I Tried AI Dropshipping For 7 Days, Copy Every Step

Executive Summary

Dropshipping expert Lauran attempted to build and run an office-products store in seven days, with a $300 budget and a goal of $1,000 in sales. Mark Tilbury selected the niche at random and reviewed the products, store, ads, and final results. The challenge generated sales and a reported profit, but did not reach the $1,000 target.

Frameworks and Playbooks

Product Research and Validation

• Look for products that solve a clear problem or have a strong “wow” factor and can be understood in about three seconds.
• Favor products that are lightweight, safe to ship, have room for a 2–3× markup, and are priced below roughly $80 for a new, untrusted store.
• Cross-check marketplace and AI-generated product data with social-media hashtag searches. Assess engagement, sales, reviews, audience, seasonality, competitor videos, and selling prices.
• Use a yes/no product checklist to score candidates before committing. A shortcut desk mat scored 8/8; an ordinary wired mouse scored 2/8 and was rejected.
• Reverse-research viral videos: start with content that attracts attention, then identify and validate the product.

Competitor Research

• Search TikTok for videos featuring shortlisted products, then read comments for customer questions and unmet needs, such as availability, price, purchase links, product dimensions, print durability, and software compatibility.
• Review competitor product pages for naming, images, copy, bundles, reviews, and purchase flow. Record weaknesses to address in your own store.
• Use Google Lens on product images to find independent sellers and compare their positioning and presentation.

Store Setup and Conversion Basics

• Use an AI store builder to get a Shopify store live quickly, then replace irrelevant default products and tailor the design and copy.
• Build clear navigation, product categories, an “About Us” page, FAQs, shipping and returns policies, contact details, and an order-tracking page.
• Use concise, benefit-led product copy, clear product variants, a visible add-to-cart button, and a sticky add-to-cart feature.
• Offer free shipping and use bundles or quantity discounts to increase average order value. Lauran moved bundle offers above the product description to make them easier to notice.
• Set up payment processing, a custom domain, and an advertising pixel. Test that the pixel records site activity before launching ads.

Creative Production and Testing

• Study successful ads for their opening hook, pacing, product demonstration, and call to action. Then develop distinct creative angles rather than simply copying a competitor.
• Make the first 2–3 seconds communicate a relatable problem or visual surprise. Keep videos concise—Lauran aimed for under 30 seconds—and use clear on-screen text, clean backgrounds, good lighting, and royalty-free music.
• Produce multiple creatives per product and A/B test them. Use results to guide the next round of creative rather than assuming the most polished-looking ad will win.
• Mark’s advice was to run the available videos and let performance data determine which to develop further.

Paid-Ad Iteration and Scaling

• Lauran used Meta/Facebook sales campaigns, directing visitors to the relevant product page rather than the store homepage.
• He tracked views, clicks, click-through rate, cost per click, spend, purchases, and frequency. He considered a frequency of about 1–1.5 a useful range to monitor; a frequency around 2+ could indicate ad fatigue.
• After the first day, he shifted budget toward the product with purchases while retaining smaller tests for the other products. He emphasized that one day of data is limited and should not be treated as conclusive.
• His general recommendation was to test organic content first when time allows, then put paid spend behind stronger creatives. In this challenge, the deadline led him to go straight to paid ads.

Products and Execution

Mark selected three products for video marketing:

• An oversized shortcut desk mat, branded Keyflow Deskmat
• A foldable desk fan, branded Frost Flow Pro
• A glittery fan, branded Diamond Breeze Pro

A power bank and label printer were judged less differentiated or harder to compete with. Lauran had also ordered extra products as backups; Mark refunded $70 for items that were not used in the challenge’s ads.

The Diamond Breeze Pro was the early ad winner, despite being the product Lauran expected to perform least well. Both video variations generated a purchase on day one. The desk mat received clicks but no early purchases, and the other fan had no early sales.

For the desk mat, comment research suggested opportunities to clarify size, compatibility, availability, and where to buy. For the fan, competitor-page research highlighted the value of clean design, concise copy, clear discounts, and a persistent purchase button.

Lauran used AI to help draft SEO text and product descriptions, Canva for branding, CapCut for video editing, and Shopify/AutoDS to import products and automate order handling.

Product pricing in AutoDS was configured around a 70% additional profit target above fees, with prices ending in .99 and a displayed comparison price set about 25% higher than the actual selling price. These were setup choices, not demonstrated realized margins.

Automatic order fulfillment and tracking were enabled. Lauran also tried AutoDS supplier sourcing after orders came in, but did not find a better supplier in that check.

Metrics and Results

• Challenge constraints: Seven days, a $300 starting budget, and a target of $1,000 in sales.
• Ad budget: Lauran said he had about $244 available for ads after setup costs. He initially set campaigns at $40 per day, then reduced two campaigns to $7.30 per day after the first day and kept the apparent winner at $40 per day.
• First-day ad results: Two purchases for the Diamond Breeze Pro, with reported ad spend of $85.77. The desk mat and Frost Flow Pro had no purchases that day.
• Final figures stated in the video: $855.75 revenue, 25 sales/items, $249.66 ad spend, $472.84 product and shipping costs, $28.90 transaction fees, $48.35 sample-order costs, and $1.99 software costs.
• Mark and Lauran described the final profit as “$541” in the subtitles, but the decimal and cost figures are unclear. The listed revenue and cost line items imply approximately $54.01 profit, so the transcript’s final profit figure should be treated cautiously.
• The daily revenue figures shown for days one through four also do not add up to the stated final revenue. The video’s totals therefore contain apparent subtitle or arithmetic inconsistencies.
• The challenge did not reach the $1,000 sales goal. Mark noted that sales improved over the reported days and saw the result as promising, while stressing that a longer testing period could provide more evidence.

Actionable Recommendations and Cautions

• Validate products before building a one-product store; keep a broader catalog or backup options while testing.
• Turn customer comments into specific product-page improvements, especially around sizing, compatibility, availability, shipping, and checkout.
• Optimize the destination page and purchase flow, not just the ad creative. Make pricing, variants, shipping terms, and the add-to-cart action easy to understand.
• Automate fulfillment and tracking early to reduce manual work and mistakes as order volume grows.
• Treat early ad results as directional. Reallocate budget carefully, keep testing alternatives, and avoid scaling based on a single day of data.
• The store displayed testimonials before it had its own customer reviews, which Mark questioned. Lauran also imported reviews from AliExpress listings for the same products. Such reviews should be clearly represented and handled transparently rather than presented as verified reviews from the store’s own customers.

Presenters and Sources

Presenters: Mark Tilbury (host and reviewer) and Lauran (dropshipping expert and challenge participant).

Tools and research sources referenced: Shopify, AutoDS, Meta/Facebook Ads, TikTok, AliExpress, Amazon, Google Lens, ChatGPT, Canva, CapCut, Artlist, Shopify Inbox, and Judge.me.