I agree with 69Franx-
Take one of your favorite extract recipes that you could normally brew in your sleep. A pale ale or brown ale is a good choice. You want something on the simpler side.
Sub out the extracts and replace them with base malts. We can help you with that here.
Brew it once, carefully measuring volumes, times, gravities, etc. Keep good notes. You'll need them later.
Taste the result. Compare the gravity, etc, with what you are used to getting from your extract batch.
Tweak the recipe (we'll help if you ask with good notes) so that it comes closer to what you are used to, and repeat the process.
This will let you dial in your system with something that is familiar to you, and you can do a side by side
with your extract batches (assuming you have some left) so you can see your progress.
A good starting point is to use 75% as your efficiency. You may hit higher or lower. The actual number doesn't matter too much as long as you get good tasting beer as the end product. What you are shooting for is the ability to ACCURATELY PREDICT your efficiency so that subsequent batches are not a crapshoot.
That is actually the most important thing to master when making the jump, IMHO. You can always make up for efficiency numbers later, but accurately hitting the numbers you put into the software is what will allow you to fine tune later.
Expect the first couple attempts to not match your extract versions.
if they do, that's awesome and rare. If not, don't be surprised. The majority take a step back for a bit while
they dial in and get the hang of AG. Just keep good notes and ask questions. The beers will improve as you gain practice.
HTH-