How to Switch From Time To Pet: Export, Dry Run, Cutover

To switch from Time To Pet, export your clients and pets as CSV files, clean a copy of each file, run the import as a dry run, check the counts and balances against the old system, then cut over and keep the old account as a reference. By the end of this post you will have a 12-point checklist you can tick off, plus a rule for when to switch now and when to wait. How long each step takes depends on how much data you have, so plan around your own row counts, not a fixed number of minutes.

Export clients and pets separately and keep the originals untouched

Time To Pet lets you download this yourself. In the Client List, use the bottom left menu to pick "Export Client Data" or "Export Pet Data", choose the fields, then click "Export Data" and a CSV file saves to your device. Time To Pet's help article lists the fields and the options for Inactive Clients, Deceased Pets, Inactive Pets and Client Fields.

Do both exports. A clients file without a pets file gives you owners with no animals, and you end up typing pets back in by hand.

Decide on purpose about inactive clients and deceased pets. Include them if you want the history in the new system. Leave them out and that history stays behind in the old account.

Then copy both files and work only on the copies. The originals are your backup if anything goes wrong.

Fix five things in the copy before any import

Most messy imports come from the file, not the software. Open each copy and check these:

• Encoding. Save as UTF-8. Salesforce's import guidance notes that files saved in another character set, such as Latin-1 or Windows-1252, can fail or corrupt text. Accents and apostrophes in client and pet names are the usual victims.
• Headers. One header row, no duplicate header names, no blank header cells. Two columns with the same name is how a phone number lands in the wrong field.
• Dates. Convert to YYYY-MM-DD. Dromo's CSV guide points out that "2026-03-31" is far less ambiguous than "31/03/26". A swapped day and month moves a vaccine due date to the wrong month.
• Duplicates. Merge or remove the same client entered twice. Vero's import FAQ suggests doing this in your spreadsheet before importing, for example with Remove Duplicates.
• Empty rows. Delete them, including hidden ones at the bottom of the sheet. Skedulo's data migration guide recommends checking the file in a plain text editor and reading the result file for row-level errors if an import fails.

Count rows before and after, because eyeballing a screen misses gaps

Run the import as a dry run first. Then compare numbers, not impressions.

The formula: rows in your cleaned file, minus the header row, equals the clients (or pets) you expect to see after the dry run. If the numbers differ, find the gap before anything goes live.

Example: say your clients file has 241 lines. Take away the header row and you expect 240 clients. If the dry run shows 236, you have four rows to find, and the result log tells you which ones. Fix those rows in your copy and run the dry run again.

After the counts match, open 5 to 10 real clients and compare them line by line against the old account: address, phone, every pet, vaccine dates, notes. Pick your oddest clients, not your cleanest.

If the mapping looks wrong, stop and fix it. Do not cut over with half-mapped columns. In Porchlight's Switch Kit, column mapping is detected for you, the import runs as a dry run, and one-click rollback undoes it, but the counts and the sample check are still yours to do with any tool.

Plan for what will not move on its own

Time To Pet's own import help page says it cannot import credit card data (for security reasons) or scheduled events, because events are structured differently in its system. Take that as a warning for any move involving cards and recurring visits, and confirm both in your dry run instead of assuming.

Make a short list of what must be confirmed or re-created by hand, using the list from Collar's switching checklist: vaccines, forms, notes and documents, open bookings and memberships, invoices, balances, deposits and payment history.

Two of these cost real money when missed. Open balances and deposits that do not carry over mean a client is not billed or is billed twice. Total your open invoices and deposits in the old account, total them again in the new one, and the difference must be zero before you cut over.

The switch checklist: 12 checks from export to cutover

Copy this into a note and tick it off.

• Freeze changes to rates, forms and message templates for the migration week.
• Export clients and pets from the Client List as CSV.
• Decide whether inactive clients and deceased pets come along.
• Save untouched originals, then work on copies.
• Clean the copies: UTF-8, one header row, ISO dates, no duplicates, no empty rows.
• Run the import as a dry run.
• Match row counts for clients and pets against your cleaned files.
• Compare 5 to 10 real clients line by line, including your messiest ones.
• Confirm recurring visits, card details and open bookings, and re-create what did not come across.
• Check open balances and deposits: old total minus new total equals zero.
• Walk one real day of visits through the new system, including a sitter in a dead zone, and read the reminder and message templates.
• Train the office and the sitters by role, then cut over.

When to switch now, when to wait, when to stay

Time To Pet is well reviewed, and a switch only makes sense if the reason is real. The usual reason is cost that rises with every sitter you hire. If that is you, what per-seat pricing really costs shows the math with your own headcount.

Overlap for a week, but never edit in both systems

A parallel run is not about running two systems forever. It earns you the confidence to commit, after which the old system is reference only, as Collar's checklist puts it. Enter new bookings in one place only. Pick that place on cutover day and tell the whole team.

Check reminders and messages during that week. Every send in Porchlight is logged with a delivery record and failures retry, which is why the notification that never arrived is worth reading before you rely on any tool to tell clients about a visit.

What a bad import costs, and the fix

If a step looks wrong after cutover, roll back to the state before the import, fix the source file and run it again. Because you kept the originals, nothing is lost.

Your next ten minutes

Open the Client List in Time To Pet, export clients and pets, save the originals in a folder, and count the lines in each file. Write the two numbers down. Those are the numbers you will check after the dry run.

Try the dry run on your own data

Porchlight's Switch Kit takes Time To Pet exports and runs them as a dry run with detected column mapping and one-click rollback, and every table you hold with Porchlight can be exported as CSV and JSON at any time. Pricing is flat, with unlimited staff, so the sixth sitter does not raise the bill. Start free for 30 days and see how your own files come across before you commit to anything.