Part of The Manufacturer's Complete Guide to Selling Automotive Products to US Retail, the operator's playbook covering retailer landscape, line review, ACES/PIES, EDI, slotting, packaging, and launch sequence.
A forecast a buyer will sign off on is built bottom up: store count times units per store per week times 52 weeks, anchored to a named comparable item already on the shelf. Top down market share math gets discounted on sight. Buyers fund unit rates, not addressable market.
Every line review packet contains a forecast, and it is the slide buyers trust least. A line review forecast, meaning the projected unit sales of your item over its first 52 weeks at a defined store count, is the only number in your submission that you control entirely. The buyer knows that. Their job is not to admire the number. Their job is to decide whether it is safe enough to build a purchase order, a distribution center allocation, and a planogram around.
Why buyers discount your forecast before they read it
A category manager running a full category reset will read somewhere between 15 and 40 supplier forecasts in a single review window. In the aggregate, those forecasts always exceed the category. The category is growing 3 percent and the assembled supplier projections add up to 40 percent growth. The buyer has seen this every cycle of their career, so they apply a standing haircut before they evaluate anything specific about your line.
The haircut is not arbitrary. The buyer compares your projected rate against the actual velocity of the item your SKU would displace, because that item's point of sale history is sitting in their system. If your forecast implies three times the velocity of the incumbent and your packet does not explain the mechanism driving that gap, the credibility damage does not stop at the forecast slide. It spreads to your pricing, your margin math, and your service claims. The Auto SKUS Group has watched otherwise strong submissions lose on this alone: not because the product was wrong, but because one unsupported number made the buyer distrust the rest of the packet.
The only forecast math a buyer actually checks
Buyers do not audit your model. They audit one equation:
Store count times units per store per week times 52 weeks equals annual units.
Work it in front of them. A 1,200 store cluster at 0.8 units per store per week produces 49,920 annual units. At a net delivered cost of 4.10 dollars, that is roughly 204,700 dollars in annual wholesale volume. Now the conversation is about whether 0.8 is defensible, which is the conversation you want, instead of about whether your growth assumptions are honest, which is the conversation you lose.
Know where your rate falls on the retailer's internal scale. In most hard parts and accessory categories, an item running under 0.3 units per store per week is a cut candidate at the next reset. Something between 0.5 and 1.0 is a healthy midshelf performer. Above 2.0 you are an anchor item and the buyer will protect your facings. Presenting a 4.0 rate for a first year item without a documented test behind it tells the buyer you have never sat on their side of the table.
Then add the number most suppliers forget: pipeline fill. That same 1,200 store cluster with a case pack of six units requires 7,200 units on the shelf before a single unit sells, plus roughly four weeks of distribution center cover. Your forecast is not the number you need to produce. It is the number you need to produce after you have already funded the fill.
Anchor the rate to a comparable, not to the market
The strongest forecast defense is a named analog. Identify an item already on that retailer's planogram in the same price band, the same pack configuration, and the same aisle position, and state your rate as a ratio to it. "We are forecasting at 85 percent of the rate of the item in the middle facing" is a claim the buyer can verify in 30 seconds using data they already own. "We are forecasting 2 percent of a 1.4 billion dollar category" is a claim they cannot verify and will not try to.
Say where your rate came from. There are only four credible sources: your own point of sale data from another national chain, a documented in store test, syndicated market data at the item level, or the retailer's own history on a comparable item. If your rate comes from a club channel or from ecommerce, adjust it down and say so out loud. Basket behavior, pack sizes, and traffic patterns do not transfer cleanly between channels, and a buyer who catches you importing a club velocity into a hard parts forecast will treat every other number the same way. This is the same discipline that governs the rest of your line review submission.
Present one number, commit to a different one
The forecast in your deck drives the planogram and the initial purchase order. The number you plan capacity against should be higher. Build supply to roughly 1.3 times the presented forecast for the first 26 weeks.
The reason is asymmetry. Overforecasting costs you credibility at the next review, which is recoverable. Underdelivering on supply costs you fill rate, triggers chargebacks, and puts a new item on a service exception list inside its first quarter, which is frequently not recoverable. Programs die from the inability to ship upside far more often than they die from a soft first year. If your item outperforms and you cannot cover it, the buyer experiences your success as a failure.
What to do when you have no velocity history at all
Do not manufacture a rate. Change the ask. Propose a test cluster of 250 to 400 stores with a defined read window of 13 weeks, commit to a specific units per store per week threshold in writing, and name what happens if you miss it, including who funds the markdown and who takes the returns. A supplier who volunteers the failure condition is telling the buyer the number is real. That posture converts more first meetings than an aggressive projection ever has, and it is the shape most successful entries into national automotive retail actually take.
How do you forecast units for a line review?
Multiply store count by units per store per week by 52 weeks. Anchor the weekly rate to a named comparable item on the retailer's current planogram and state your data source. Add pipeline fill separately: it is not part of the sales forecast.
What is a good units per store per week number in automotive retail?
Under 0.3 units per store per week is generally a cut candidate at the next reset. Between 0.5 and 1.0 is a solid midshelf performer. Above 2.0 is anchor item territory. Rates vary by category, price band, and purchase frequency.
Should I forecast at full chain distribution or a test cluster?
Forecast at the distribution you are actually requesting. If you have no retail velocity history, request a 250 to 400 store test cluster with a 13 week read window rather than projecting chain wide numbers you cannot support.
How much inventory do I need before the first purchase order ships?
Pipeline fill equals store count times case pack, plus roughly four weeks of distribution center cover. A 1,200 store cluster on a six unit case pack needs about 7,200 units on shelf before any sales occur, funded ahead of revenue.
What happens if I miss my line review forecast?
Missing low is survivable if you flag it early and bring a corrective plan. Missing high without supply is worse: it drives fill rate failures and chargebacks. Buyers penalize surprises more than variance, so report against forecast quarterly rather than waiting for the review.