How to design a data driven budget

We know about data driven investing. Investments always have data behind them, like historical returns, P/E ratio, technical analysis and stock price.

But data driven budgeting? Yes there are numbers in budgeting, but possibly very few people use data to create their monthly or annual budgets, if they budget at all.

In this post, we are going to elaborate on this topic on what data to use and how to apply it to create a budget.

Let us say you are new to budgeting, or simply want to start afresh. I start afresh almost every year. It gives me a fresh perspective on the current financial situation and an opportunity to correct past mistakes.

Like any data driven analysis, the first step is to collect as much data as we can.

What are the sources of data for a good budget?

• Last 3 months of credit card statements.
• Last 3 months of checking and savings account statements.
• Paycheck statements
• Tax bills
• Holes in your savings plan – for example, if you have not yet accumulated 6 months of emergency funds or another goal is underfunded.
• How much investments are you making over the last one year? Are they regular or sporadic?
• What kind of debt do you have, how much and at what interest rates?
• What are your recurring payments – insurance, debt and utilities.
• Subscriptions on the online platforms – Amazon, Google, Apple

If your head is spinning, don’t worry. Just get these ready (or at least mentally) and we will organize all of this into a nice little budget.

Organizing the data around objectives.

There are 3 areas of a budget in which most of the data can be dropped and further refined.

• Debt reduction and obligations
• Expenses
• Savings and Investments

Let us drop the data we collected into these 3 buckets.

Now take the data in each bucket and come up with a monthly/annual figure.

First you need to strive to reduce the bill as seen from the data for previous months/years.

For example, in the first bucket:

• Refinance your mortgage, if it makes sense with extra costs.
• Negotiate better rates for car and home insurance.
• Check property and tax bills if you are paying extra or can be reduced.
• See if you can transfer balance from one card to another to reduce interest rates, or simply pay it off.

In the second bucket:

• Check your paycheck and make sure everything is as expected. https://smartasset.com/taxes/paycheck-calculator
• Check subscriptions and cancel all that are not needed or not used in last 3-6 months.
• Check credit card statements and checking accounts to identify unusual cash flow per month.
• Make sure your accounts are not charged for overdraft, savings withdrawal etc.
• Are the auto-debits all making sense?

In the third bucket:

• Check the fees for investment products, for example, expense ratio of mutual funds and ETFs.
• Interest rates on savings accounts.
• Are all auto-transfers expected and need fine tuning?
• 401k cost.

The next step is to come up with numbers for each, either monthly or annual.

Let’s take the first bucket. Typical numbers could be something like:

• Mortgage Payment – $800 (monthly)
• Property Tax Bill – $2400 (annual), $200 (monthly)
• Home insurance – $600 (annual), $50 (monthly)
• Car insurance – $600 (bi-annual), $100 (monthly)
• Credit card payoff – $150 (monthly)

With above, the monthly bucket-1 bill is $1300.

The second bucket is more tricky and difficult, since it involves variable expenses.

Assuming you have reduced all extraneous subscriptions, the following should be calculated.

• Value of all auto-debit (utilities, phones etc.) – $300 (monthly)
• Food and groceries – $500 (monthly)
• Transportation expenses (gas, car maintenance) – $200 (monthly)
• Non-discretionary expenses (entertainment, hobbies) – $300 (monthly)

This will vary quite a lot from month to month, but it will be good to take an average for each category over 3-6 months. As from the example, we have another $1300 allocated to this.

Last but not the least, you need to save and invest. This is non-negotiable part of a budget. If the above two buckets do not leave anything to save/invest, then go back and try to adjust till you have at least some surplus.

Remember with the Pay Yourself First, this is for your future – both short term as well as long term. Even though it is bucket number 3, while automating the budget this will come up on top, that is, the amount will be skimmed off the account as soon as the paycheck hits.

For this bucket, there are few considerations and allocations to be done.

• Cutting out the non-discretionary expenses, we have $2300 of monthly essential expenses. If the savings does not have $2300*6=$13800 to create an emergency fund, then that is the first item to allocate for.
• Let’s assume the savings account has $4000 of savings, so monthly we have to allocate $800 to build the corpus over next 12 months. Of course, if this is difficult, then the time horizon can be increased but not more than 18-24 months.
• Let us also assume we have $1300 surplus to allocate to this bucket. So rest $500 can now be saved or invested for other goals, medium or long term.
• After 12 months, the emergency fund contribution can be diverted to further goals or boost the long term investments.

So finally we arrived at a 1/3 budgeting style, that is, $1300 in each bucket. The below post describes this budgeting method in more details.

Budget – Grow the tree upside-down

The above example is for anyone earning $5000-$6000 gross monthly pay, so that after tax and deductions, the paycheck will be close to $3900.

Conclusion

This post has shown a data driven way of constructing a budget or refining an existing budget. The top down approach starts with no assumptions and rely on the data first to drive the various areas of the budget. It also takes away emotional bias like excuses – I cannot save, the debt is huge for me to pay off, I need a bigger house etc. These emotional elements cloud the judgement of the budgeteer and ultimately fails the entire exercise in next two to three months.

When you start from data, you know how to monitor and you always have a reference ($1300 in our example) for each bucket. Anytime one goes up or down, something changed and should be either reallocated or rebalanced.