How to choose what to sell online (and check there’s demand)

The best thing to sell online is a product people are already searching for, that you can sell at a healthy margin, that’s easy to post, and that you can explain better than most other sellers. Check demand with free search tools before you buy any stock, then test with a small batch.

Here’s a practical way to go from a list of ideas to one product worth trying.

Start with what you know

Your own experience is an advantage. A keen cyclist knows which accessories are fiddly and which last. A nurse knows what carers struggle to find. Knowing your customer makes everything easier: product choice, descriptions, photos and ads.

Write down 10 to 20 ideas from:

• Hobbies and interests you know well.
• Problems you or people around you have had to solve.
• Products you’ve bought online and been disappointed by.
• Your job or trade, where you understand what professionals need.
• Local makers or suppliers you could work with.

Check people are searching for it

Demand you can measure is better than demand you hope for. These free tools help:

• Google Trends shows whether interest is growing, falling or seasonal. Set the country to the UK.
• Google Ads Keyword Planner gives rough monthly search ranges for terms. You need a Google Ads account, but you don’t have to run ads.
• Google’s own search box. Type your product and look at the autocomplete suggestions and the “People also ask” questions.
• Marketplaces. Look at best-seller lists and the number of reviews on similar products. Lots of reviews usually means steady sales.
• Forums and Reddit. Real people describing real problems, in their own words.

Watch out for seasonality

Some products sell all year. Others spike for a few weeks (Christmas gifts, garden furniture, Halloween costumes) and go quiet. Seasonal products can work, but you need to time stock carefully and have a plan for the quiet months. Google Trends over five years shows the pattern clearly.

Check the competition (in a useful way)

Competition is a good sign: it proves people buy. What you want to know is whether you can stand out. Look at the top shops and listings and ask:

• Are the product photos poor or the descriptions thin?
• Do reviews keep mentioning the same complaint you could fix?
• Is delivery slow because products ship from overseas?
• Is nobody serving a specific group well (left-handed people, small spaces, older users)?

If you can’t find any way to be better, faster, clearer or more specific, think twice.

Do the margin maths

A product can be popular and still lose you money. Work out the numbers for one order:

1. 1Write down the price shoppers typically pay for similar items.
2. 2Take off VAT if you’re (or will be) VAT registered.
3. 3Take off what the product costs you, including getting it to you.
4. 4Take off packaging, postage, and payment fees.
5. 5Look at what’s left. That has to cover advertising, returns, your time and profit.

As a rough guide, products with a very low price struggle when you pay for ads, because the cost of winning a customer can swallow the profit. Higher-value products, bundles and items people reorder give you more room.

Think about the practical side

• Size and weight. Small, light items are cheaper to store and post.
• Fragility. Glass and ceramics break. Breakages cost you a replacement and a customer.
• Returns. Clothes and shoes are often returned for fit. Factor that in.
• Shelf life. Food, cosmetics and supplements have use-by dates and stricter rules.
• Sizes and variants. Ten colours in five sizes is fifty things to keep in stock.

Check the rules for your product

Some products need safety testing, labelling or marking before you can sell them in the UK. This includes toys, electrical items, cosmetics, food, children’s products and anything making medical claims. Getting this wrong can mean recalls or fines.

Test before you commit

You don’t need to guess. Try a small test:

1. 1Order samples and check the quality yourself.
2. 2Buy the smallest sensible first batch, or start with made-to-order or dropshipping to test demand.
3. 3Build a simple product page with good photos and an honest description.
4. 4Spend a small, fixed amount on targeted ads, or share it with an audience you already have.
5. 5Measure: visits, add-to-baskets and sales. Read every question customers ask.

If people visit but don’t buy, look at price, photos, delivery time and trust. If nobody visits, the problem is reaching people, not the product itself.

A hosted platform with a free trial makes this cheap to try. Merchella, for example, has a 30-day trial with no card needed, so you can build the page before you pay anything.

Signs you’ve found a good product

• Steady searches all year, or a season you’re ready for.
• A clear reason someone would buy from you rather than the big sellers.
• A margin that still works after ads and delivery.
• Easy to store, post and explain.
• Customers who might buy again, or buy something that goes with it.

In short

Start from what you know, check real search demand, do the margin maths honestly and test small. The goal isn’t a perfect product on day one. It’s a product with enough promise that you learn something from every sale, and can then improve or expand.