Almost every church we've onboarded onto TheRehoboth was running on some combination of a spreadsheet, a notebook and a WhatsApp broadcast list before they switched. That's not a criticism — it's what's available when the alternative is expensive, generic business software.
The migration itself is usually the biggest fear. Nobody wants to spend a weekend retyping years of membership records, and nobody trusts a system enough to abandon the spreadsheet on day one anyway.
That fear is usually well founded, because most software treats a migration as a one-time import: upload a file, hope the columns map correctly, and live with whatever broke. We treat it as a conversation instead. Someone on our side sits with the church's existing spreadsheet, asks what each odd column actually means — a "status" field that turns out to encode three different things depending on who filled it in that week — and cleans the mapping before anything gets imported, rather than importing garbage and asking the church to fix it after the fact.
So the onboarding process is built around parallel running: import what exists, keep both systems for a few weeks if needed, and let the office move over gradually as trust builds. The goal isn't a dramatic cutover — it's a quiet point, usually a few weeks in, where nobody opens the old spreadsheet anymore because the new system already has everything and more.
During that parallel period, the two systems usually diverge in small, telling ways — someone gets added to the spreadsheet but not the app, or vice versa. Rather than treat that as a failure of the migration, we treat it as useful signal: it tells us exactly which workflow habit hasn't moved over yet, and lets us go fix that specific habit — a reminder, a simpler form, a shortcut — instead of guessing at what's missing.
That's a more honest measure of adoption than any onboarding call: the day the spreadsheet just stops getting updated. Nobody announces it. It just happens, usually a few weeks after go-live, and it's the actual signal that the migration worked, not the 'import successful' message from week one.
