It's kind of crazy that it's been nearly a year since last doing a batch of work on this - not sure where the time has gone. My learnings in the past year:

On Behaviours

• Opening up Clerk on a regular basis just never became a habit that stuck. I would leave it unopened for months and then open it when I was having anxiety around money or wondering where my spending was going. I definitely personally have avoidance issues around money, but I'm guessing this is also true for most people.

• At some point I wanted to do some research for business expenses and I threw my raw revolut export into a google sheet. This behaviour persisted much more than opening up Clerk - I think the combination of G sheets being familiar, and the amount of control that comes from a spreadsheet, had more of a draw.

• Over time I got really sophisticated with the spreadsheet, eventually landing on a formula based approach that almost mimicked the Clerk UI. I had all of my categories listed in the first sheet and then a category column in the transactions spreadsheet that pulled them in dynamically, and a table that shows category breakdowns over time.

On Features

• Despite doing reasonably well this year, there were still multiple months where I dipped into my savings before the month is out. I really regret this and wish I'd have been more controlled/intentional around that. My problem really stemmed from making poor decisions at the beginning of the month and accumulating more spend that I should have.

• That said, I still haven't found a system simple enough to give me a no brainer answer to "Should I make this purchase". The ideal would be some set of the following:

• When salary gets paid, subtract fixed and recurring costs to get the total amount of discretionary spend available for the month. Then divide this by the amount of days in the month and show a "You have $X per day". Then as the month progresses show some indicator as to whether you're adhering to that or going over. Then you can make decisions about purchases based on how much you've gone over or under your baseline per day.

Tying this Together

• What I'm now thinking about is Clerk the app being a companion to a Google Sheet.

• The sheet could either be given away for free or sold as a one time payment. There's enough value in the sheet itself to exist separately to the app. And we could do up a little help center and set of videos on how to use it.

  • Continuous Sync: No need to export as csv and import into G sheets which is a pain

  • Auto Categorization: Currently you have to go through and manually assign every transaction. This would be trivial with an AI API.

  • Multi currency normalization: Bank exports don't always offer a unified currency column, so totals can be massively wrong because they're adding, for example, Yen amounts alongside Euro amounts.

  • Backups & Rollback: The big downside of a g sheet is that it's easy to break. Giving users the ability to rollback in case they break something by accident could be super valuable.

Thinking:

• At some point I should combine Blog & Dev Blog - currently the posts on Blog are pretty out of date. On the other hand I could also just hide the dates altogether - I kind of loathe the idea that a significant amount of the implied value of a post comes from its recency. At the same time, people don't want to see a flat list of posts as it's just a wall of text.

• Idea: What if I break down the posts by month, but remove the year. So posts from March 24 would show alongside posts from March 22.

• Update profile photo

• Clean up/remove some old posts

• Hide year on posts index page

•