GRR vs NRR: How Retention Metrics Impact Forecast Accuracy

Revenue growth used to be easy to explain. Sales teams closed new deals, and finance tracked total revenue. However, in subscription and contract-driven businesses, this approach no longer works.

Today, most revenue comes from existing customers. In fact, 73% of companies now prioritize growth from their current customer base, according to Gartner-backed research. As a result, retention directly affects forecast accuracy.

For example, a pharmaceutical distributor may report stable growth based on signed contracts. However, if key customers reduce their order volumes or delay renewals, future revenue will drop. If finance does not track these changes early, forecasts become too optimistic.

This is where retention metrics become essential:

• GRR (Gross Revenue Retention) shows how much revenue you keep from existing customers, excluding expansion
• NRR (Net Revenue Retention) shows whether your existing customer base is growing, including upsells and cross-sells

Both metrics answer different questions. GRR shows how stable your revenue base is. NRR shows whether that base is expanding or shrinking.

Read more: What Great Financial Reporting and Analytics Actually Look Like

The problem is that many companies track only one of them. As a result, they either miss churn or overestimate growth. This leads to a common issue in finance teams: unclear and inconsistent KPIs. When teams rely only on top-line revenue, they miss early signs of decline. Over time, this reduces forecast accuracy, especially in companies with complex operations and fragmented data.

In the next sections, we break down GRR and NRR and show how to use them in financial planning.

Gross Revenue Retention (GRR) measures how much revenue you keep from existing customers over a given period. It focuses only on revenue loss, so it excludes any upsells or expansions.

GRR answers one question: Are we keeping the revenue we already have?

You calculate GRR by starting with your existing revenue base. Then, you subtract revenue lost from churn and downgrades. Finally, you divide that number by the original revenue.

• Includes: churn and downgrades
• Excludes: upsells, cross-sells, and expansions

Because of this, GRR gives a clear view of revenue stability. It shows whether your customer base is shrinking, even when total revenue appears stable or growing.

Read: Annual Recurring Revenue vs Revenue: How Each Metric Impacts Financial Forecasts

For example, consider a manufacturing company that offers maintenance contracts. At the start of the year, it had €10M in contract revenue. During the year, it loses €1M due to cancellations and reduced contract scope.

In this case, GRR is 90%. This means the company retained 90% of its existing revenue, regardless of any new sales or expansions. At first glance, this may not seem critical. However, it means 10% of the revenue base is eroding. If this continues, future revenue will decline, even if new sales remain strong.

This is why GRR matters for planning. It highlights revenue risk early, before it appears in top-line results.

In practice, many companies miss this signal. They focus on total revenue and assume stability. However, without isolating churn and downgrades, they cannot see what is really happening in the existing customer base.

This issue becomes more severe in companies with fragmented data and manual tracking. Finance teams often rely on spreadsheets, which makes it difficult to accurately measure revenue loss accurately.

As a result, GRR becomes one of the most reliable indicators of hidden revenue risk.

Net Revenue Retention (NRR) measures how much revenue you keep and grow from existing customers over a given period. Unlike GRR, it includes both losses and gains. Therefore, it gives a more complete view of how your customer base performs.

NRR answers a different question: Are we growing revenue from the customers we already have?

You calculate NRR by starting with your existing revenue. Then, you subtract churn and downgrades. After that, you add any upsells, cross-sells, or price increases. Finally, you divide that number by the original revenue.

• Includes: churn, downgrades, upsells, and expansions
• Reflects: both revenue risk and growth within your existing customer base

Because of this, NRR shows whether expansion revenue can offset losses.

Read: What Is Revenue vs. Marginal Revenue? A Simple Guide for Finance Professionals

For example, consider a distribution company with €10M in existing customer revenue. During the year, it loses €1M due to churn and downgrades. However, it generates €2M in upsell revenue across regions and product lines.

In this case, NRR is 110%. This means the company has replaced lost revenue and grown its existing customer base. At first glance, this looks strong. However, it can hide underlying issues. The company is still  losing customers or reducing contract value. Growth comes from expansion, not from a stable base. If expansion slows down, total revenue will drop.

This is why NRR needs context. On its own, it does not show whether the underlying customer base is healthy.

In practice, many companies focus on NRR because it reflects growth. However, without understanding what drives that growth, they risk overestimating future performance.

This problem becomes more visible when data sits across CRM, ERP, and reporting tools. Inconsistent definitions and delayed insights make it difficult to separate churn from expansion.

As a result, NRR is a powerful metric, but only when used together with GRR.

GRR and NRR start from the same revenue base, but they answer different questions. Looking at only one of them creates an incomplete picture.

Because of this, you should always analyze both metrics together.

• High GRR + High NRR → Stable revenue with strong expansion
• High GRR + Low NRR → Stable base, but limited growth
• Low GRR + High NRR → Growth is masking churn
• Low GRR + Low NRR → Revenue base is declining

The third case is the most risky.

For example, a telecom company may expand its contracts with large clients while losing contracts with smaller ones. NRR stays above 100%, but GRR declines. Over time, the customer base becomes weaker.

At this point, the company is not truly growing. It is replacing lost revenue with expansion.

This is why both metrics matter for planning. GRR highlights risk, while NRR shows growth potential. Together, they provide a more accurate view of future revenue.

Retention metrics often sit with sales or customer success. However, they directly shape financial outcomes. Because of that, finance teams need to treat GRR and NRR as core planning inputs.

Forecast accuracy depends on what happens to existing revenue, not just new sales. According to Gartner, most organizations already prioritize growth from existing customers. However, many still fail to model retention correctly.

When finance teams ignore churn and downgrades, they assume revenue will continue as planned. In reality, even small changes in customer behavior reduce future revenue.

For example, a distribution company may lose key customers while expanding others. Total revenue still grows, but the underlying risk increases. Without separating GRR and NRR, finance cannot detect this early.

As a result, forecasts become systematically too optimistic.

Retention directly affects demand, which drives cost planning. According to PwC, inaccurate demand assumptions are one of the main causes of budget variance.

When retention is not tracked properly, finance teams base budgets on expected volumes that may not materialize.

For instance, a manufacturing company may plan production based on contract expectations. However, if customers reduce orders or delay renewals, the company overproduces. This leads to excess inventory and margin pressure.

Because of this, retention metrics help align budgets with actual demand, not assumptions.

Cash flow stability comes from existing customers, not new deals. KPMG highlights that predictable revenue streams are a key factor in financial resilience.

Companies with strong GRR can rely on recurring inflows. In contrast, companies with weak retention depend on new sales, which are less predictable and more sensitive to market conditions.

As a result, GRR often provides an earlier and more reliable signal of cash flow risk than top-line revenue.

In many companies, retention data sits across CRM, ERP, and reporting tools. This fragmentation limits visibility and delays decision-making.

For example, a pharmaceutical company may track contracts in one system and invoicing in another. Finance then consolidates the data manually. By the time churn appears in reports, it is too late to react.

Because of this, retention issues often surface only after they impact financial results.

Read: Enterprise Reporting: Why It Fails and How to Fix It

Unclear KPI definitions create misalignment across teams. Sales, finance, and controlling often calculate retention differently, leading to conflicting reports.

Aside from the fact that this is a big reporting issue, it also affects decisions. Leadership may rely on metrics that do not reflect actual performance.

Over time, this creates a gap between reported growth and real underlying trends.

Read: 15 Key Metrics to Track as KPI for Finance Department

GRR and NRR serve different purposes:

• GRR shows risk by highlighting revenue loss from existing customers
• NRR shows growth by capturing expansion within the current base

Looking at only one of these metrics creates blind spots.

A company can grow while losing customers. It can hit revenue targets while weakening its base. This is why NRR alone is not enough, and why GRR needs equal attention.

When used together, these metrics provide a complete view of revenue performance. They help finance teams detect risk earlier, improve forecast accuracy, and make more reliable decisions.

However, this only works if retention is part of the planning process. If GRR and NRR are tracked only for reporting, their value remains limited.

If retention is not built into your planning model, your forecast is incomplete.