We’ve long observed that Q4 tended to be a heavy-hitter on paid vacation time, but I was surprised by the distinct seasonality pattern I found in our data. In our software development consultancy, time is quite literally money; when makers are on vacation they obviously can’t be working on billable projects. In this post, I look at vacation time seasonality and its impact on revenue.
Vacation time seasonality
I use the term “vacation” to mean both paid vacations and paid holidays. If vacation was taken uniformly over the course of the year, each quarter would see 25% of the total, and there’d be no seasonality. As it turns out, the amount of vacation taken in 2016 increased almost perfectly linearly (R^2 of 99%) through the year, as the graph below shows.
Since we don’t expect to generate any revenue while people sip rum drinks on the beach, we don’t think of vacation as lost revenue. But since vacation isn’t taken uniformly across the year, we do have some seasonality in our revenue. Our capacity to generate revenue has a seasonal component. One way to measure this is to consider how each quarter differs from an “equal share” of 25% of the vacation. By multiplying these differences by our utilization and hourly rate, we see how each quarter’s revenue varies against the “equal share” case.
In 2016, vacation had the following effect on our quarterly revenues: Q1 saw approximately $88,000 more revenue because Atoms took less vacation, and Q4 saw approximately $97,000 less revenue for the opposite reason. These represent a swing of +4.2% and -4.6%, respectively, from the “equal share” situation. Q2 and Q3 showed smaller differences.
Vacation isn’t the only thing that affects revenue. Market conditions, capacity of the company, billable rates, sick time, project credits—these factors have a potentially much greater impact on revenue than use of vacation time. Unlike those factors, the effect of vacation on revenue is very predictable.
The graph below shows our revenue variation by quarter compared to uniform distribution of revenue across the year (25% of annual revenue falling in each quarter). Clearly in 2016 other factors were dominant over vacation seasonality.
Atomic’s vacation policies
While weekends are great, everyone needs longer breaks from work. We provide paid holiday and vacation time to all employees so they can remain fresh, creative, happy, and productive. Taking your vacation is a basic expectation of employment.
Atomic’s vacation policy is on the generous side of typical for our industry. Some aspects of our policy play a role in the seasonality of vacation.
Employees with less than three years of full-time professional experience (not just Atomic tenure) have two weeks of vacation. With three years of professional experience, people have three weeks of vacation. After 10 years Atomic experience, they have four weeks of vacation. In practice, this means only recent college grads have less than three weeks.
Sick days are unlimited and paid. Personal business time is unpaid.
We don’t insist, but we strongly encourage everyone with three or four weeks to take at least one full week of vacation (nine contiguous days off).
Everyone gets nine days of holiday, the six majors and three floating. Q3 and Q4 have slight surpluses of fixed holidays compared to an even distribution (33% versus 25%); Q1 and Q2 have slight deficits (17% versus 25%).
Everyone receives their full vacation allotment on January 1st each year. New employees receive a pro rated amount on their hire date. This allows maximum flexibility for using vacation time whenever you’d like during the year.
With rare, planned exceptions, you must use your vacation the year you receive it. We switched to a “use it or lose it” policy a number of years back when some people weren’t taking their vacation, or were turning it into cash (since we use hourly pay, this was a simple matter of working a full week and claiming a full week of vacation pay.) This policy also means we don’t have to worry about tracking unused vacation or the wage liability associated with it.
Year end is a typical time to take family and holiday related time off, partly explaining the higher use of vacation in Q4. In addition, people who find themselves with unused vacation going into December may take time off at the end of the year so they don’t lose it. We encourage planning.
Sick time seasonality
Sick time shows a different pattern than vacation time. Q1 and Q3 have twice as much paid sick time as Q2 and Q4. Perhaps those quarters are flu/cold season and back-to-school germ exposure? At any rate, vacation time is nearly 9x greater than sick time, and sick time is only 10% of total PTO (paid time off). The impact on revenue by looking at total PTO versus only vacation time is that Q1 and Q4 are slightly less differentiated; the overall seasonality pattern I found remains.
While the revenue impact of vacation time seasonality is very predictable, it’s also likely to be small compared to other factors that affect revenue. Our policies may affect the timing of vacation usage, but that effect is likely a small portion of an already small impact.
- Atomic Ownership, Part 3: Valuation - January 2, 2019
- Atomic’s purpose: to be a company where work matters - November 5, 2018
- Elevating & distributing “glue work” flows out of our core principles - October 18, 2018
- Atomic Ownership, Part 2: How the Atomic Plan & our ESPP work - July 16, 2018
- Atomic Ownership, Part 1: Context, motivations, and accomplishments to date - May 17, 2018