A route to market is the path a product takes to reach the buyer. It makes place, one of the 7Ps from Module 4, concrete: who sells your product, and who controls the sale.

The main routes.

Read the table as trade-offs, not a ranking. Control and data usually fall as reach rises, and fees and rules change, so check them at the source.

With direct to consumer you also carry the work of bringing people to your shop, which is the marketing you planned in Module 7. A marketplace gives you shoppers who are already looking, but takes a fee and puts you beside rivals. A retailer keeps a share of the shelf price, called the retailer margin: you trade some of the price for reach you couldn't afford alone.

UK example: Graze. Graze began as an online snack-box subscription business. In spring 2015 it ran a trial in Boots stores, and in July 2015, the trade press reported, its first shop range launched in Sainsbury's, Boots and WHSmith Travel. Unilever bought Graze in February 2019 and sold it to Katjes International in a deal announced in December 2025 and completed in February 2026. A business that started direct chose to add a route with different reach, control and economics.

Your organisation's routes. Most organisations use two or three routes and ignore the rest. Larkfield Coffee Roasters (fictional, with illustrative figures) sells through its own online shop, wholesale to cafés, and farm shops and delis. It doesn't use an online marketplace yet.

Whatever mix of routes you choose, your own online shop is the one you control. Run it well, because everything else is measured against it. This lesson is a quick audit of the basics.

Product pages. The page has to answer the questions a buyer would ask in a shop. Clear photos that show the real product, a plain description in the customer's words, the price including anything extra, and what's included. Put the main action, such as "Add to basket", where a thumb can reach it. Use the usability checks from Module 8: clear labels, one obvious next step and a layout that works on a phone first.

Checkout. Checkout is a journey like the sign-up you mapped in the UX topic, and it loses people in the same ways. Offer guest checkout, ask only for what you need, show the full cost before payment, and explain errors where they happen. The consent rules from Module 5 still apply here: no pre-ticked marketing boxes.

Delivery promises. Say when the order will arrive, what delivery costs and who carries it, before the customer pays. Then keep the promise. A delivery promise is part of your proposition; for Larkfield it is "at your door within three days of roasting". Late or vague delivery is one of the quickest ways to lose a repeat buyer.

Returns and refunds. Make the returns policy easy to find and written in plain English. In the UK, buyers who purchase online generally have a right to cancel within 14 days, with exceptions such as goods that deteriorate rapidly. The rules sit in the Consumer Contracts Regulations 2013, and the exceptions differ by product, so check what applies to yours. This isn't legal advice.

Customer service. Say how to reach a person and how fast you reply. Questions people ask often belong on the product page.

Genuine reviews. Reviews help buyers decide, and they are one of the reasons marketplaces work. You met the fake reviews rules in the regulation topic. Since 6 April 2025, the Digital Markets, Competition and Consumers Act 2024 bans fake reviews, including paying for reviews without saying so. The CMA published guidance on the ban in April 2025. The safe approach is simple: ask real customers for honest reviews, publish the bad ones as well as the good, and never write or buy reviews.

The one-journey check. Pick one product and buy it on your organisation's site, on a phone, as a new customer. Score it on what you find: the page, the checkout, the delivery promise, the returns policy, how to get help and the reviews.

Read the worked example later in this topic first: it works one figure through all three routes.

Revenue isn't profit. A sale through a marketplace or a retailer can look like growth while earning far less than the same sale through your own shop. The price the customer sees isn't what you keep. To compare routes you need the contribution per unit: the price the customer pays, minus the costs that come with each sale. Those costs are the product itself, the fee or margin the route takes, delivery and fulfilment, and payment fees.

Four questions for one sale.

1. What does the customer pay?
2. What does the product cost to make and pack?
3. What does the route take: a marketplace fee, a retailer margin, delivery or payment costs?
4. What's left?

What's left is the contribution. It goes towards costs you pay regardless of volume, such as salaries and rent, and then profit. Contribution margin is that contribution as a share of the price the customer pays. Module 12 builds on it with lifetime value.

Fees and margins you can't see. You often don't know a marketplace's fee for your category or a retailer's margin. Look up the marketplace's own seller fee page, or ask the retailer, and until you have it use a stated assumption, labelled as one. For retailers, work from the shelf price: if they keep 35% of £8.00, you receive the other 65%.

VAT. If your product carries VAT, take it out of the price first, because VAT goes to HMRC, not to you. Larkfield's roasted coffee is assumed to be zero-rated in the UK, so its prices carry none. Check your own product's rate on GOV.UK.

What the number leaves out. Contribution per unit shows what a sale earns once it happens. It doesn't include the marketing needed to win the sale. Read the contribution beside the reach: how many sales each route can realistically deliver.

Retrieval: choose the measure. In Module 6 you set baselines before judging change. A route decision needs the same discipline. Before you grow a route, note the baseline you'll measure it against: orders, contribution per order, and orders from new customers.

AI activity. Ask an AI assistant to summarise a real marketplace's seller fees for your product category. Then check every figure against the marketplace's own official fee page, with the date. Record what the AI got wrong or out of date, and whether it presented any figure with more confidence than the page justifies. Then decide which fee figure you'd use and label it an assumption, or confirm it from the page.

Having worked out what each route earns, you now decide which to grow, which to test and which to decline. The numbers inform the decision. They don't make it.

Grow, test or decline. For each route, ask three things.

• Does it earn enough per sale to be worth the effort, once every cost is counted?
• Does it reach people you can't reach another way, or only people you already sell to?
• Can you serve it well, with stock, delivery and customer service, without harming the routes that already work?

A route with strong contribution and proven demand is one to grow. A route that looks promising but has gaps in the evidence is one to test, small, with a baseline and an end date. A route that earns too little or puts the brand at risk is one to decline, for now. Saying no to a route is a decision, and worth writing down with the reason.

Channel conflict. Channel conflict is tension between your own routes. It shows up in three ways.

• Price: one route sells at a lower price than another and pulls customers across.
• Availability: a product is on every route, so no route feels special.
• Brand control: a partner presents the product in ways you wouldn't, from the photo to the description.

The fix isn't always a policy of identical prices. It might be a product only one route carries, such as a gift set, or a clear difference in what each route offers.

Leadership: making a call others may not like. A route decision usually has someone who wants a yes. A sales colleague likes the big listing; a customer service lead worries about new complaints. Lead the decision by showing the numbers, saying what you assumed, giving the reason for the choice and saying what would change your mind. People can disagree with a call and still trust how it was made.