Bringing your old shop across

There are two doors. A Shopmonkey shop can connect to Shopmonkey with an API key and let Torqix read the account. Every other system — and Shopmonkey exports that are not available through the manual connection — comes in as CSV. Both doors stage a preview before writing and match source identities so a repeat does not quietly duplicate the same records.

Import is in the second column beside the rail, in the customer band. You need to be able to open the customer file to use it, which owners, managers, advisors and front desk can.

The direct Shopmonkey connection

1. 1

   Step 1: Paste the key

   Under Bring it straight from Shopmonkey, paste Your Shopmonkey API key. The one-off connection uses it for this pull and does not store it. Pick customers first, import them, then return for vehicles so every car has an owner to find.

2. 2

   Step 2: Queue or preview

   Queue or preview reads the account but writes no shop record. The result page reports what arrived, what Shopmonkey claimed existed, what Torqix will add or match, and every refusal. Commit only after those totals agree.

3. 3

   Step 3: Keep it in step only if an owner chooses to

   The owner-only Keep Shopmonkey in sync until cutover card can seal a credential and run daily or weekly. It is off by default. The owner chooses customers, vehicles and order history separately, then ticks Enable recurring re-import. Those future runs commit through the same identity and refusal rules without another person pressing Import, so turn it off and rotate the Shopmonkey token at cutover.

How a CSV migration goes

1. 1

   Step 1: Drop your exports

   Under Bring everything at once, pick every CSV you have and press Start the migration. Twenty megabytes per file. Do not open and re-save them in Excel first — it mangles ZIP codes, dates and long part numbers.

2. 2

   Step 2: Read the preview

   Four numbers: Rows, Will add, Already there, Skipped. Underneath, the columns it matched, the first few rows as they will land, and every row it cannot take with the reason spelled out. Nothing has been written at this point.

3. 3

   Step 3: Fix the column matching if it guessed wrong

   Open Wrong guess? Fix the column matching, set the ones it missed, and press Re-run the preview with these columns. Your corrections are remembered and applied to the next file automatically.

4. 4

   Step 4: Import

   The button counts what it is about to do — Import 412 customers — and the next file's preview runs once this one lands.

You never have to name your columns up front. There are no required headers for a migration file; Torqix reads the header row and matches it against the names these systems actually use. A column it does not recognise is left out and listed, not guessed at.

What comes across

Customers

Names, companies, every phone number and email on the row with its label kept, addresses and notes.

Vehicles

VIN, year, make, model, plate, colour, mileage, attached to their owner.

Service history

Past visits, as a read-only record of what the other system said happened, shown on the vehicle with a label saying where it came from.

Parts inventory

SKU, part number, description, brand, bin, cost, price and an opening count.

For CSV, that is the list. The direct Shopmonkey connection also brings appointments, open estimates, closed order, invoice and recorded-payment evidence, and customer-visible message history. Imported messages are marked as Shopmonkey history, arrive read, never enter a send queue, and never create texting consent. Remote photo links are counted with their reason but are not fetched from arbitrary hosts. Before ending your Shopmonkey access, download the original arrival photos there and attach the ones you need to keep to the matching repair order in Torqix. Live invoices, inspections and internal Shopmonkey notes remain outside this path; still-open invoices are not resumed as live Torqix invoices.

Imported history is deliberately kept apart from your live numbers. It is shown on the car and it is searchable by its old order number, but it is not fed into totals, never recalculated, and never turned into a repair order. Fifteen years of somebody else's arithmetic is testimony about what happened, not a ledger to run a shop from.

Duplicates, and the cases it refuses to guess

Customers are matched on phone number, then email. A match merges into the record you already have and leaves it exactly as it was — an import never overwrites live data.

• Two people on one landline are a household, not a duplicate. Same number, clearly different name, and you get two records with a note saying why.
• A phone number that turns up on five or more rows is treated as a placeholder and never used to match anybody.
• A vehicle whose owner is ambiguous is refused rather than attached to a guess: more than one customer named "J Lopez" — attach by hand. Putting somebody else's service history on a stranger's car is the one mistake a switching shop would not forgive.
• A part with a SKU you already have merges, and the live count and price win over the spreadsheet.

For history, the preview shows a link rate — how many visits found a car, and how many found a customer. That is the number to check before importing. Low means you should import customers and vehicles first, then preview the history file again.

Taking it back

A CSV import of customers, vehicles, parts or read-only service history can be undone from its own page: Undo this import, then Yes, delete the imported customers. A whole multi-file migration can go back in one go, newest file first. The undo also deletes the CSV you uploaded from our database — a shop that took its import back should not leave its customer list behind.

Anything you have already put to work is kept and reported. A customer who now has a repair order, a car with a visit on it, a part that has moved — those stay, and the result line tells you how many.

Direct Shopmonkey order history is different. It lands as financial evidence with source totals and line reconciliation attached, and Undo this import refuses it. Deleting a historical ledger because the connection was later turned off would rewrite what the old system said happened. Review the preview and its discrepancy counts before that order run commits.

Changing a lot of things at once

The same screen does bulk edits, and it is the easiest way to reprice a parts list. Download the list with Download it as a spreadsheet, edit it, bring it back, and you get a diff — old value beside new — before anything changes.

Two rules that catch everybody. An empty cell means leave this alone; it does not clear anything. To genuinely empty a field you type (clear) in the cell and tick a separate box confirming how many values you are emptying. And leave the Torqix ID column alone — it is how a row finds its record. Clear it and that line becomes a brand-new record instead.