New product forecasts are exercises in structured reasoning, not historical extrapolation. You have no win rate history, no established sales cycle, and no pricing data from real deals. If you build a model the same way you would for an established product line, you will produce a number that looks precise and is probably wrong. The right approach is to be explicit about your assumptions, build from observable evidence, and present leadership with a range they can make decisions from.
Step 1: Identify What You Do Know
Even without historical data for the new product, you have signal worth using. Catalog it before touching a spreadsheet.
Early pipeline data. If you have run any discovery calls, demos, or pilot conversations for the new product, count them. Note how many accounts are in active conversation, what stage they are at, and whether any have moved to proposal or closed. Even a handful of early deals gives you cycle length and conversion signal. Existing product comparables. If the new product is adjacent to something you already sell, your existing win rates and cycle lengths are a starting point. Adjust for the fact that new products typically sell more slowly and convert at lower rates until the motion is proven. Compensation-set data. For products in established categories, public competitors' reported conversion rates, deal sizes, and sales cycle benchmarks provide a sanity-check range. These are imprecise but useful for bounding your assumptions. Committed accounts. Any account that has signed a pilot agreement, LOI, or verbal commitment is worth breaking out. These are not forecasted deals. They are near-certain revenue with known timing.Step 2: Build the Bottom-Up Pipeline View
Structure the forecast from your actual pipeline, not from a market-share calculation.
For each opportunity in the new product pipeline, assign:
- Estimated deal size. Use the pricing framework if established, or the size of early conversations as a proxy. - Estimated close quarter. Based on where the deal is in the conversation, not a wish. - Stage-based probability. Use your existing opportunity stages as the framework, but apply lower win rates than your established product line until you have data to justify higher ones.
If you have no named opportunities at all, your near-term forecast should reflect that honestly. A pipeline that does not exist yet cannot produce revenue in the current quarter.
Step 3: Apply Conservative Conversion Rate Assumptions
This is where new product forecasts most commonly fail. Teams apply their existing win rates to new product pipeline and produce an overstate.
New products face structural headwinds that established products do not:
- The sales team is still learning the pitch. - Buyers are evaluating a product without a reference customer base. - The competitive response is unpredictable. - Legal, security, and procurement reviews take longer for unfamiliar products.
For a product in its first two quarters of selling motion, apply a meaningful discount to your established win rates. Document the discount explicitly. As you close more deals, update the assumption with real data.
Step 4: Set Explicit Assumptions for Each Input
A new product forecast is only as trustworthy as its assumption log. For every key driver, write down:
- What the assumption is. - Why you chose it. - What would have to be true to change it.
| Input | Assumption | Rationale |
|---|---|---|
| Win rate (pipeline to close) | Lower than established product | No reference customers yet, longer eval cycle expected |
| Average deal size | Based on early pilot conversations | Adjust up as pricing crystallizes |
| Sales cycle length | Longer than established product | Buyer unfamiliarity adds evaluation time |
| Pipeline coverage needed | Higher than established product | Lower conversion means more pipeline per dollar of expected revenue |
Step 5: Build Three Scenarios
With your assumptions documented, run three versions of the model.
Conservative case. Lower win rates, longer cycle lengths, smaller deal sizes. Assume the deals that feel likely slip one quarter. This is the number the business can plan around without risk. Base case. Your most defensible assumptions given current evidence. This is what you are prepared to defend in a forecast call. Upside case. What happens if the early pilots convert, the sales motion tightens faster than expected, and one large deal closes in the period. This is not the plan. It is the ceiling.Present all three to leadership with the assumptions behind each. A point estimate for a first-year product is fiction. A range lets leadership plan around real uncertainty.
For the underlying forecasting frameworks used in this approach, see revenue forecasting and bottom-up forecasting. For how a top-down view can bound your scenarios, see top-down forecasting.
Common Mistakes
Anchoring on the TAM. Market size has no bearing on what your team will close in the next two quarters. Build from pipeline, not from market share math. Applying established win rates to new pipeline. Your existing conversion rate belongs to your existing product and selling motion. New products are not there yet. Presenting a single number. Leadership will ask how confident you are. A scenario band answers that question. A point estimate forces you to defend a guess. Ignoring cycle length. If your new product takes longer to sell than expected, deals that look like this quarter's revenue become next quarter's. Model cycle length explicitly, then plan around it.Frequently Asked Questions
Why is it hard to forecast a new product launch?
New products have no historical conversion rates, average selling prices, or cycle lengths to draw from. Without those inputs, standard forecast models return meaningless outputs. The solution is to replace missing historical data with structured assumptions sourced from comparable products, early pipeline signal, and external comp-set data, then present results as a range rather than a point estimate.What is a scenario band for a new product forecast?
A scenario band presents three versions of the forecast: a conservative case built on pessimistic assumptions about win rates and cycle length, a base case built on the most likely assumptions, and an upside case reflecting favorable conditions. The band gives leadership a range to plan around rather than a single number that will almost certainly be wrong.When should you use top-down vs. bottom-up forecasting for a new product?
Bottom-up is more reliable for a new product launch because it grounds the forecast in actual pipeline activity and named accounts. Top-down (market size times assumed share) tends to produce aspirational numbers that do not reflect what your sales team can actually execute in the near term. Use top-down for sanity checking the ceiling, not for building the operational forecast.Frequently Asked Questions
Why is it hard to forecast a new product launch?
New products have no historical conversion rates, average selling prices, or cycle lengths to draw from. Without those inputs, standard forecast models return meaningless outputs. The solution is to replace missing historical data with structured assumptions sourced from comparable products, early pipeline signal, and external comp-set data, then present results as a range rather than a point estimate.
What is a scenario band for a new product forecast?
A scenario band presents three versions of the forecast: a conservative case built on pessimistic assumptions about win rates and cycle length, a base case built on the most likely assumptions, and an upside case reflecting favorable conditions. The band gives leadership a range to plan around rather than a single number that will almost certainly be wrong.
When should you use top-down vs. bottom-up forecasting for a new product?
Bottom-up is more reliable for a new product launch because it grounds the forecast in actual pipeline activity and named accounts. Top-down (market size times assumed share) tends to produce aspirational numbers that do not reflect what your sales team can actually execute in the near term. Use top-down for sanity checking the ceiling, not for building the operational forecast.
See how ORM turns these insights into action
ORM builds custom revenue forecast models for B2B SaaS companies. Not dashboards. Prescriptive analytics that tell you what to do next.
Schedule a Demo