Consumer goods: the full guide

A consumer brand sells through a chain. It makes or buys the product, sells to a distributor or straight to a retailer, and the retailer sells to the shopper. Each step takes its share, and each step holds information the others want. Beer, wine and spirits sell through a legally required distributor tier and have their own guide.

Four pools of money decide how a brand's year turns out.

Trade spend is usually the largest line after the cost of goods. Industry surveys put it between 11% and 27% or more of revenue, and estimate that a quarter to more than two thirds of it does not pay back. For a $100 million brand, that is roughly $11 million to $27 million a year, much of it spent on promotions the brand cannot fully measure.

Pricing power sits with whoever holds the best view of demand. For decades that was the retailer, through its point-of-sale data. Brands bought that view back through data-sharing programs and syndicated data. The rest of this guide is about how AI widens or narrows that gap.

Two cautions apply to every row.

Capacity is not cost. A sales team that spends more time selling has more capacity, not lower cost. That capacity is worth money only if it produces more orders, or if the brand needs fewer people or fewer brokers. A proposal that reports hours saved has not yet shown a saving.

Vendor figures are best cases. Every number in this table comes from the company that sold or used the tool, measured its own way. They show what is possible, not what a mid-sized brand should expect. The section on where to invest shows how to test a claim before paying for it.

Two patterns stand out. The companies with results are the ones that own the ordering channel or the largest budgets. And the evidence is almost all self-reported. A brand buying a tool on the strength of a case study should assume it is looking at the vendor's best customer.

Scenario 1: The reorder decides itself. Distributor and retailer platforms move from suggesting orders to placing routine ones. Brands with steady, well-recorded sales get reordered automatically; new or seasonal brands need a person to argue for them. Assumes: stores accept automated reorders for staple items. Signal to watch: a major distributor reports that most reorders at small accounts are placed without a sales rep. Last checked: September 2026. Mondelez says its AI order suggestions already reach more than a million stores.

Scenario 2: Shoppers ask an AI assistant instead of browsing. A shopper asks for a gluten-free snack for a kids' party under $10 and an assistant picks. Brand choice shifts from the shelf and the search page to whatever the assistant can read about the product: its listing, its reviews, its ingredients. Assumes: shoppers trust assistants with routine purchases. Signal to watch: a large retailer reports meaningful sales through an assistant or a chat app. Last checked: September 2026. Walmart announced shopping through ChatGPT in October 2025.

Scenario 3: Retailers sell AI recommendations, not just data. Retail data platforms move from reports to paid recommendations on pricing, promotion and assortment. Brands that pay get advice from the retailer's model; brands that do not are judged by it anyway. Assumes: retailers price these tools within reach of mid-sized brands. Signal to watch: a retailer publishes supplier tiers or prices for AI recommendations. Last checked: September 2026. Walmart has described this direction for Scintilla.

None of these requires a brand to act this quarter. All three reward the same preparation: clean sales data by account, accurate product information everywhere a machine reads it, and a clear view of which promotions pay.

Retailers and large distributors gain. They see every brand's sales in their stores and increasingly decide reorders through their own software. The more routine ordering they automate, the less a brand's sales call counts on staple items.

The largest brands gain. They can afford data teams, digital copies of their products and paid retailer tools. They also supply the steady, high-volume sales history that ordering models favor.

Mid-sized brands are squeezed in the middle. Too big to be a local favorite a store owner orders by hand, too small to pay for every retailer tool. Their defense is narrow: accurate data on their own sales by account, fast product information, and promotions they can prove pay.

Small and craft brands can gain. Cheap content and direct-to-consumer tools let a small brand look and sell like a bigger one. In specialty and independent stores, buyers still want what is different, and recommendation systems built on best-sellers work against that.

Agencies and brokers face pressure. When content costs fall and routine orders place themselves, the hours agencies and brokers bill for shrink. Expect them to reprice or rebundle their services, and use that moment to renegotiate.

The cost ranges are illustrative and will be replaced with reviewed figures. The pattern holds regardless: the tool is rarely what fails. The data under it, or the person meant to act on it, usually is.

One test before any contract. Ask the vendor to run the tool on one season of the brand's own past data and show what it would have predicted against what actually happened. A vendor confident in its product will do this. One that will not is telling you something.

Keep the rules that do not change. The Federal Trade Commission has said there is no AI exemption from the laws on the books. A health claim, a fake review or a misleading "made with" statement is the brand's problem whether a person or a model wrote it. Check every AI-made claim against the rules of each market you sell in.

Your customers: distributors. As routine reorders move onto distributor platforms, a brand's standing at a distributor depends on what its system can see: sales history, promotion calendar and product data. Read the wholesale distribution guide for how distributors are changing their own economics.

Your customers: retailers, restaurants and cafes. Retailers are furthest ahead and set the terms through their data platforms. Food service buyers change more slowly and still reward relationships and what is different. Read the food service and hospitality guide.

Your suppliers: co-packers, packaging and ingredients. Suppliers using AI for planning can quote faster and hold less stock, which can shorten lead times. It can also mean less slack when a brand's forecast is wrong. Read the manufacturing guide.