From spreadsheet forecasting to a 48-month view of future revenue
Most businesses can tell you what they’ve already sold. The harder question is what happens next.
For one global enterprise software company, answering that question had become increasingly difficult. Revenue forecasting relied heavily on spreadsheets, renewal income wasn’t always easy to distinguish from new business, and important commercial data was spread across different parts of Salesforce.
The information existed. Turning it into a dependable view of future revenue was the problem. Sweet Potato Tec worked with the business to change that, creating a predictive forecasting model inside Salesforce CRM Analytics that could look as far as 48 months ahead.
Organisation
Global enterprise software company
Sector
Enterprise software, data infrastructure and cloud
Location
US headquarters with global operations
Annual revenue
$50m to $100m
Enterprise software, data infrastructure and cloud solutions
About the client
The client is a global enterprise software company specialising in data infrastructure and cloud solutions. Headquartered in the United States, the business works with Fortune 500 organisations, technology companies and high-growth businesses across multiple industries, generating annual revenues of approximately $50 million to $100 million.
At that scale, forecasting isn’t simply a sales reporting exercise. It affects hiring, investment, renewals, targets and the decisions leadership makes about where the business goes next.
The company needed a better way to understand what its revenue data was telling it.
The challenge
The situation
Forecasting had become too dependent on spreadsheets. That made producing forecasts slower, but the bigger issue was confidence.
01
Renewals weren’t always clearly separated from net-new business, making recurring revenue projections harder to interpret.
02
Revenue-critical information was spread across opportunities, products and accounts.
03
Duplicated, incomplete and inconsistent Salesforce records created further uncertainty.
04
The existing approach could show what had happened and some near-term pipeline, but couldn’t reliably look further ahead.
For leadership, that created a fairly fundamental problem. How do you make confident decisions about the future when you’re not completely confident in the forecast?
What would have happened if nothing had changed?
The business could have continued forecasting in spreadsheets. Plenty of organisations do. But as the company grew, the limitations of that approach would have become more significant.
Finance, Sales and Operations would continue producing different interpretations of future revenue.
Renewal projections would remain difficult to separate from new business.
Forecasting would continue to depend heavily on manual work and individual knowledge.
Leadership would still be looking predominantly at what had already happened.
The data was there. The opportunity was to start using it differently.
Our approach
We started with the questions leadership needed to answer
Before building a predictive model, Sweet Potato Tec worked with executives and operational teams to understand how forecasting was currently carried out and where confidence began to break down. What information did leadership rely upon? How were renewals treated? Which data could be trusted? Where were the gaps?
Those conversations helped define the business requirements before the technical work began. From there, the objective became clear: build forecasting around the way the organisation actually sells and renews.
1
Building a reliable data foundation
Predictive analytics isn’t particularly useful if the underlying data can’t be trusted, so a significant part of the project happened before any predictions were made. We used CRM Analytics recipes to bring together data from the Salesforce Opportunity, Product, Account and User objects. The data was then prepared and transformed, with irrelevant records removed, gaps addressed and renewal activity clearly distinguished from net-new opportunities. The work also surfaced duplicate records, missing product mappings and inconsistent renewal flags that were weakening existing reporting.
2
Looking 48 months ahead
With the data in better shape, we built a predictive forecasting model using Einstein Discovery and Salesforce CRM Analytics. Rather than forecasting only at company level, the model predicts revenue at account and product level, covering both renewals and new business. The resulting 48-month forecasting application gives a rolling view of predicted future revenue that moves forward automatically as time progresses, with interactive dashboards to compare predicted against actual performance and drill into account, region and product line.
3
One forecast for Sales, Finance and Operations
One of the most important changes wasn’t technical — it was organisational. Sales, Finance and Operations now work from the same predictive forecast rather than maintaining separate interpretations of future revenue. Renewals are clearly identified, data quality issues are visible, and predicted and actual revenue can be compared. Finance can export what it needs into CSV or Excel, while scheduled prediction refreshes keep the model current.
Scope of work
What Sweet Potato Tec delivered
Spreadsheet-based forecasting, inconsistent renewal visibility, fragmented Salesforce data and data-quality issues were making it difficult for leadership to build a dependable long-term view of revenue. The work addressed each of them.
- 48-month predictive forecasting model
- Einstein Discovery predictive modelling
- Unified Opportunity, Product, Account and User data
- Predicted versus actual revenue reporting
- Data-quality monitoring and governance
- Salesforce CRM Analytics implementation
- Revenue Intelligence dashboards
- Renewal and net-new revenue classification
- Account, region and product-level analysi
- Scheduled prediction refreshes and model retraining framework
Key integrations
CSV and Excel exports for Finance
Scheduled prediction refreshes
Salesforce capabilities
Sales Cloud
Revenue Intelligence
CRM Analytics
Einstein Discovery
Data Cloud
Agentforce
The outcome
The difference it made
The company now has a much clearer view of its future revenue position. Forecasting is more accurate and forward-looking, renewals can be separated clearly from new business, and Sales, Finance and Operations are better aligned because they’re working from one predictive forecast.
Perhaps most importantly, the model wasn’t built to answer a question once. It was designed to keep moving forward.
- Improved forecasting accuracy
- Greater confidence in revenue data
- Clearer visibility of renewals versus new business
- One predictive forecast across Sales, Finance and Operations
- Long-range revenue visibility
- Scenario modelling for leadership
- A forecasting model designed to keep rolling forward into future years
“Before this project, forecasting was manual, inconsistent, and reactive. Bringing predictive intelligence together with Salesforce CRM Analytics let us anticipate revenue patterns months ahead, trust our data and align teams around one forecast — it has prepared us not just for this year, but for the future.”
VP of Sales Operations — Global Software Company
What happened next
Keeping the forecast useful
Predictive models need looking after. Businesses change, buying behaviour changes and the data behind a model changes with them.
Sweet Potato Tec built governance and ongoing management into the project, including scheduled prediction refreshes, data-quality monitoring, documentation and a process for regular model retraining.
That gives the business a forecasting capability that can continue evolving alongside Salesforce, Revenue Intelligence and newer AI capabilities. The objective wasn’t to create an impressive dashboard — it was to build something leadership could continue using to make better decisions.
What could your Salesforce data tell you about the next 48 months?
If forecasting still depends on spreadsheets, manual reporting or a handful of people who understand how the numbers fit together, there may be far more value sitting inside your Salesforce data than you’re currently able to use.
Sweet Potato Tec helps organisations use Salesforce CRM Analytics, Revenue Intelligence and predictive modelling to turn existing data into clearer, more useful commercial insight.
Let us make your technology work harder for your organisation
Tell us what you are trying to improve, where your current systems are holding you back and what success would look like.
We will arrange an initial conversation to explore your requirements and determine whether Sweet Potato Tec is the right partner to help.