Step 0: Making sure your data is what we call "clean."

Your data must be valid, accurate, complete, consistent, and uniform. We can clean data to a certain extent, but eventually, we must reach those five characteristics.

If none of your numbers/processes/people have these characteristics, skip my posts for the next 2 weeks. Find a bookkeeper ASAP and start over. This is non-negotiable. I'm serious.

If you don't have any sort of logical consistency, you do not have business - you have a dumpster fire. Actually, you don't even know if you have a dumpster and can't tell if it's on fire. Also, your accountant hates you.

Think of this like cooking a meal. We need the right ingredients. We need the right amounts. If we don't know either of those, we will never be able to find out which recipes we can make, if our food is any good, or how to fix it if it's bad.

This is half the battle. If you can do this, you are set up for success.

Below are a few items to look out for. We can solve some of these, but it will add unnecessary complications.

// Each account has a unique ID (and name). // Each financial transaction includes all key details — like date, amount, account, and category. // Each sale record includes all key details — like date, customer, product, product category, and sale amount.

// Fields are used correctly. Product names go in the product name field, not in the notes or description field. // Names are consistent. If I sell CoInsight's world-class burger, I will classify that as "World Class Burger" every time — not "WCB" one day and "classy burger" the next. Pick one version and stick with it. // Categories are consistent. If I call something "Office Supplies" in one place, I shouldn't call it "Office supply" or "Supplies - Office" elsewhere. // No duplicates. Customers, vendors, products, or accounts should only exist once in your system — not spelled slightly differently or with minor variations. // No impossible or illogical values, like future transaction dates, negative sale amounts (unless it's a return), or customer age = 200. // "Miscellaneous" and "Other" are used sparingly and only when truly appropriate. Overuse of these is a sign of unclear data.

From this point forward, I will assume that you have been in business before, you keep records, and you know the importance of accuracy and uniformity.

Step 1 starts tomorrow.