in a price comparison and all of it appears in your week.

Two traps worth naming:

• A one-off licence is not a one-off cost. The maintenance beside it is what keeps the updates coming, and updates are how a change in the rules reaches your invoice at all. A licence without maintenance ages into a program you may not stop using and cannot safely keep using.
• Setup is a real line, and it is not always shown. Workshop Software advertises “no costly setup fees” on its own home page — a sentence nobody prints unless the fee is common enough elsewhere to be worth denying. Ask for the number, and ask whether it is a fee or a project with days in it.

Ours is the hosted shape, and the honest description of it is a trade rather than a win. You install nothing, you replace no server, and the backup is not your job. In exchange, the update you did not ask for is also the update you cannot postpone, and where the machines physically stand becomes a question you have to ask instead of answer. Ask it — of every supplier, in writing — and see the main product page for why the answer belongs in a contract rather than in a headline.

Term, notice, and the clause about the price

The monthly figure is the part of a software contract everybody compares. These four are the part that decides what it costs you over five years, and all four fit on one sheet of paper:

• The term, and what happens at the end of it. Not just how long, but what it renews into. A one-year term that renews for another year is a different product from one that renews monthly.
• The notice period, and how notice is given. A month before the end and a quarter before the end are the same sentence with very different consequences, and the day you remember is usually the day after.
• The price-adjustment clause. How much, how often, with how much warning — and whether a rise gives you the right to leave. A term without a price lock is the version worth arguing about; a longer term that buys a fixed price is a trade you might well want.
• What leaving costs and what you take with you. Both halves. An exit that is free but hands you a stack of PDFs is not free.

One observation that helps you know which conversation you are in: Orderry publishes its plans; GaragePlug prices by a call. Where the price comes out of a conversation, so do the terms. That is not a reason to distrust anyone — it is a reason to ask for all four points in that same conversation, and to have them in writing before the trial ends rather than after.

The last point is the one people discover late: your duty to keep invoices does not end when the contract does. If nobody settles the export at signing, a retention duty quietly becomes a subscription you are not allowed to cancel. The invoicing page sets out how that happens.

What is actually negotiable

Less than you hope on the monthly figure, more than you expect on everything around it. The published per-unit price is usually the least movable thing in the whole offer, which is why asking for a discount is the weakest opening available to you.

These move more often:

• Term against price. If you commit for longer, ask what you get for it — and take a fixed price rather than a percentage off, because the fixed price is the one that is still worth something in year three.
• Onboarding and data import. The setup fee and the migration of your customers and vehicles are work, and work can be included, halved, or turned into a fixed price instead of days.
• The notice period. Shortening it costs the supplier nothing today and is worth a great deal to you on the one day it matters.
• A second location, if one is realistic within the term. Priced when you are still deciding, it is a negotiation; priced when you already depend on the system, it is a quote.
• Time. An extended trial, or a start date that begins when you actually go live rather than when you sign. Both are free for the supplier to give and directly reduce your risk.

The one to reach for first is not a discount at all: a shorter term, or a clean exit. What you are really buying at this stage is the option to have been wrong, and every month of term you accept is a month of that option sold back.

Two practical notes. Where a supplier publishes its plans, most of the price is already fixed and the predictability is itself worth something — you can compare without a meeting. Where a supplier prices by a call, nothing is fixed until the call, so the comparison costs you the meeting either way. And whatever is agreed, get it into the order document rather than the email thread: the person who promised it is not necessarily the person who will be there when you need it.

What free costs, and where the bill turns up

Free comes in three shapes and they behave completely differently, so the first thing to establish is which one you are being offered.

• A trial — the whole product with a clock on it. Nothing is limited except the time.
• A free tier — the product with a ceiling on it: a number of users, of vehicles, of invoices a month. Nothing expires; you simply grow into the wall.
• Free because the money is made elsewhere — on a transaction, on a bundled service, on something you have to buy alongside. That is not automatically a bad deal, but you should be able to say out loud which of them it is before you build your shop on it.

What each one costs, regardless:

Setting up is paid in your own hours, and a ceiling makes you pay it twice. Master data, templates, the way you name things — that work is the real investment, and it does not transfer when you outgrow the tier and move. Before you take a free plan, ask what the paid one costs, because that is the price you have actually chosen.

A user ceiling is the per-seat problem wearing a different hat. The login that does not get created is the apprentice's or the Saturday cover's, for exactly the reasons set out under per-seat billing above, and free makes the pressure worse rather than better because there is no invoice to argue with.

Ask whether the export is in the free tier or above it. One question, and it tells you what the plan is for. A free product you can leave is a genuine offer; a free product you cannot leave is an acquisition cost you paid in data.

And a trial is not free either, it is a fitting — so use it as one. A real week, real customers, real parts, and at least one real invoice you actually send. A trial run on demo data proves that the demo data works.

The cost of changing systems, and how to keep it small

Every comparison you make is really a comparison plus a move, and the move has a price that no pricing page shows. Three parts to it, only one of which is money:

• The overlap. A sensible switch runs both systems side by side for a few weeks, which is a few weeks of two subscriptions and, more expensively, of divided attention at the counter.
• The learning. Mostly not the software's fault — it is habits, and habits are slower than menus. Budget it in weeks of slightly slower work rather than in hours of training.
• The tail. The old data have to stay reachable long after the old contract ends, and whoever only notices that at the switch pays an old supplier for years purely to be able to look things up.

How to keep each of them small is a sequencing question rather than a money question, and it is set out on the main product page: master data first, then intake from a fixed date, billing last and on a month boundary. One step from that list belongs here as well, because it is the one that saves the most and is skipped the most often: pull the full export while the old system is still running and still paid for. It cannot be done later, and on the day of the switch everything else will look more urgent.

Then put the number where it belongs in the comparison. A switching cost is paid once. A per-seat premium, a second supplier for your website, an add-on for a standard invoice format — those are paid every month for as long as you stay. After enough months the one-off is the smaller figure, and the calculation that ignores it is the one that keeps people in a system they stopped choosing years ago.

Finally, the switching cost you can actually prevent is the next one. Test the export in your first month with the new supplier, while nobody is under pressure and nothing depends on the answer.