How to Forecast Sales for a New Product

Flying blind. That’s how it feels when forecasting sales for a new product. Without historical data to guide you, creating accurate projections feels more like guesswork than science.

Yet forecasting remains essential. Your company needs directional visibility to align go-to-market teams, plan resources, and set realistic expectations. The good news? Methodical approaches exist to forecast new product sales with reasonable accuracy, even without past performance data.

This guide will walk you through proven forecasting methods, when to use them, and how to implement them in your favorite spreadsheet tool.

Key factors to consider before forecasting

Start with solid ground. Before building any forecast model, gather these foundational inputs:

1. Market size: How many potential customers exist for your product? Industry reports, market research, and competitor analysis help quantify your total addressable market (TAM).
2. Comparable products: Have you launched similar products before? Do competitors offer similar solutions? Past performance of comparable offerings provides valuable benchmarks.
3. Marketing and sales efforts: Be realistic about your promotional muscle. Consider:
   • Marketing budget
   • Sales team capacity
   • Channel strategy
   • Launch timing
4. Pricing strategy: Your price point dramatically impacts adoption rates. Premium pricing may yield higher margins but slower adoption, while competitive pricing might accelerate growth at lower margins.
5. Economic conditions: Current market trends, economic outlook, and industry dynamics all influence buying behavior.

Gather these inputs first. They’ll form the foundation of any forecasting model you choose.

Step-by-step process to forecast sales for a new product

Definethe forecast period

Set your time horizon first. Most new product forecasts span:

• Monthly projections for the first year
• Quarterly for years 2-3
• Annual for years 4-5

Shorter periods improve accuracy but require more detailed work. Match your forecast period to your business planning cycle.

Estimate unit sales or conversion volume

Start with unit sales. This is typically easier to estimate than revenue because it removes price variability from the equation.

Consider:

• Industry standard conversion rates
• Sales team capacity and quotas
• Seasonality factors
• Market penetration goals

For SaaS products, focus on new accounts or licenses. For physical products, estimate units sold.

Apply pricing model

With unit estimates in place, apply your pricing structure. Account for:

• Tiered pricing
• Discounting strategies
• Bundle offers
• Promotional pricing

For subscription products, factor in annual versus monthly payment options and their impact on cash flow.

Adjust for ramp-up and sales cycle

New products rarely sell at full velocity immediately. Build a ramp-up curve that accounts for:

• Awareness building period
• Sales cycle length
• Adoption resistance
• Market education needs

A simple approach: start with 10-20% of steady-state sales in month one, then increase gradually until reaching full velocity.

Create best/likely/worst-case scenarios

One forecast isn’t enough. Build three:

1. Best case: Everything goes right; early adoption exceeds expectations
2. Likely case: Realistic projection based on most probable outcomes
3. Worst case: Significant headwinds, slower adoption than expected

This scenario planning provides flexibility when communicating with stakeholders.

Forecast revenue

Finally, calculate revenue by multiplying units by price across your forecast period.

For subscription products, include:

• New customer revenue
• Expansion revenue
• Churn impact

For one-time purchases, revenue typically mirrors unit sales more directly.

Models to consider for new product sales forecasting

Traditional forecasting relies heavily on historical data. Without it, consider these alternative approaches:

Top-down model

Start big, then narrow. Begin with your total addressable market (TAM), then apply filters to reach a realistic sales estimate:

TAM × Market Penetration Rate × Win Rate = Estimated Sales

For example: If your TAM is 10,000 companies, you might target 5% penetration in year one with a 20% win rate, yielding 100 new customers.

This approach works best for established markets with clear boundaries.

Bottom-up model

Start small, then build. Estimate sales based on your operational capacity:

Sales Reps × Leads per Rep × Conversion Rate = New Customers

For example: With 5 sales reps handling 50 leads monthly at a 10% close rate, expect 25 new customers monthly.

This model excels when your sales capacity constrains growth more than market opportunity.

Comparable product model

Borrow from experience. Use the adoption curve of similar products you’ve launched previously:

Similar Product First-Year Sales × Adjustment Factor = New Product Forecast

Adjustment factors account for differences in market conditions, product features, or pricing.

Adoption curve model

Apply established technology adoption patterns. The Bass diffusion model or Rogers’ adoption curve can predict how quickly innovations penetrate markets based on:

• Innovators (2.5%)
• Early adopters (13.5%)
• Early majority (34%)
• Late majority (34%)
• Laggards (16%)

Map your forecast onto these segments for a research-backed adoption timeline.

Weighted model combining multiple inputs

No single approach is perfect. Combine models for increased accuracy:

(Top-down × 40%) + (Bottom-up × 40%) + (Comparable × 20%) = Weighted Forecast

Adjust weights based on your confidence in each input method.

Post-launch sales forecasting

First data changes everything. After 2-3 months of real sales, your forecasting can shift from theoretical to empirical.

Linear regression for consistent trends

When early sales show steady growth, linear regression offers simplicity and clarity:

=FORECAST.LINEAR(future_date, known_y_values, known_x_values)

This Google Sheets function predicts future values based on existing linear relationships. Perfect for steady, predictable growth patterns.

Exponential smoothing for momentum

When recent data matters most, exponential smoothing gives more weight to recent observations:

=FORECAST.ETS(future_date, known_y_values, known_x_values)

This captures momentum and recent changes, ideal for products gaining traction rapidly.

ARIMA/ETS for seasonal patterns

If your early data suggests cyclical or seasonal behavior, more sophisticated time-series models help:

=FORECAST.ETS(future_date, known_y_values, known_x_values, seasonality_length)

Add the seasonality parameter to account for monthly, quarterly, or annual patterns.

Living models with Coefficient

Static forecasts quickly become outdated. Coefficient transforms your spreadsheet into a living model by:

1. Automatically syncing CRM data into your forecast model
2. Refreshing actual sales numbers daily or hourly
3. Recalculating projections based on the latest data

This eliminates manual exports and ensures decisions are made with the freshest data available.

For example, connect Salesforce opportunity data to your Google Sheet to automatically update your forecast as deals progress through your pipeline.

How to continuously track and update the forecast

Smart forecasting doesn’t end at launch. Establish a rhythm of review and refinement.

Actual vs. forecast tracking

Create a dashboard that compares predicted versus actual sales. Track key metrics:

• Unit sales variance
• Revenue variance
• Conversion rate reality
• Customer acquisition cost

This comparison helps identify where your assumptions were wrong and how to adjust.

Scenario-based dashboards

Don’t scrap your original scenarios. Instead, track performance against all three:

• Are you trending toward best case?
• Holding steady at likely case?
• Slipping toward worst case?

This context helps stakeholders understand performance in relation to expectations.

Coefficient-powered tracking

Manual updates create lag time between performance and visibility. Coefficient eliminates this gap by:

1. Syncing real-time sales data from your CRM
2. Automatically calculating variances against forecast
3. Sending Slack alerts when performance deviates significantly
4. Enabling self-service data access for stakeholders

This automation ensures everyone works from the same, current numbers without manual data wrangling.

New sales forecast tracker

Start faster with our pre-built template. It includes monthly projections, scenario planning, and variance tracking—all ready for your data.

Download the template and customize it for your new product launch.

Take the guesswork out of forecasting

Forecasting new products will always involve uncertainty. But methodical approaches reduce the guesswork.

Remember these principles:

• Start with multiple forecasting models
• Create scenario-based projections
• Update regularly with actual data
• Communicate confidence levels clearly
• Learn from variance to improve future forecasts

The best forecasts evolve. They start with educated estimates and improve with real data over time.

Ready to transform your static forecast into a dynamic decision-making tool? Get started with Coefficient today and keep your new product forecast continuously updated with live data.

FAQs

How to make a sales forecast for a new product?

Start with market size and penetration goals. Create a multi-model approach using both top-down (market-based) and bottom-up (capacity-based) methods. Develop best, likely, and worst-case scenarios to account for uncertainty. After launch, refine forecasts with actual sales data using regression or time-series models in your spreadsheet.

How to estimate sales of a new product?

Look to similar products in your portfolio or competitor offerings for benchmark data. Apply adoption curve models that follow established patterns for new product uptake. Factor in your marketing spend, sales capacity, and pricing strategy to adjust these benchmarks. Use weighted averages from multiple methods rather than relying on a single approach.

How do you forecast revenue for a new product?

Multiply unit sales projections by your pricing model, accounting for discounts, tiers, and payment terms. For subscription products, build cohort models that include new sales, expansion, and churn. Factor in seasonality and sales cycle length when distributing revenue across time periods. Use spreadsheet tools with Coefficient to automatically pull in pricing data from your CRM for accurate calculations.

How do you forecast sales for a new product with no history?

When history is absent, rely on market research, competitive analysis, and analogous products. Apply standard technology adoption curves modified for your specific industry and product type. Build multiple scenarios with clearly stated assumptions. As soon as initial sales data becomes available, use Coefficient to automatically feed this data into time-series models that improve accuracy with each new data point.