Most practices change their client management software too late, because the current setup is annoying rather than broken and the cost of moving is vivid while the cost of staying is diffuse. Four signals mean it is genuinely time, and none of them is a feature you wish you had. When one of them appears, the migration itself is a smaller job than people fear, provided you decide in advance what is actually coming across and resist the urge to bring everything just in case.
The four signals worth acting on
A second list has grown up beside the system to hold something it cannot represent. A client was missed because two places disagreed. Nobody can answer a simple question about a client without opening several things. Or the person who understood the setup has left. Each of these is a structural failure rather than an inconvenience, and each gets more expensive every month it is tolerated.
What to bring, in order
Clients first, complete and deduplicated, because everything else attaches to them. Documents second, each attached to its client with its date, because a document that arrives without those is filed nowhere. History third, and only what you would actually consult. Anything beyond those three belongs in a read only archive of the old system, which costs nothing to keep and removes the temptation to migrate it.
Run both for one cycle, then stop dead
Overlap for one full cycle of your work so the new system is proved on real cases, then close the old one properly on a named date. The failure mode is an indefinite overlap, where some clients live in one system and some in the other and nobody is sure which. That is worse than either system alone, and it is what happens when nobody sets the date in advance.
Questions people ask about software for client management
How long does migrating client management software take?
The data usually moves in days. The habits take a cycle of your work, which is the number worth planning around.
Should we clean the data before or after moving?
Before, and use the move as the reason. Duplicates and stale records imported into a new system are much harder to find once they look like normal data.
What if the old software will not export properly?
Get whatever it will give you, and keep a working copy or full screenshots of what it cannot. Discovering the export is poor is also the strongest possible argument for never being in that position again.