How to scenario plan for Covid-19

Scenario planning is an exercise in deciding what to do before you get punched in the face.

It’s always good to do scenario modeling… if this ends up being 6 months instead of 2 months, have you modeled out what that means and do you have a plan? It’s always better to make those decisions calmly and with plenty of forethought than to be forced into critically difficult decisions very quickly.”  - Bill Gurley on Invest Like the Best

In the last two weeks, virtually every business in the US was forced into a new round of scenario planning. Given how quickly the situation is evolving, they’re likely to continue to do so in the weeks to come.

At Basis One, we're working with companies whose revenue has gone to zero, and others whose growth has accelerated beyond even their most optimistic projections. Here are a few of the lessons from this process, with an emphasis on how growth teams can project user and revenue growth.

Make sure that you get both the positive and negative extremes on the page

Job #1 is to build and understand the worst case scenario. What would happen if you had to sustain 6 months of zero revenue, and 18 more without the ability to profitably acquire new customers?

However, it is also about understanding the upside case. Putting aside businesses that are obviously benefiting, I suspect many tech companies can emerge stronger from this. Their legacy competitors have been hit even harder. Consumer behavior is shifting rapidly, leading to lots of discovery and trial (if not much purchasing yet). Customer acquisition is less competitive, including CPMs down about 50%. All of this means that for many businesses, by 2021 or 2022, the upside case could be better than your pre-crisis forecast.

Before it’s clear which scenario you’re in, it’s all about preserving optionality. Cut spend to extend runway. To the extent you can, balance cost cutting with keeping as much of the team in place as possible to be able to execute. Maintain trust with your customers by not making knee-jerk changes to product, policy, or pricing.

Build your scenarios around the potential shapes of economic recovery

We’re likely to see one of three patterns:

  1. V-shaped recovery: A shock displaces demand but does not impact it in the long run. Historically, most epidemics have followed this pattern, but this is highly unlikely given that Covid-19 has cascaded into broader economic impact.
  2. U-shaped recovery: A shock breaks the growth trend. Growth resumes on the same path, but later than planned. Financial recessions typically follow this pattern.
  3. L-shaped recovery: A shock impacts a fundamental component of growth and results in a lower growth rate even after recovery. Japan’s “lost decade” is the canonical example.

This slide from BCG helps illustrate:

The recovery will not be evenly distributed

Different industries and companies are going to see different shapes of recovery. Here are two questions to ask yourself to help understand which you’ll see:

Are my customers actually churning, or just going dormant?

The more it is the former, the more likely you are to see a U or even L-shaped scenario. This may be the unfortunate reality for many businesses that serve small businesses, for example, because do not have the cash to survive a month without revenue.

Has underlying customer behavior actually changed?

This is likely to be true in the extreme negative cases such as travel and events, and in the extreme positive cases such as online education and food delivery. We will certainly see regression to the mean once full optionality returns for customers, but some new behavior patterns are being hardwired as we speak. If this is the case, businesses are more likely to see the positive or negative version of an L-shaped recovery.

Your customer acquisition machine is currently broken, and there is a premium on quickly understanding when it is working again

Most companies get big by getting one distribution channel to really work. The metrics that make it “work” currently aren’t performing how they typically do. This means that many businesses currently have zero sustainable acquisition channels, and as a result they’re sitting on the sidelines.

How can you build confidence that your payback periods are back within an acceptable range? You’ll get faster feedback on the cost side of the equation, especially if you’re able to keep some campaigns live so you can get a pulse on how the market is evolving.

The LTV side of the equation is harder to model. One useful approach is to break down the components of LTV, and build a new set of metrics that serve as early indicators for each. Here are a couple of examples that we’ve seen used effectively:

  1. Activation rates: if you typically look for customers to take an action a certain number of times in their first week or month, try looking at the rate at which customers take that action on their first day. You’ll often get 60-80% of the accuracy as the slower metric in predicting if customers will retain
  2. Transaction volume: use metrics that are upstream of purchasing, such as browse, cart add, and saving as an indicator for latent demand that may not be converting today due to high uncertainty.

By continuing to monitor these early indicators, you may be able to build the confidence to go back on the offense faster than others.

The error bars on your scenarios are going to be very high at first. That is actually a good thing, because it will force you to think at the extremes of how the pandemic could play out, and how you would respond.

Going through the process will give you the scaffolding on which to hang new information as you receive it, and you’ll be able to drive to greater and greater clarity over time.