Design win pipeline forecast
Track design-in opportunities by stage, probability, volume and price, and forecast the revenue they bring year by year.
AI, systems and startupFor membersHow it worksSourcesWhere it helps
Register or sign in to use this tool
The working tool, its inputs and its results are for registered members. Registration is free and takes a minute. You come back to this page afterwards.
How it works
A design win is a customer's decision to use your part in a product. It becomes revenue only after the customer builds, qualifies and ramps that product, often years later, and many opportunities never get there. This tracker holds your pipeline of design-in opportunities by stage, turns each into revenue year by year from its volume, ramp and price, weights it by the chance that opportunities at its stage convert, and shows the total by year and by stage, the opportunities that have gone quiet, and what the forecast would be if the probabilities were different.
- Record each opportunity. For each opportunity enter the customer, application and part, the stage, the expected production start date, the annual volume at the peak, the years to ramp up, the years at the peak, the starting price per unit and the price erosion a year, plus the competitor, owner, next step and its date. The stages are identified, engaged, sampling, design win, in production and lost.
- Turn volume into revenue by year. From the start date, volume rises in a straight line from zero to the peak over the ramp years, holds for the peak years, and then stops. If you turn on decline, it instead falls in a straight line from the peak to zero over the decline years. The price starts at the launch price and falls by the erosion share on each anniversary of production start. Each calendar year is cut at every change in slope and at every price step, and each piece is the average volume times its length times its price.
volume(t) = peak x t / ramp for t < ramp; peak while t < ramp + hold; then 0, or peak x (1 − (t − ramp − hold) / decline) with decline on; price(t) = ASP x (1 − erosion)^floor(t); revenue = sum of volume x price over each piece - Weight by stage. Each stage has a probability that an opportunity at that stage ends up shipping. The defaults are 10% identified, 25% engaged, 50% sampling, 75% design win, 90% in production and 0% lost. They are editorial figures that rise with the stage; replace them with your own win rates from past opportunities. Weighted revenue is the unweighted revenue times the stage probability.
weighted revenue = revenue x probability of the stage - Total by year and by stage. The forecast starts in the calendar year of today's date and runs for the horizon you set, up to 15 years. The line chart shows three lines by year: weighted revenue, revenue if every open opportunity were won (lost ones left out), and revenue from design wins and parts in production only. The bar chart shows the weighted revenue over the horizon for each stage, with the count and probability beside it.
- Find the stale ones. An opportunity that is not lost is stale when its next step date is before today. The tracker lists them with the days overdue, and warns about open opportunities with no next step or no owner, and about opportunities whose production start date has passed before they reached design win.
days overdue = today − next step date, when positive - Try other probabilities. The what-if table multiplies every stage probability by 0.5, 0.75, 1 or 1.25 (never above 100%) and shows the weighted revenue over the horizon each time, so you can see how much of the forecast rests on the probabilities you chose.
weighted revenue (factor) = sum over opportunities of revenue x min(1, probability x factor)
Assumptions
- The stage probabilities are editorial defaults, not industry statistics. Public sources give timelines but no standard conversion rates; use your own history.
- Volume ramps and declines in straight lines, and the price steps down once a year from the production start date. Real programmes have bumps, end-of-life buys and mid-year price deals.
- A date is read as the first day of ramp. Revenue is booked as units shipped in the calendar year; the time between shipment and invoice is ignored.
- Probabilities apply to the whole opportunity. A programme won in part, or shared with another supplier, should be entered at your share of its volume.
- The example pipelines are invented and describe no real customer.
Limits and warnings
- A forecast from your own entries is only as good as the dates, volumes and probabilities in them. Customers' volume forecasts are usually optimistic.
- Revenue after the horizon is not counted, and no discounting or cost is applied. This is revenue, not profit or value.
- One price per opportunity. Tiered pricing, rebates, last-time buys and currency changes are not modelled.
- Wins that share a customer's volume with a second source, or products that cannibalise each other, are not linked; adjust the volumes by hand.
Sources
The methods and default figures rest on these public sources. The wording here is our own; read the source for its full text.
- Valens Semiconductor Ltd., Annual Report on Form 20-F (opens in a new tab), US Securities and Exchange Commission (EDGAR), for the year ended 31 December 2024. A public company's own account of how a design win becomes revenue. In automotive it describes the customer's evaluation and decision at about 12 to 18 months, development at about 24 to 30 months, and revenue over a product life of about 5 to 7 years once in mass production. Used for the timing and the shape of the funnel, not for any probability. One company's figures: a guide to scale only.
- Editorial stage probabilities, The Semiconductor Hub, October 2026. The stage probabilities in this tool are the Hub's own editorial defaults. No public source gives standard conversion rates by stage, so replace them with the rates from your own past opportunities.
Where it helps
The lifecycle stages this tool serves: 3 stages in 2 lifecycles.
More ai, systems and startup tools
Terms on this page
- Design-in
- The engineering engagement period during which a customer evaluates, prototypes and integrates a part, often while a competitive selection is still under way.
- Design win
- Confirmation that a customer has selected a specific component for a production programme.
- Pipeline
- In commercial jargon, the set of identified sales or design-in opportunities at various stages of maturity, weighted by probability.
- Price erosion
- The structural annual decline in the price of a given part as it ages, competitors qualify alternatives and customers negotiate reductions.
- Second source
- An alternative qualified supplier of a functionally equivalent part, kept available to protect against supply interruption or price leverage.
Tell us what you're after and leave your email: we'll connect you with the right person on the Hub.