How to to Build a Worst-Case Scenario Using a SaaS Financial Model
"Shit is hitting the fan", isn't a phrase I've heard regularly from customers over the years. But based on the number of requests for worst-case scenarios we've gotten from customers in the past two weeks, there's a non-zero chance you are losing sleep over this if you are a SaaS founder.
As a financial modeling company, we have helped our customers navigate through tough times in the past. What’s different this time is that we are having these conversations with nearly everyone, and the changes can be very sudden. We want to help, and I believe my experience with other SaaS companies may be useful to you as you plan out your company’s new future.
This post will guide you through the process of building scenario-based forecasts in a SaaS financial model. We will also show you how you can evaluate your new plan. If you are looking to create multiple scenarios by using Flightpath, follow the support article instructions here instead.
Here’s what we cover:
1. Why create scenarios?
2. How to create scenarios
3. Adjust your revenue forecast
- Base-Case scenario
- Worst-Case scenario
- Review bank balance before expense adjustments
4. Adjust your expense forecast
- Freeze hiring
- Cut non-essential expenses
- Pay cuts & unpaid time off
5. Evaluate your new plan
- Expense breakdown
- Comparison to 2019
- Comparison to 2020 Target
- Quarterly view
- % of Revenue
6. Other things to consider
- Payment terms
- Annual plans
Why create scenarios?
Think of scenarios as possible futures for your company.
Because we can’t know what will happen, we need to build out multiple scenarios to get visibility into what could happen.
At this time, the majority of our customers are focused on conserving cash. As they should be. Having cash in the bank will give you enormous optionality to adjust to changing conditions that may last a long time, while enabling you to move quickly once you have weathered the worst of the storm.
The ultimate goal of any financial model should be to understand what your cash balance is each month over the next 12-18 months.
How to Create Scenarios
We are going to build the scenarios on top of my guest post on Baremetrics blog - The SaaS Financial Model You’ll Actually Use. If you aren’t familiar with the template yet, I suggest you keep the original post open in another tab as a reference, but make a copy of the completed version of the financial model here.
We are going to modify the original SaaS model to include your old forecast, and create two new forecast scenarios.
1. Preserve a copy of your old forecast
First, take your original 2020 goal and call it 2020 Target.
Now forget it. Every company is going to be affected by the COVID-19 pandemic, it’s just a question of how much, so expect that your old target is no longer relevant. However, your 2020 Target is going to be useful to compare against as we start building a more realistic outlook.
Next, create a new tab, and call it something like this.
It doesn’t need to be this wordy, but the idea is that you can easily identify this is a former scenario, and it’s hard-coded so you don’t accidentally edit it.
Copy your existing Operating Model values and formatting to the new 2020 Target tab. Don’t just duplicate the tab - make sure to first copy over the values and then the formatting. This avoids duplicating the named ranges in the original Operating Model, which makes it messier to use them in formulas later on.
2. Create Worst-Case scenario
Next, we are going to create a carbon copy of your Operating Model - formulas and all - for your Worst-Case scenario. Instead of copying values, copy over formulas and formatting. Repeat for Hiring Plan, and Revenue Model tabs.
This is going to be a scenario to help you figure out what to do if everything goes wrong. I’ve seen our customers name this scenario as Worst-Case, Fallback or Covid-19 to give you some examples. The idea here is to build a contingency plan - what will you do if everything goes wrong?
3. Create Base-Case scenario
Lastly, rename your Operating Model to Operating Model - Base-Case. I recommend you do the same for the Revenue Model, and Hiring Plan for clarity. This is going to be your operational forecast for the next 12 months, and you’re going to iterate and improve it monthly.
Adjust Your Revenue Forecast
Start by adjusting your revenue. At times like this your revenue growth may not be under your control, as customers are cutting costs everywhere they can.
For our Base-Case scenario, I recommend you find out what happens if your recurring revenue stays flat or contracts a little over the next 3-4 months, after which it starts gradually growing again.
With our example company Southeast Inc, we are assuming that we will have trouble attracting new customers, and existing customers will churn at a higher rate. I’ve also brought down the projected expansion revenue. Overall, MRR will shrink 4% over the next few months.
This is what it would look like in a visual format:
Below are the same changes represented in the Revenue Forecast Model. Note that I’ve hard-coded these changes directly on top of our existing formulas. While you should eventually try to figure out how these changes take place in your marketing funnel / sales pipeline, we need to get the first version out as quickly as possible to understand the magnitude of the problem.
The goal here is not to try to estimate the entire effects of the global pandemic on your business, but what you need to do in order to keep the lights on. Keep in mind that we are adjusting this forecast monthly as new information comes in, which means that at this stage you want to err on the side of caution.
In the Worst-Case scenario, you need to cut much more. Bring out your worst, paranoid fears and include them in your plan. Is it possible your monthly recurring revenue might shrink 20-30% over the next few months? What if the recovery to your current MRR takes a year?
Our example company Southeast Inc sells SaaS to airports. There’s no way they are going to come out of this without significant negative impact on their business.
Here’s the Worst-Case scenario revenue model with a 26% reduction in MRR. Note that the MRR isn’t back to normal even by the end of the year.
As a reminder, you are doing this to get an understanding of the levers at your disposal if everything goes wrong. You need to understand how you will save your business if your worst fears come true.
Review bank balance before expense adjustments
After these changes, it’s useful to look at your cash forecast. We have brought down your revenue but haven’t made any adjustments to your expenses.
Looking at your newly forecasted net profit/loss alongside your bank balance will give you an idea of the scale of cuts you need to make in either scenario. The situation isn’t as bad as it looks here, but it’s meant to tell you that if you continue business as usual, you run out of money in a matter of months.
In a normal business environment, your expenses would be gradually growing, something like this:
When a business appears to take a turn for the worse, we need to figure out how to “stop the bleeding.” This means cutting down your expenses enough so your business doesn’t go under. In Base-Case, this doesn’t necessarily mean drastic measures - think of finding ways to postpone your expense growth. In the Worst-Case scenario it is almost certainly going to result in cutting costs in all areas of your business.
This is one of the first things I’ve seen CEOs do during a time of crisis. Payroll is usually your biggest expense and will move the needle the most. A pause is also a fast decision to make, unlike changes to your current team.
A hiring ban isn’t permanent either. Once you have a new plan in place, you can gradually begin to lift the freeze should your finances allow it.
In the Base-Case scenario, I paused all hires in the Hiring Plan until August 2020, and changed the speed of hiring to be every two months instead of every month.
Cut non-essential expenses
First, evaluate expenses by type and bring down any other types of expenses you expect to decrease. Many of your variable expenses are going down anyway (not a lot of people are in need of a flight ticket right now), so all you need to do is to make sure these expense cuts are reflected in your new financial plan.
You’ll also want to adjust for expenses that will go up. For example, your team working from home might need new equipment.
You will also benefit from evaluating expenses by vendor. I rarely meet companies where the founder or CEO isn’t at least a little bit trigger happy with signing up for new, shiny SaaS tools. Those tools get forgotten, and yet the vendor continues to bill you. If you software spend is, say, $5,000 a month, it’s hard to figure out how much of that is actually needed just by looking at the number alone.
Luckily, there is a better way. Pull up your Expenses by Vendor Summary report from QuickBooks, and export the past 12 months of data, separated by month.
This view enables you to get a better understanding which tools or services are still in use, and what annual renewals are about to hit soon. I usually first look at the total, which lets me see the overall cost of the service over the past 12 months. For example, here I have highlighted how Southeast pays $26,400 a year to Adobe for their Creative Cloud licenses. Given they are already on their way to switching to Figma, this would be the time to cut off a service that’s no longer needed.
I also recommend looking at individual months to identify any annual plans. You can see that Algolia was paid $125,000 in April 2019, which means the renewal is likely hit soon. Given it’s a critical part of Southwest’ application search, this is not an area where they can save.
If your report in QuickBooks doesn’t look like this, you might not be assigning vendors to your expenses. Here’s a walk-through how you can alleviate this going forward by creating bank rules in your accounting system. To apply these for the past transactions, you’d need to “undo” the initial categorization before you can benefit from the rules.
After making these changes to non-essential expenses, take a look at the cash forecast again. These changes have helped enormously in your Base-Case scenario where the revenue contracts relatively little.
Worst-Case improves as well. Instead of running out of money sometime in June/July, we are looking at a zero bank balance in the September/October time-frame. Since we want you to continue operating your business, you will need to find other ways to cut costs.
Pay cuts & unpaid time off
What if cutting everything non-essential is not enough? Given payroll is likely one of your largest expenses, at some point you will have to consider making changes.
We have seen companies where only the leadership team takes a pay cut, but also team-wide cuts have been on the table. I haven’t personally seen unpaid time off offered yet, but I could see it as an option for some employees who want to work reduced hours because of home-schooling or other similar needs. Delaying raises has been commonplace.
Start by modeling out what would be the impact of implementing pay cuts. This is an iterative process, so you can begin by adding only the cuts you know you can make and see if it’s enough to keep you afloat. If not, you’ll come in and go through a second round of changes.
Now, say that you have modelled in all of the cost-cutting strategies from above, and yet more and more customer cancellations trickle in. Like many startups, you might have been operating with a low bank balance, and you rely on the consistent cash flows coming in from your SaaS.
What are your options?
If you don’t touch your payroll, your risk running the company to the ground. At that point, you can forget about taking care of anyone in your team. From what I’ve learned from our CEOs, there are two schools of thought on how to go about navigating this.
In the first option, you only lay off the minimum amount of people required for your company to survive, given everything you know about your business right now. If the revenue slump is temporary, this will be the only layoff you need to do, and you minimize the impact on your team.
The counterpoint to this approach is what if this isn’t enough? What if revenue continues to take a nose-dive, and now you need another round of layoffs? And then another? This will sap your team’s morale as everyone is going to be asking am I going to be next and when are we going under? The "Do it once" school of thought is also the take of David Ulevitch on his write-up about managing layoffs.
You are the best person to figure out what’s the right approach for your company. Modeling out the financial impact of either option is easy - it’s the decision having to part with some of your team members that is the hard part.
In the Worst-Case scenario, Southeast needs to do a 25% executive pay cut, combined with letting go of nine of their 34 employees.
After all of these changes, we finally have a plan that has a shot of keeping your business going.
Evaluate Your New Plan
After making all of these changes in your model, evaluate the new plan. It’s not enough that your cash balance forecast appears to steady out over time - what if your assumptions are flat out wrong?
There’s no way to get everything right with the first try, but luckily we have ways to iterate and improve on your plan. Start by looking at your forecasts with a broader view, but from multiple angles. This is an easy way to spot errors compared to your original 2020 plan, and sanity-check your new assumptions. Let’s look at five quick ways of evaluating your new plan.
Expense Breakdown Over Time
First, your Base-Case expenses should look something like the chart below. There’s a clear drop in near-term expenses, after which expenses stay flat for a few months. Finally, as you expect the business to get back on track in mid-to-late summer, the expense forecast should begin to grow as well.
Same thing here, but for the Worst-Case. Expenses come down significantly, stay flat for a long time, and even when they begin to grow they stay below the current levels for a while.
Comparison to 2019 Numbers
Second, compare your forecasts by using the Reports tab. I recommend starting by comparing each of your forecasts to 2019 numbers. This anchors the newly created forecasts to something familiar, and makes it easier to compare them to each other.
As you can see, you’re still expected to grow your revenue by 30% in your Base-Case scenario, whereas Worst-Case is virtually flat despite the hit in your ARR. This view lets you also make your first adjustments. For example, does it seem realistic that in your Worst-Case scenario, your Support, Engineering and Sales & Marketing team costs go down 7-12%, but your General & Admin (G&A)doesn't shrink at all?
Turns out that this does make sense, since even though G&A expenses are otherwise going down, Southeast has locked into a new lease starting in April 2020 that offsets the savings elsewhere. Now would be the time to figure out can they back out of that lease.
Comparison to 2020 Target
This is where you truly begin to see the impact of the changes you have made. For Southeast, their EOY projected ARR is 40% lower than predicted, and we are talking about just the Base-Case. This is almost $8M in recurring revenue they don’t have available in 2021 to grow their business with.
Southeast was on track to become profitable in 2020, but in the new plan that's no longer possible.
The Worst-Case paints an even more drastic picture with a 60% decrease in ARR vs the 2020 Target.
That said, I’d argue that the Worst-Case scenario might be making too deep cuts to expenses. Southeast is still predicted to have $812k in the bank by the end of 2020, after which the balance starts growing. This is more than in the Base-Case, and more than the entire year's Net Loss (which here equals net cash outflow). In other words, the company could stay flat for the entire 2021 and still have money in the bank.
You need to keep in mind this is not an exercise to maximize cost-cutting measures, but it’s about figuring out what are the minimum cuts you need to make that gives your company a fighting chance once the crisis blows over.
Before we go back and remove a layoff or two, let’s continue our review and see if there are other changes we should make.
Next, I’d look at both scenarios through a quarterly lens. It’s easier on the eyes than a monthly forecast, yet it provides more detail than just the annual overview.
We should go over each line-item on the table below, and ask ourselves: “Do we understand what each change means?” If not, review and adjust.
For example, we understand that Support costs go down from $231k in Q1 to $200k in Q2 because of layoffs you are forced to make.
But do we know why the Hosting and Service delivery costs are coming down? Looking at the forecast more closely, we can see hosting is forecasted out as a percentage of your revenue. When your revenue is projected to come down, so will your hosting costs. In reality, the relationship might not be as straightforward. Tying hosting costs to revenue is a good forecasting method for a perpetually growing company, but what if you just purchased a bunch of reserved instances that you can’t quite shut down if there are less customers using it? In other words, your hosting costs might not actually go down even if your usage does (or at least not right away).
If we change the Hosting forecast formula to never go below your lowest monthly hosting cost over the past three months, the quarterly forecast looks like this.
The hosting cost change alone brought down your end-of-year bank balance down from $812k to $705k. That’s about the cost of one employee for Southeast.
Follow the same logic for all of the other accounts, and do deeper dives as necessary. Keep an eye out for the bank balance, and make sure it doesn't touch zero. Normally, you'd want a buffer of 3-6 months of expenses in the bank, but when modeling out your Worst-Case scenario such rules of thumbs will go out the window.
Quarterly View - Percentage of Revenue
Looking at your spending as a percentage of revenues is another useful way to normalize your forecasts and make them more comparable between each other.
Comparing the two scenarios here lets us quickly see that our Support spending is proportionally similar in both scenarios. After making the adjustment to minimum hosting costs, we can see the hosting costs are actually proportionally higher in the Worst-Case scenario.
Review the rest of the line-items to spot any outliers. For example, is it justified to have 40% of your revenue spent to Sales & Marketing when your revenues are tanking? Or perhaps you have to continue to spend on S&M to be able to attract new customers?
Other Things to Consider
Here are a couple of other considerations that may help you with the tough times.
Many of our larger, profitable customers have utilized loans or lines of credits in the past. They have used these funds to maintain a comfortable cash cushion while investing into sales, marketing, or other areas of the business.
In the current crisis, taking advantage of a loan is not as straightforward. You will first need to understand what operational changes you should make to continue running a sustainable business.
That said, a loan can give you enormous option value. What if things seem to be picking up after this summer and you’d want to continue making new hires without dropping too low cash-wise?
If you can get a loan with reasonable interest, payment terms and not too restrictive covenants (measurements of your ability to pay back the loan), I recommend you take it. You don’t actually have to use the money for anything, but it’s there should you truly need it. Keep in mind that once the economy enters deeper into recession - and perhaps your business looks worse than it is today - a loan may not be in your cards anymore.
In contrast, do not get a loan to delay the inevitable. Say you take out a loan to avoid making layoffs, but the business does not pick up in a month or two. Now you have spent a large chunk of the loan, have a low bank balance, large expenses, and a loan payment plus interest to take care of. A huge exception would be loans such as the US Government Paycheck Protection Program (PPP), which is specifically designed to be forgiven if you are able to keep your team onboard. Here’s a great overview of business relief programs by the Microconf Team.
The key here is to model out your options. You should obviously try to take advantage of the loan if it’s merely there to help you move faster once the worst is behind us, and avoid it if the loan won't really help anyway. However, most of us operate somewhere in the middle, which makes it so important to get clear visibility into what a loan would do to your business if you were to take one.
I’ve included a loan tab in the financial model so you can see yourself what the impact of taking out a loan would be. Principal payments flow through your Balance Sheet and are automatically taken into account in your future bank account balance projection. Similarly, the interest payments flow through your Profit and Loss statement and bring down your Net Income.
Ask for longer payment terms & collect A/R
This is as standard business operations as it gets. Your vendors might typically offer NET30 payment terms - ask for NET60 or NET90. Try to ensure your own customers are paying on time - you can even offer a couple of % points discount for paying early.
Offer discounts for paying annually
I’ve seen customers do this successfully in the past with discounts being anything between 1-3 months of free service. Keep in mind that offering three months free is in some ways similar to a loan with 25% annual interest, so don’t go crazy with the annual discount.
Other things to consider
I’m not a tax advisor, so make sure to speak with your accountant if this is applicable to you.
- Delay Income Tax Payments. The IRS has granted a 90-day filing and payment extension for many pass-through entities and corporations. If you aren’t in the U.S., ask if your tax authority offers something similar
- Tax Credits. If it’s your first five years of business and you are building something novel and innovative, R&D tax credits can be a significant chunk of cash you’ll get back from the IRS in the form of a payroll tax credit.
Conserving cash doesn’t just mean pinching pennies everywhere you can - although you absolutely have to do that too. You’ll need to look into the ways to get more cash in the door, be it in the form of annual prepayments, loans, or negotiating better terms with vendors.
That said, you can’t flip a switch to get your customers to pay annually, and the financing options aren’t always on the table during uncertain times. Therefore it’s essential that you use a financial model to forecast out many possible futures for your company.
While I’m not a healthcare professional, I want to do my part to help out. I’ve seen many ways how companies deal with the crisis and might be able to help you think this through.
I’m offering pro-bono consulting sessions to help founders of any company affected by the downturn. It can be looking at your model and evaluating your assumptions, or figuring out how to add a Worst-Case scenario to your existing model. Book a time for a free help session here and feel free to ask me any questions or request additions to the model on Twitter as well.
Good luck and stay home.