Salesforce Opportunity Stages and Their Probability

Clari logo

Clari Team



Ready to take your revenue to new heights?

Photograph of a yellow table with paper graphs and a plant
Photograph of a yellow table with paper graphs and a plant

Want better forecasts?

Try creating universal opportunity definitions, setting benchmarks for your team, and embracing automation.

It's not uncommon to find Salesforce admins and business analysts struggling with defining Salesforce opportunity stages and the associated probability values—but why is it so important?

One of the core purposes of a CRM system is forecasting. This means that the company needs to understand as accurately as possible how much money they are going to make and, therefore, how much money they can afford to spend. This is called "revenue forecasting." Other forms of forecasting are "post-sales resources forecasting" (e.g., "How many deployment specialists are we going to need?") or "supply chain forecasting." Fortunately, there are two simple measures to get to a significantly better place when it comes to forecast accuracy: Salesforce probability values and Salesforce opportunity stages. The good news is that these measures can be used independently of Salesforce's more recent and complex forecasting features. Before we dive into the two measures, let's look at how forecasting works at a high level in Salesforce.

Salesforce forecasting basics

Every opportunity object in Salesforce has four important fields related to forecasting: Amount, Stage, Probability (controlled by the Stage field), and Expected Revenue. If Amount and Stage are populated, the Expected Revenue automatically evaluates to Amount x Probability.

To illustrate how this works, let's say you have a $5,000 deal in the pipeline that will be closing with a 50% probability. With those numbers, the CFO can expect $2,500 revenue from that deal at this point in time. This forecast is the value of Expected Revenue.

This example illustrates the importance of two things to ensure accurate forecasting:

  1. Assigning the right percentage value to each sales stage
  2. Making sure users pick the right value at the right moment

Now that we've covered some forecasting basics, let's dive into the two key measures keeping Salesforce admins up at night.

Probability values

As noted earlier, the Probability field is controlled by the Stage field. That means that every pick list value of that field has a specific percentage defined. These pick list values are chosen to reflect the stages/milestones in your unique sales process.

Salesforce opportunity stages

The assigned percentages to the individual pick list values should not be done at random. Instead, for every stage, you should look at past opportunities, and for each sales stage ask yourself the question, "What's the percentage of deals that closed after they reached this stage?"

After carefully evaluating a representative sample of past opportunities, you might come up with a table like this:

  1. Discovery: 14%
  2. Product Demo: 21%
  3. Trial: 63%
  4. Contract Negotiations: 92%
  5. Closed – Won: 100%
  6. Closed – Lost: 0%

The example table above indicates that only 21% of all opportunities that reached the stage Product Demo eventually closed. The remaining 79% dropped out.

At first, these "unround" numbers might seem odd, but this breakdown of percentages by stage can produce a surprisingly realistic Expected Revenue number without any extra effort for your users.

Salesforce opportunity stages

Because the Probability field is set by the Opportunity stage, it is critically important that sales reps have a clear understanding of when to pick each stage—from Discovery to Closed Won or Lost. Ambiguity must be avoided at all costs.

One of the most common problems stems from uncertainty around whether a stage refers to a "completed milestone" or a "task in progress." For example, if you have a stage called "Product Demo," is it clear whether the stage means, "a product demo has been scheduled," or the quite different outcome of, "we gave them a product demo and now we are working on the next milestone (e.g. product trial)"?

The "tasks in progress" paradigm, where a sales stage is defined by a number of individual tasks that are supposed to be completed without any particular order, can be especially effective. For example, the Product Demo stage could be defined by the tasks "schedule demo," "prepare the demo," "conduct the demo," and "analyze feedback." Creating a checklist of required steps that must be completed before the opportunity can move to the next stage can help eliminate confusion and errors that can negatively impact forecast data. The point is to make sure everyone is on the same page when it comes to what is included in each sales stage.

Having too few stages can also impact your ability to generate accurate forecasts. In the earlier examples of stages and assigned percentages, there are huge gaps between Product Demo (21%), Trial (63%), and Contract Negotiations (92%). These gaps introduce a lot of vagueness that can impact your Expected Revenue. To help ensure more accurate forecasts, you would ideally find sales stages that fit in between these. For example, you could look at splitting the Product Demo stage into two stages, each with a percentage value that can be assigned based on completed tasks in a checklist.

Accommodating multiple sales processes with different percentage values

If your company is selling different types of goods or services that all have the same sales stages but the percentage values are different, you will have to create more than one sales process with the specific percentage values. Salesforce won't let you create the same Stage Name twice, so you will have to create a variant of the stage name to assign a different percentage.

As your business changes, the percentage values will change over time too. In order to keep your forecasts as accurate as possible, create a schedule to verify the percentages at regular intervals and make any necessary changes.

Improving opportunity stage data and forecasting accuracy with automation

The measures above can have a significant impact on improving your revenue forecasting, but they are still susceptible to user error. Sales reps may accidentally select the wrong stage or forget to advance a stage altogether. And honest mistakes aren't the only concern—sometimes reps will skip these steps because they see them as being administrative and taking time away from actually selling. If opportunities aren't properly advanced to the next stage, the probability values won't be calculated accurately and sales leaders won't have the data they need to effectively forecast what will and won't close. If you use Salesforce as your CRM, one powerful way to reduce human error is by automatically advancing an opportunity stage based on an action, such as the type of meeting set or completed with a prospect.

A Salesforce-native sales engagement platform like Groove can bring standard and custom Salesforce fields right into Google Calendar, as seen in the image of a Google Calendar event below.

Salesforce fields in Google Calendar via Groove

These fields are a direct window into your Salesforce instance and update bi-directionally in real time before or after the event is complete. Once a rep makes an update, Salesforce can automatically respond with a number of different automation options including creating an opportunity, updating a field or stage of an existing record, and/or the majority of other manual tasks that they'd have to otherwise go into Salesforce to change. For example, if a rep books a meeting, Groove can automatically sync that activity back to Salesforce. Groove also makes it possible for a rep to specify right in their calendar that it was a demo meeting and automatically have the corresponding opportunity move to the demo stage.

Logging those meeting types back to Salesforce also helps provide more granular reporting. For example, that data can be used to understand how many discovery meetings lead to how many demo calls, to how many qualification meetings, pricing meetings, etc. Similarly, this same Salesforce integration could also enable a rep to kick off a Salesforce automation from their inbox that creates an opportunity based on the type of email being sent to a prospect.

By bringing this automation into your sales reps' existing workflows, you are able to capture more accurate analytics in reporting and eliminate time-consuming steps for reps like creating and updating items in Salesforce manually.


A powerful way to improve your revenue forecast is by avoiding ambiguity around sales stages and using empirical values as percentage values. Additionally, using a sales engagement platform to automatically update opportunity stages can go a long way in avoiding human error in selecting stages or forgetting to advance a stage. Even better, all of these recommendations improve your revenue forecasts with no extra effort from your users.

For more information on how Groove is helping sales organizations improve their revenue forecasts by automating Salesforce opportunity stage advancement based on Inferred Account Status, request a demo today.

Note: These recommendations were adapted from an original post by Alex Kerschhofer.