- Use the simplest methods to minimize sources of error when calculating probabilities.
- Add ARR from closed/won deals, estimated ARR from in-pipeline deals, and projected ARR from future leads to estimate your ARR.
- Examine historical data over two to four quarters to calculate conversion rates per deal stage for best accuracy.

Where will you land by the end of the year?

Every growth-oriented SaaS CEO routinely faces this question from their board. Aside from assuring the board, the answer to this question also reveals gaps to address.

SaaS business environments are dynamic and you must adjust quickly to sustain growth. You’ll need to check and revise your forecasts monthly to identify growth bottlenecks and correct your course. For instance, if you have failed to reach a revenue checkpoint, you could hire and build your pipeline before time runs out.

The sum of the four revenue metrics below will help you figure out where you are and how far you have to go:

- Revenue from closed/won deals
- Expected ARR from in-pipeline deals
- Projected ARR from future opportunities
- ARR lost from cancellations and downgrades

**Expected ARR = Revenue from closed/won deals + Expected ARR from in-pipeline deals+ Projected ARR from future opportunities - ARR lost from downgrades and cancellations.**

- SaaS companies experience a considerable degree of churn over time. Reactivation ARR is a fifth metric you must consider if you are operating in a highly competitive space where churn is common over several years. Given the complexity of organizing data sources that reveal this metric and the volume of reactivations needed to make it significant, newer companies can safely leave this out of the picture.

Let's look at how you can calculate these metrics separately.

## Calculating ARR from closed/won deals

The first variable is the easiest to figure out. You can retrieve this number from your CRM and plug it into the formula above.

## Calculating estimated revenue from in-pipeline deals

The second variable in our expected ARR formula poses a few challenges. You must calculate the probability of closing deals in your pipeline and sum the expected ARR from those deals. The challenge is: How can you calculate deal closing probabilities with reasonable accuracy?

You can use historical pipeline data to create reasonable estimates of projected ARR.

Deal close probabilities rely on a few factors:

- Deal cycle time
- Deal value
- Win rates - How many deals do you close/win from the opportunities you receive?
- Average deal stage times
- ARPA

The more complex your deal cycle, the greater is your ability to dive deeply into your data. However, our purpose is to create a reliable estimate, not conduct deep data analysis.

Here's a sample ARR projection for pipeline deals based on stage-weighted closing probabilities:

- You can add dimensions based on sales cohorts critical to your ARR. A cohort is a group of customers sharing a set of attributes such as region, market segment, industry, product line, etc. For instance, if win rates across market segments are significantly different, you can break down the numbers in the table above per market segment to calculate a more reliable ARR estimate.

This example results in a projected ARR of $1.4 million. While the calculation here is quite simplistic, it serves our purpose of reaching a reasonable estimate.

You could create more sophisticated models by measuring the average time a closed/won deal spends at each stage. For instance, if the average closed/won deal spends 10 days in the first stage, an existing deal that has spent 20 days is less likely to convert.

However, you must also assign weights to each deal stage, with advanced deal stages receiving more weight. You could give the first stage a 10% weight, reducing its impact on total calculations.

Note that while this method will yield more accurate probabilities, it still relies on estimates. Thus, the marginal return on the additional time you spend modeling is unlikely to deliver an outsized impact on your projections.

## Calculating expected revenue from future opportunities

The third component of our expected ARR formula is the most complicated and error-prone. While the previous step looked at existing deals, this one considers future opportunities and their probability of entering your pipeline.

In short, you'll be leaving your sales funnel and looking at average lead volumes. We recommend a lookback period of two to four quarters. Let's assume the following:

You have an estimated $12.5 million entering your pipeline's first stage. From the previous step, we assigned a conversion probability of five percent for this stage. Thus, our projected ARR from future leads is $625,000.

Note that this calculation assumes your deals will close by the end of the year. You can use a moving average of leads generated per month to reliably account for changing performance over time.

Like the previous step, these calculations are straightforward and help you arrive at a ballpark figure. However, the marginal return on the additional time you spend doing this is low.

- You can use a forecast category-weighted method instead of a stage-weighted one. For instance, if you have classified deals in your CRM based on forecast categories such as “Commit,” ”Pipeline,” or “Best-case,” you can sum deal ARRs per their probabilities.

## Calculating ARR lost from cancellations and downgrades

The fourth step is to account for ARR lost from cancellations and downgrades. Your Customer Success Management (CSM) platform will give you customer health data that you can use to estimate this number.

If you don’t have a CSM platform in place, you can use a moving average of the churn rate over the past two to four quarters.

You can sum the results of the previous three steps and calculate lost ARR as a percentage.

- Revenue from closed/won = $6 million
- Expected ARR from in-pipeline deals = $2.5 million
- Projected ARR from future opportunities = $1.5 million
- Total projected ARR = $10 million
- Churn rate = 11.42% (We’re assuming this number applies till the end of the year)
- ARR lost due to churn = 11.42% of $10 million = $1.42 million

## Putting it all together

In the final step, we sum the results of our calculations to arrive at the projected ARR. You must apply a sanity check to your final figures. For instance, if your projected ARR shows a growth rate wildly out of proportion with your target growth, re-examine your data.

Here are a few tips to help you create accurate projections:

- Deal stage probabilities are the biggest source of error - Examine conversion rates thoroughly.
- A moving average of leads generated will build seasonality into your projections.
- Consider deals whose cycle times fall within the end of the fiscal year.
- You can retroactively calculate projections for previous years and compare them to actuals to verify your model's accuracy.

Estimating ARR can become tedious if you dive into data rabbit holes. However, your objective is to create mile-markers that tell you whether you're on track to meet your goals.

Thus, you're best served using simple methods that minimize assumptions and potential sources of error.

## How Drivetrain increases ARR forecast accuracy

- This post is the first of our Revenue Forecasting series. In the next post, we outline how you can improve forecasting accuracy.

The method we have outlined here is simple and you can implement it using a spreadsheet. However, if your business is experiencing rapid growth, maintaining a spreadsheet and adding complexity to your models is costly and challenging.

For instance, we highlighted a case where you can break down in-pipeline ARR projections by adding dimensions pertinent to critical sales cohorts. Adding these additional data to a spreadsheet, verifying accuracy, and even formatting data is tedious and time consuming. Tracking actuals versus the plan is more cumbersome since you’ll have to enter data manually and query inputs from several business systems.

Even worse, tiny errors can compound and create projections out of sync with business reality.

Drivetrain makes scaling complexity easy by pulling data automatically from your CRM, ERP, Billing, HRIS and CSM systems. Thanks to this seamless integration, you can easily account for new regions, modify territories, add new product lines, market segments, update pricing plans and more when projecting ARR.

Drivetrain helps you compute moving averages of any metric over any time period and build forecasts quickly using business formulas that don't rely on cell references thus eliminating errors. Drivetrain gives you the insights you need to figure out whether you’re on track to hit your ARR targets.

Curious how Drivetrain can help you increase year-end forecast accuracy? Get in touch with us for a free demonstration today.