Product

What happens to your jobs and leads if you switch booking or CRM tools

An AI-era checklist for moving your history, your lead records, and the prompts that depend on them without losing anything that matters.

Ohio AI Lab Staff · 3 min read
The quick hit
  • Inventory every place your job and lead data lives before touching the old tool
  • Export each record type as files you can open and read
  • Your AI prompts and instructions point at old data; rewrite them before the cutover
  • Keep the old account read-only for a while so nothing is lost in the gap

The moment is familiar. An owner sits in a scheduling or CRM tool they have complained about for a year, with the cursor hovering over the cancel button, and stays put. The reason is not loyalty. It is five years of job history, lead notes, and email threads that live inside that tool, plus the AI prompts and instructions that now depend on all of it. A careful checklist makes the switch safe. Here is that checklist, in order.

Why the stakes are higher now

Before AI, a tool switch risked losing records. Now it can also break the instructions you built on top of them. If you taught an AI assistant to pull job details from your CRM, or wrote prompts that assume your lead notes live in a particular place, moving tools without adjusting those instructions gives you wrong answers that look right. The work splits into three parts: audit what you have, export it, and rewrite the AI layer for the new home.

Step one: inventory what lives where

Start by listing every record type the old tool holds. For most owners that means customer records, job or appointment history, lead records with their notes, quotes or estimates, and any email or text threads stored inside the tool. Do not rely on memory for this. Open the tool and walk its menus, writing down what you actually see. Note where each record type lives and whether the tool lets you export it.

Include the AI layer in the inventory. If you keep prompts, custom instructions, or a saved business context file that describes how you work, list those too. They may live inside the tool, inside a chat account, or in a document. The point of this pass is to end with a single list of everything that must survive the move.

Step two: grab the exports

Every serious booking and CRM tool offers some form of export, usually to a spreadsheet or a file you download. Look in the settings or account area for words like export, download, or data. Export each record type separately rather than trusting one bulk file you cannot read. Open each export after downloading it and confirm the columns make sense and the rows are all there.

Check the date range on each export. Some tools limit exports to recent activity or require you to request a full archive. If a record type cannot be exported, write that down now rather than discovering it after the account closes. You may be able to copy that data by hand before the switch, or you may decide the loss is acceptable, but make that decision with eyes open.

Store the exported files somewhere the old tool cannot touch, such as your own computer or your own cloud storage. Keep copies. These files are your insurance policy for the whole move.

Step three: rewrite the AI instructions before cutover

AI instructions, meaning the prompts, custom settings, and context files that tell an assistant how your business works, are written against the world as it exists today. When your job history moves from one tool to another, those instructions stop matching reality. Rewrite them after the data lands in the new tool but before you rely on the AI day to day.

Go instruction by instruction. For each prompt or saved instruction, ask what data it references, whether that data now lives in the new tool, and what changed in the naming or structure. A prompt that says look at the lead notes may need to point at a different field name in the new tool. Test each rewritten instruction against a real record and check the answer by hand before trusting it.

The checklist, in order

Run the switch
  1. 01
    List every record type

    Open the old tool and write down each kind of record it holds, from customers to notes.

  2. 02
    Add the AI layer to the list

    Note every prompt, instruction, or context file that depends on the tool's data.

  3. 03
    Export each record type separately

    Use the tool's export function and download files you can open and read.

  4. 04
    Check every export

    Confirm the rows and columns are complete and note anything that will not export.

  5. 05
    Copy files out of the tool's reach

    Store exports on your own computer or storage, with backups.

  6. 06
    Move data into the new tool

    Import or enter the records, then spot-check a sample against the exports.

  7. 07
    Rewrite each AI instruction

    Update every prompt and context file to match the new tool, then test against real records.

  8. 08
    Keep the old account read-only

    Leave the old tool open for reference for a month or two before closing it.

Trade-offs worth naming

This checklist costs time up front. Counting record types and checking exports is an afternoon or more, and rewriting AI instructions adds another pass. The alternative is worse. Losing a lead note that explains why a customer hesitated, or running AI instructions against data that is no longer there, costs more than the audit does.

One more caution. Do not close the old account on cutover day. Reduce it to a read-only reference and give yourself a month or two of overlap. When someone asks about a job from two years ago, you will still be able to look it up while you confirm the new tool holds everything.

If you want a hand with the AI side
25minute free fit call
2hour working calls each month
90minutes, minimum training session

Questions people ask

How long should I keep the old tool open?
Long enough to confirm the new tool holds everything, typically a month or two of overlap, with the old account kept read-only for reference.
What if the old tool will not export something?
Write it down during the inventory. You can copy that data by hand before closing the account, or accept the loss, but the decision should be deliberate.
Why do AI prompts break when I switch tools?
Prompts and context files are written against where data lives and what the fields are called. When records move, those instructions stop matching reality, so each one needs review and testing against real records.
Should I import into the new tool before rewriting my AI instructions?
Yes. Rewrite them once the data has landed and you can see the new field names, then test each instruction against a real record before relying on it.

Contact us

Have a question about AI for your business?

Ask about team training, AI Concierge, or whether AI fits a task in your week. A real person on the Ohio AI Lab team will get back to you.

  • Team AI training
  • AI Concierge for your business
  • Not sure yet? Ask anyway