How to Use a Headcount Planning Model to Determine Your Next Office Location

To date, you’ve built a 30-person team colocated with you in California, prioritizing speed over cost efficiency, but now it’s time to consider whether to keep that approach, or start to build out a back office in a lower cost geography.

But how do you weigh the headcount cost of each option? In this post, we’ll walk you step-by-step through how to think about this. We used our free headcount planning template to run the numbers. You can download it to follow along or create scenarios that work for your business, so you can consider all the moving parts as you plan your strategy.

Let’s start by laying out the options

Base case: Headcount planning template for hiring at headquarters

Over the next 18 months, you forecast hiring 32 back-office staff across finance, customer success, sales development and sales ops, and marketing ops, all in your home office in California.

Alternate scenario: Headcount planning template for hiring in a new office location

Same number of hires, but in Arizona, plus a senior manager for each team to account for higher coordination costs with HQ

New hires are added in the Input_USA Hires tab, as you can see in the screenshots below:

 

 

 

 

 

Our analysis needs to account for three differences between the scenarios: salary levels, the number of hires, and tax costs.

1. Salary differences. Because salaries for forecasted hires are based on job type and level, not geography, we need to add a multiplier for high cost geographies in CA. We’ve done that where the new hires are added. If a future hire is assigned to CA or NY, we assume a 20% higher salary compared to the standard rate.
2. Number of hires. In the geographic expansion scenario, we have to hire more management staff in AZ.
3. Taxes. This one cuts in two directions. Higher salaries translate to higher FICA expenses per employee in CA, but for more employees in AZ. Also, the Arizona state unemployment tax (SUTA) rate is more than 3x the rate in California. (see table below)

So what’s the upshot? Using variance analysis to compare your options

In our example, the incremental managers hired in Arizona don’t join until June 2023, so if we look at May, the month before they join, there’s an almost 8% (~49,000) higher payroll cost in CA, a 2% (~$1,000) higher payroll cost, and an 8% (~$3,500) higher FICA. It’s slightly offset by 53% ($2,500) lower SUTA. It nets out to about $50,000 higher costs in CA.

But once incremental managers join in Arizona in June 2023, all the savings disappear and in fact the Arizona option ends up costing even more, by ~$3,500 , eaten up by those managers’ salaries and payroll taxes!

What are the takeaways for thinking about location strategy for your business?

• Make sure salary deltas are significant enough to warrant offsetting costs!
  • Think carefully about any differences in staffing levels you’ll need to accommodate multiple offices. They could be the biggest driver of cost in the long term.
  • Consider non-headcount costs (beyond the scope of this post) like additional office space, extra travel between offices, or offsites.

Exploring the limitations to conducting variance analysis in Excel or Google Sheets

We hope you found the discussion above illuminating, and the spreadsheet files behind it helpful if you’re contemplating a similar shift in work location.

But as helpful as spreadsheets are in building a headcount model, relying on spreadsheets while trying to collaborate with your executive team will hold you back. When your CEO asks for the impact of a decision on each department, you’ll kill your credibility if you have to leave the C-suite for your cubicle for several hours. To persuade your leadership team, you need to answer questions in real time.

Why can’t spreadsheets get you there?

1. Spreadsheets stink for managing multiple scenarios. In our case, we had to create three: a base model, an alternative, and a third to calculate the difference between them. If we wanted to change some of the assumptions, like hiring growth or the structure of benefits, or adding more than 100 current employees and 50 new hires, we’d need to make those changes in exactly the same way to both scenarios to avoid miscalculation.
2. Looking at different time horizons is hard. In this example, there was a huge difference between the first 18 months and beyond that. If we only took a yearly view, or only forecast out for a year, we would have totally missed it! But jumping between monthly, quarterly, and yearly views isn’t trivial. You have to rework formulas and recreating charts. That’s a major hassle.
3. It’s a pain to keep a model like this up to date with the latest staff list and compare the forecast with actuals as time progresses. You have to import staff data manually each time, copy/pasting it into the right tabs in each scenario. 
4. You can’t share the spreadsheet in whole across the org—it contains salaries for every employee in the company. Instead you have to create and keep track of redacted copies.
5. We’ve relied on just a few charts in our analysis above. What if we want to see the impact by department, or change the timing of different hires? Or consider an option in a totally different state (or states, or countries?) Such ad hoc analysis can take many different directions, so being able to really quickly build new views into the data and feed the insights back into the assumptions is critical.

The good news is that OnPlan exists precisely to keep what you love about spreadsheets—the ability to build models with unlimited flexibility in an environment and with a syntax you know—while addressing the limitations of spreadsheets, like those above.