Start Here:
The AI Ecommerce Blueprint

Start with a buyer and a problem they already pay to solve, not a product you happen to like. A good first niche has people actively searching, spending and complaining about the current options. Aim for something you can explain in one sentence.

Act as an ecommerce product researcher. I am based in [your country] and want to sell online. Give me 10 niches where buyers have a clear, repeat problem. For each, list who the buyer is, the problem, what they currently pay, and one reason it could fail. Then rank the top 3 and explain why.

Pick your top niche and write one sentence: I help [buyer] solve [problem] with [product].

Before you build anything, check three things: is there demand, how crowded is it, and is there margin left after product cost, shipping, fees and ads. A product with a thin margin needs a much higher ad performance to survive. Work this out on paper first.

Here is a product I am considering: [product]. Selling price: [price]. Product cost: [cost]. Shipping to customer: [shipping]. Payment fees: about 3 percent. Work out my gross margin per sale and the maximum I can spend on ads per sale to break even. Then list the top 5 risks (returns, competition, shipping time, compliance, seasonality) and give a TEST or SKIP verdict with reasons.

Run your best product through that prompt. Write down the break-even ad cost per sale. That is your first key number.

Trending products on TikTok and in marketplaces can show you demand early. Check that sales are spread across many sellers, not one viral video. Then find a supplier, confirm the real landed cost and delivery time, and set up a simple one-product test store. It only needs to be clean, trustworthy and fast, not perfect.

I want to test this product: [product]. Write me a supplier checklist (questions to ask, samples to request, what delivery time is acceptable for [your country]) and a one-product test store plan: the pages I need, the trust items buyers expect (contact, shipping, returns), and a simple product page outline with headline, bullets and FAQs.

Send your supplier questions to two suppliers and note the landed cost and delivery time from each.

If tracking is wrong you cannot trust a single number. Before you spend, make sure your Meta pixel and conversions API are installed and the key events fire: view content, add to cart, initiate checkout and purchase. Test with a real test order and check the events show up in Meta.

Give me a step-by-step checklist to set up and test the Meta pixel and Conversions API on a Shopify store. Include how to confirm each event (ViewContent, AddToCart, InitiateCheckout, Purchase) fires correctly, what a healthy event match quality looks like, and the 5 most common setup mistakes.

Place a test order and confirm all four events appear in Meta Events Manager.

You only need a handful of numbers. ROAS is revenue divided by ad spend. CPC is the cost of a click. CPM is the cost of 1,000 views. CTR is the share of people who click. CPA is what one sale costs you in ads. Break-even ROAS is 1 divided by your margin, so a 40 percent margin needs a ROAS above 2.5 just to break even.

Explain ROAS, CPC, CPM, CTR, CPA and break-even ROAS in plain English with one worked example. Then calculate my break-even ROAS and target CPA using: selling price [price], product cost [cost], shipping [shipping], fees 3 percent.

Write your break-even ROAS and target CPA on a sticky note and keep it next to your screen.

A test buys data, not hope. Keep it simple: one product, a few different creatives, a small daily budget you can afford to lose, and a set number of days. Do not change things every few hours. Judge each ad against your break-even numbers from lesson 5. Results vary and nothing here is a promise of income.

Design a simple Meta ads test for this product: [product], budget [daily budget] per day for [days] days, targeting [country]. Give me the campaign structure, 4 different creative angles to test, the primary text and headline for each, and the rules for when to pause, hold or keep an ad running, based on a target CPA of [your CPA].

Write your test plan on one page: budget, days, creatives, and your pause rule. Do not launch until you have it written.

Every day, look at each ad against your target numbers and make one of four decisions: scale it, hold it, cut it or refresh it. A short daily routine beats a long weekly panic. Give each ad a clear status so nothing sits there losing money by accident.

Act as my ecommerce media buyer. Here is yesterday's data for each ad: [paste ad name, spend, purchases, ROAS, CTR]. My break-even ROAS is [number] and target CPA is [number]. For each ad, tell me SCALE, HOLD, CUT or REFRESH and give the reason in one sentence. Then list my top 3 actions for today.

Set a 10-minute daily slot to run this check, and keep a simple log of the decisions you make.

Real people and real reviews build trust faster than polished ads. Brief a few creators to show the product in use, collect genuine customer reviews, and then brand the store properly once a product proves itself. Keep all claims honest and compliant, especially for health and wellness products.

Write a short UGC creator brief for [product]: who the video is for, a 20 to 30 second structure (hook, demo, benefit, call to action), things the creator must not claim, and 5 hook ideas. Then write a polite email asking recent customers for an honest review.

Send your brief to 3 creators or friends, and request reviews from your first 5 customers.

Keep this master prompt. It gives Claude or ChatGPT the context of your business so every answer is specific to you. Fill it in once and paste it at the start of any new chat.

You are my ecommerce advisor. My business: [what you sell]. My customer: [who buys]. My country: [country]. My price and margin: [numbers]. My main goal this month: [goal]. My biggest problem right now: [problem]. Always give specific, practical next steps, flag risks, and ask me for any numbers you need before answering.

Fill in the master prompt, save it in a note, and use it for your next question about your business.