GRR vs NRR: Which Retention Metric Investors Weigh

What Is Driver-Based Revenue Forecasting?

A revenue forecast is a projection of the money your business will earn over a defined future period. There are two ways to build one:

Top-down forecasting starts with the total addressable market and works down to a market share assumption: “The UK B2B software market is worth £10 billion. If we capture 0.1%, we generate £10 million in revenue.” Useful for sizing the opportunity, useless for operational planning. Investors have heard thousands of 0.1% market share projections and are rightly sceptical.

Bottom-up, driver-based forecasting starts with the specific activities that generate revenue: “We have capacity to run 20 outbound sales conversations per week. Our conversion rate is 10%. Our average contract value is £12,000 per year. That gives us 2 new customers per week, or roughly 100 new customers per year, generating £1.2 million in new ARR.” Every assumption in that chain is testable, improvable, and explainable.

Driver-based forecasting is also the input layer for your 3-statement model — your revenue drivers feed the income statement, which integrates with the balance sheet and cash flow statement.

Why a Revenue Forecast Startup Needs a Different Approach

Established businesses forecast revenue by extrapolating historical data. Startups do not have historical data. The entire forecast must be built on forward-looking assumptions rather than trend lines. A driver-based model built on transparent assumptions is actually more useful to an early-stage investor than a statistical extrapolation, because it makes the business logic explicit and discussable.

The Core Framework: Identify Your Revenue Drivers

What Is the Difference Between GRR and NRR?

The direct answer: GRR excludes expansion revenue and NRR includes it. Both metrics start with the same cohort, the recurring revenue from customers you already had at the start of the period, and measure what that cohort is worth at the end of the period.

GRR formula: (Starting MRR - churned MRR - contraction MRR) / Starting MRR. Expansion is ignored entirely, so the maximum possible GRR is 100%.

NRR formula: (Starting MRR - churned MRR - contraction MRR + expansion MRR) / Starting MRR. Because expansion is added back, NRR above 100% means your existing customers are worth more this year than last year even before you sign a single new logo.

Both calculations exclude new customer revenue. That is the whole point: retention metrics isolate the health of the base you already own. If you are still deciding whether to report in monthly or annual terms, our MRR vs ARR guide covers which denomination fits your motion.

Why Do Investors Look at Both Metrics?

Investors weigh GRR and NRR together because each one can hide a problem the other exposes.

A company reporting NRR of 115% looks healthy on the surface. But if that same company has GRR of 78%, the picture changes completely: it is losing more than a fifth of its base revenue every year and papering over the hole with aggressive upsells to the accounts that stay. That is a fragile structure. Expansion revenue tends to concentrate in a small number of large accounts, so if one or two of them churn or stop expanding, NRR collapses toward the weak GRR underneath.

The reverse pattern is more forgivable. GRR of 93% with NRR of 101% describes a sticky product that has not yet built an expansion motion. Investors generally treat that as an execution gap (fixable with pricing tiers, seat expansion, or new modules) rather than a product problem. Weak GRR is a product or ICP problem, and those are much harder to fix with go-to-market changes. Our post on logo churn vs revenue churn explains the related distinction on the churn side of the same coin.

What Are Realistic GRR and NRR Benchmarks?

Benchmarks vary heavily by customer segment, because SMB customers churn more than enterprises regardless of how good the product is.

Segment	Healthy GRR	Healthy NRR	Best-in-class NRR
SMB (low ACV)	80-85%	95-105%	110%+
Mid-market	85-90%	100-115%	115-120%
Enterprise (high ACV)	90-95%+	110-120%	120-130%+

[Source ranges: widely published survey data from OpenView, KeyBanc, and public SaaS filings -- verify exact figures before publish]

The widely cited rule of thumb is that best-in-class SaaS companies post NRR of 120% or higher, and public market darlings like Snowflake at IPO reported NRR well above 150%. But those are enterprise, usage-based outliers. Judging a $2M ARR SMB tool against Snowflake's NRR is a category error, which is why sophisticated investors always ask for your segment and ACV before reacting to the number.

How Do GRR and NRR Change What Your Model Projects?

Retention assumptions are among the highest-leverage inputs in a startup financial model. A one-point difference in monthly revenue retention compounds dramatically over a three-year projection.

Consider two companies each starting at $1M ARR with identical new-business plans. Company A holds NRR at 100%; Company B holds NRR at 120%. After three years, Company B's existing base alone has grown roughly 73% (1.2 cubed) while Company A's base is flat. Company B's growth stacks new logos on top of a compounding base; Company A must replace nothing but also gains nothing. This is why investors describe strong NRR as "growth you do not have to pay for," and why it directly improves metrics like the SaaS magic number and your Rule of 40 score: the same sales spend produces more net new ARR when the base is expanding underneath it.

When you build retention into your model, model GRR and expansion separately rather than plugging in a single NRR assumption. A single NRR line hides the mix, and diligence teams will unbundle it anyway.

Which Metric Should Founders Lead With at Each Stage?

Pre-seed and seed. You likely do not have enough customer-months of data for either metric to be statistically meaningful. Report what you have honestly, cohort by cohort, and flag the small sample size yourself before investors do.

Series A. Lead with GRR if it is strong, because at this stage investors are underwriting product-market fit and stickiness more than expansion. An NRR above 100% is a bonus; a GRR below 80% will dominate the conversation regardless of what NRR says.

Series B and beyond. NRR becomes the headline number because investors are now underwriting efficient growth at scale. But expect diligence to decompose NRR into its GRR and expansion components, by cohort and by segment, so have that breakdown ready in your data room.

What Mistakes Make Retention Metrics Untrustworthy?

Three errors show up constantly in diligence and each one damages credibility:

1. Mixing cohort windows. NRR should compare the same customer cohort at two points in time, typically twelve months apart. Blending in customers acquired mid-period inflates the number.
2. Counting one-time revenue. Services, implementation fees, and one-off charges do not belong in a recurring revenue retention calculation.
3. Reporting monthly NRR annualized from one good month. Retention is noisy month to month, especially on a small base. Trailing twelve-month calculations are the standard investors expect.

How Do You Improve GRR and NRR Without Gaming Them?

The two metrics respond to different levers, which is another reason to track them separately.

GRR improves through churn and contraction prevention: tighter ICP qualification so the wrong customers never enter the base, structured onboarding that gets users to first value quickly, health scoring that flags at-risk accounts before renewal, and multi-year or annual terms that reduce decision points. None of these are fast, which is why GRR is treated as a product and ICP signal: it moves on quarters and years, not weeks.

NRR improves through expansion architecture: pricing that scales with a value metric (seats, usage, records) so customer success naturally becomes revenue growth, packaging with a clear upgrade path, and a defined expansion motion with owners and quota rather than passive hope that accounts grow. Companies frequently discover several points of latent NRR simply by instrumenting expansion opportunities that already exist in the base.

What does not work is gaming. Forced migrations to higher-priced plans spike expansion for a few quarters and then show up as elevated churn when the repriced customers hit renewal. Discount-heavy annual prepays defer churn rather than preventing it, flattering this year's GRR at the expense of next year's. Excluding "non-core" customers from the cohort produces a retention number diligence will recompute and reject. Investors have seen every one of these patterns; the durable move is to improve the inputs, report conservatively, and let the trend line do the persuading.

One practical sequencing note for early-stage teams: fix GRR before investing heavily in expansion. Expansion motion layered onto a leaky base produces the high-NRR-weak-GRR profile described above, and the work is largely wasted if the accounts being expanded churn eighteen months later. Retention first, expansion second is the ordering most post-mortems endorse.

Frequently Asked Questions

Can NRR be above 100% while GRR is falling? Yes, and this is exactly the pattern investors screen for. Rising expansion from a concentrated set of accounts can offset a deteriorating base for several quarters before the NRR headline finally cracks.

Is 100% GRR possible? Only if literally no customer churned or downgraded in the period, which happens on small bases over short windows but is not sustainable at scale. Sustained GRR in the mid-90s is considered elite for enterprise SaaS.

Should I calculate retention on revenue or logos? Both, because they answer different questions. Revenue retention weights your large accounts; logo retention treats every customer equally. Our churn rate guide walks through when each view matters.

Do usage-based businesses report NRR differently? The formula is the same, but usage-based NRR is more volatile because consumption can contract without a formal downgrade. Many usage-based companies report NRR alongside a committed-revenue retention figure for that reason.

What NRR do I need to raise a Series A? There is no hard gate, and many strong Series A companies have thin retention data. Directionally, NRR at or above 100% with GRR above 85% removes retention as an objection.

Model your metrics with Raise Ready's free financial model tool. Build GRR and NRR assumptions directly into your startup financial model and stress-test how retention changes your runway with the runway and burn calculator.

Further Reading

• What Is a Startup Financial Model?
• The Rule of 40 Explained
• Churn Rate Calculation and Cohort Analysis
• Runway and Burn Calculator
• Financial Model Pro