The 10 Questions Every Manufacturer Should Answer Before Building an Ecommerce Store

Most manufacturers do not fail at ecommerce because they picked the wrong platform. They fail because nobody asked the hard questions first. The site gets built, the launch date lands, and then reality arrives. Nobody owns the inbox. The stock number is wrong. The pallet economics do not survive contact with a single-unit parcel. Six months later the project is quietly described as “a bit of a learning experience”.

The businesses that get this right start somewhere less exciting. They start with an honest look at where they are today.

When we run a readiness audit with a manufacturer, we score the business across the areas that make or break a direct-to-consumer move. Below are 10 questions drawn from that audit. They are deliberately uncomfortable. If you can answer all 10 with confidence, you are ready to build. If you cannot answer three or four of them, you have just found your project plan.

Question one: Who owns this on Monday morning?

Not who signed off the budget. Who actually owns it.

Ecommerce needs someone accountable for the website, the content, the ads, the analytics and the customer emails. In most manufacturers that work quietly lands on whoever seems most comfortable with a computer, on top of their existing job. That is not ownership. That is a hobby with a deadline.

No named owner is the single most common reason D2C projects stall. Not budget. Not technology. Ownership.

What good looks like: one named lead with the authority to make decisions, protected time in the diary, and a clear route to whoever holds the budget.

Question two: Why would anyone buy direct from you?

Your retail partners have shelves, footfall and existing customer relationships. Amazon has next-day delivery and a billion reviews. You need a reason for someone to bypass all of that and buy from your site instead.

“Because we make it” is not a reason. It is a fact. Better value, exclusive lines, configuration and personalisation, richer information, faster access to new products, direct support from the people who built the thing: those are reasons.

This is also where you decide your pricing guardrails. Undercut your stockists and you win a few orders and lose a channel. Get the rules clear before the first product goes live, not after the first angry phone call.

What good looks like: a proposition you can say in one sentence, a price position that does not sabotage your trade accounts, and products that are genuinely ready for individual parcel delivery.

Question three: Can a stranger find, understand and buy in four minutes?

Four minutes. That is the test.

Take someone who has never heard of you. Put them on your site on a phone. Can they work out what the product does, whether it fits their situation, what it costs delivered, when it arrives, and how to complete the purchase, in four minutes without asking anyone?

Manufacturers usually struggle here for a specific reason. Your product content was written for buyers who already know the category. Spec sheets, part numbers and trade language. A consumer needs to know whether it will work for them, and they need to know it fast.

What good looks like: mobile-first, fast, honest about delivery and cost, with trust signals in place and a checkout that does not ask for anything you do not need.

Question four: Do you make money on one unit, or only on a pallet?

This is the question that changes projects.

Your unit economics almost certainly work at trade volume. Now strip it back to one item, in one box, shipped to one house, with card fees, packaging, pick and pack labour, and a real probability that it comes back. Add the cost of acquiring that customer through paid media or search. Is the order still profitable?

If you do not know your landed cost per SKU, your fulfilment cost per order and your target acquisition cost, you cannot answer this. And if you cannot answer this, you are about to scale something that loses money on every sale.

What good looks like: a per-order P&L you trust, a known break-even acquisition cost, and a view of what a customer is worth over time rather than just on the first order.

Question five: Where is the first order actually coming from?

Every business plan assumes traffic. Very few explain where it comes from.

Be specific. Which search terms do you want to rank for, and who currently owns them? What will you spend on paid media, and what return do you need for it to work? Do you have an email list, or a database of trade contacts you cannot legally market to consumers with? Is there an audience on social who already care about what you make?

“We will do some SEO and a bit of Facebook” is not a demand plan. It is a wish.

What good looks like: named traffic sources, a realistic view of how defensible each one is, and a measurable return you can actually track back to orders.

Question six: Can you ship one?

Your warehouse is built to move volume to trade addresses on scheduled runs. Consumer ecommerce asks for something different: single items, picked accurately, packed to survive a courier network, dispatched against a cut-off time, tracked, and occasionally sent straight back.

Ask the practical questions. What time is your next-day cut-off? Who packs it? Does the outer carton survive without a pallet around it? Is your trade packaging covered in barcodes and case quantities that look wrong on a doorstep? How does a customer start a return, who authorises it, and how quickly is the refund processed?

Returns are the part manufacturers most often forget. Consumers return things. It is normal, it is legally protected in most cases, and if your process for it is “email Sharon”, you have a problem waiting.

What good looks like: a documented pick, pack and dispatch process, packaging designed for a parcel network, a published cut-off, and a returns route that works without anyone improvising.

Question seven: What happens the second time someone buys?

Acquisition is expensive. The profit is usually in the second, third and fourth order.

So what actually happens after order one? Is there a welcome sequence? Does anyone ask for a review? If a customer emails on a Friday afternoon, how long do they wait? Do you have any mechanism for bringing people back, or does every sale start again from zero?

Most manufacturers launch with a checkout and no plan for what follows. That is an expensive way to rent customers rather than own them.

What good looks like: captured first-party data with proper consent, automated flows that do the obvious jobs, a response time you would be happy to publish, and reviews you collect on purpose.

Question eight: Does your stock number tell the truth?

Your website says 14 in stock. Your ERP says 6. Your warehouse says the last two are damaged. Somebody is about to have a bad day.

Data flow is where ecommerce projects break silently. Ask how information moves between the website, the ERP, the warehouse system, the CRM and finance. Ask who owns each connection. Then ask the question everyone avoids: what happens when one of them breaks at 2am on a Saturday?

This is also where your product data lives. Structured, complete, consistent product information is what feeds your site, your marketplaces, your ads and increasingly the AI tools your customers use to find products at all. Messy data limits every channel you might want to add later.

What good looks like: stock and order data that moves both ways without a human copying a spreadsheet, structured product information, analytics configured properly, and named ownership of every integration.

Question nine: What happens when the person who does it is on holiday?

Ask how a new product gets set up on the site. Ask how a promotion gets built. Ask how a complaint gets escalated. If the answer to all three is a person’s name rather than a process, you do not have a system. You have a dependency.

Undocumented processes are survivable at low volume. They become the bottleneck the moment the channel starts working, and they guarantee an inconsistent customer experience in the meantime.

What good looks like: the handful of processes that repeat every week written down clearly enough for someone else to run them.

Question ten: Who is accountable if it goes wrong?

Selling direct to consumers means taking on obligations you never had when you sold to trade. Consumer rights and returns law. Your lawful basis for capturing and using customer data. Cookie consent that is actually respected rather than just displayed. Safety marks and labelling for a consumer audience. Warranty terms written for a person, not a purchase order. Accessibility standards on your site.

None of this is difficult. All of it is expensive to retrofit, and some of it carries real risk if you get it wrong. Deal with it before launch, when it costs an afternoon.

What good looks like: policies that reflect what you actually do, consent handled correctly, and one person who knows they own this.