How to Find Products to Sell Online: A Repeatable Research Process

The best way to find products to sell online is to start with a specific group of buyers, collect ideas from places where those people already spend money, check that demand is steady, and throw out anything that can't leave you a healthy margin. That process takes a few evenings and costs almost nothing. Most beginners do it backwards: they fall for one product, then go looking for evidence that they were right.

This guide gives you a process you can repeat every time you need a new product. By the end, you should have a scored shortlist of three to five ideas, with one clear favourite to test.

What makes a good first product

Before you collect ideas, decide what you're looking for. Here is the filter I'd use for a first product, assuming you have a small budget and no audience yet:

These are my rules of thumb, not laws. A $15 consumable can work if people reorder it, and a $300 product can work if you know the niche well. But if you have no experience, breaking several of these at once is a bad idea.

I'd also avoid a few categories for a first store. Supplements, cosmetics, food, children's products and anything with a battery can bring extra safety rules, testing requirements or carrier restrictions. They aren't forbidden, but they're a poor place to learn. Clothing has its own trap, because sizes multiply your stock and raise your returns.

How to find products to sell online in six steps

Step 1: Pick the buyer before the product

"Pet products" is a category. "People with an older dog who struggles on slippery floors" is a buyer with a problem. The second one suggests specific products, specific search terms and specific places to find customers.

Write down five groups of people you understand. Use your job, your hobbies, your family situation and anything you've spent too much money on. Your own experience counts as an advantage here, because you'll spot complaints and gaps that an outsider would miss.

For each group, answer three questions:

• What annoys them regularly?
• What do they already spend money on?
• What do they buy as gifts for each other, or get bought for them?

Step 2: Collect ideas from where people already buy

Aim for 20 rough ideas. Don't judge them yet. These sources are free and show real buying behaviour:

• Amazon Best Sellers. Browse by category, then go several levels deep. The top of a broad category is dominated by big brands. Narrow subcategories are far more useful.
• Amazon and Etsy reviews. Read the three-star reviews first. They tend to be the most balanced, and "I like it, but..." is where product ideas live.
• Etsy search suggestions. Type your niche into the search bar and note what it suggests. Those suggestions show you the words shoppers use.
• Reddit and niche forums. Search for "recommend", "alternative to" and "does anyone make" within communities your buyer uses.
• Your own store receipts and browser history. Look at what you bought and were slightly disappointed by.
• Meta Ad Library. Search product terms to see what stores are currently advertising. An ad that has been running for months is a better signal than a new one, because few people keep paying for ads that lose money.

Treat social media trends with suspicion. A viral video can show that something is interesting without showing that people will pay your price for it, and by the time you've sourced stock, the moment may have passed.

Step 3: Check that demand is real and steady

Now cut your list to about ten ideas and look at search behaviour. Two free tools do most of the work.

Google Trends shows whether interest is rising, falling or seasonal. Set the range to five years and pick your target country. Keep in mind that Google normalises the data and scales it from 0 to 100, so the number tells you relative interest and not how many people searched. A term at 100 might still have few searches. Use Trends for the shape of the line:

For rough volumes, use Google Keyword Planner. It sits inside Google Ads and gives estimated monthly searches and costs. As of November 2026, Google says you need to finish account setup, including billing information, before you can get keyword ideas. You don't have to run ads to do your research, but do watch that you haven't left a campaign switched on by accident.

You're looking for search terms that show buying intent. "Non-slip dog socks" is better than "why does my dog slip on floors", although the second is useful later for SEO content.

Step 4: Look at the competition

Competition is a good sign. If nobody sells your product, the likeliest explanation is that nobody has managed to make money from it. What you want is a market with several sellers and a visible weakness.

Search your main term on Google and Amazon, then visit the top five sellers. For each, write down:

• Price, including shipping
• Number of reviews and their average rating
• Common complaints
• Photo and description quality
• Whether they appear to be one big brand or a mix of small stores

A market is attractive when small stores appear on the first page, buyers repeat the same complaint, and listings look lazy. A market is a bad fit when a few famous brands hold every top spot, or when several sellers offer the same item at near-identical low prices. That second pattern usually means you'd be competing purely on price, which is a fight a new store loses.

Write one sentence that finishes "Mine is better for this buyer because...". If you can't, drop the idea. "Cheaper" doesn't count. Good answers are things like a better size, stronger material, a useful bundle, clearer instructions or a design for one specific use.

Step 5: Run a quick margin check

Margin kills more ideas than anything else, so check it before you get attached. Find the likely selling price from your competitor research, then get a rough unit cost from a wholesale marketplace or supplier list. Add packaging and shipping.

Here's a quick example using a $30 product:

If that final number is under roughly a third of the selling price, I'd usually move on, because paid ads will struggle to work. Here it's about 46% of the price, which gives you room. Our guide on how to calculate profit margin for an online store shows how to work this out properly, including returns and ad costs.

While you're at it, decide how you'd source the product. Holding stock and dropshipping give you very different margins and risks, so read dropshipping vs holding inventory if you haven't chosen yet.

Step 6: Score your shortlist

You should now have three to five survivors. Score each from 1 to 5 on the same criteria so that your favourite doesn't win on enthusiasm. Here's what that looks like with three made-up ideas. The scores are examples, not research:

Look at what happens to the phone case. It has the strongest demand and it's easy to post, but it loses because it has no way to stand out and little room for profit. High demand alone isn't enough, and forgetting that is an easy way to waste a research session.

If two ideas tie, pick the one you'd be happier talking about every day for a year. You'll be writing about it, answering questions about it and looking at photos of it.

Where to source the product once you've chosen

You have four main options, and the right one depends more on your budget than on your product:

• Make it yourself. Best for handmade goods and small batches. Your time is the limit, so charge for it.
• Buy wholesale. Domestic wholesalers cost more per unit but offer lower minimums and faster delivery. Overseas manufacturers can be cheaper at volume, but you'll take on freight, duties and longer lead times.
• Dropship. You hold no stock, but you give up control over quality and delivery, and margins are often thin.
• Print on demand. Suits designs on existing products, with little risk and limited margins.

Whichever route you choose, order samples first, and compare at least two or three suppliers. Ask about the minimum order quantity, lead time, defect policy and whether the price includes shipping. Get those answers in writing.

Mistakes that waste the most time

• Copying a trending product list. Lists of "winning products" are read by thousands of other sellers, and the advantage disappears the moment everybody has it.
• Mistaking a big market for an easy one. Lots of searches for a broad term usually means strong competitors. Look for a narrower angle.
• Selling something only you want. Your taste can guide you to a niche, but it isn't evidence that strangers will pay.
• Researching forever. Research only reduces risk. After a week or two, you need a real test with real buyers.
• Comparing supplier prices and ignoring landed cost. A cheap unit price can become an expensive product after freight, duties and defects.

What to do next

Set a limit of one week for this process. Spend the first evening on your buyer list, two evenings collecting ideas, and the rest on demand, competitors and margins. At the end, you want one product with a clear buyer, a reason to choose you and numbers that work.

Then stop researching and find out whether real people will buy it. Our guide on how to validate a product idea walks through cheap tests you can run before you spend money on stock. If you're still at the planning stage, how to start an online store shows where this step fits into the bigger picture, and how much it costs to start an online store will help you set a budget for your first order.