Can you talk about e-commerce in general regarding AI and what parts would you automate?
I'd automate anything that drives growth—for e-commerce that means getting total eyeballs on your brand. I also automate other things, like a campaign that scanned Reddit posts in 30‑40 relevant subreddits, compiled the last 24 hours of posts, and had humans post soft‑sell snippets recommending the product. Those are genuine, human‑sounding comments (not the obvious AI slop like “Everything was lost until I read XYZ book”), and they work well because they’re just more eyeballs on the brand.
Yes, I've done lots of consulting with e-commerce brands. A common AI-driven easy win is generating virtually infinite ad creative—using video models, you can feed your product and have it do anything, like a giant bottle falling from the sky crushing buildings. It's entirely up to your imagination. I value ad creative because removing that bottleneck unlocks reach, which is a core pillar of growth. While some organic-only brands are impressive, creating high-quality content, copy, and creative remains a bottleneck that limits reach.
Most low‑hanging e‑commerce fruit—like automated product descriptions, recurring emails, and checkout flows—is already handled by platforms, so you need to look beyond those basics. If the store is mostly B2C, reframe the problem as B2B: help the owner reach retailers, distributors, or influencers and set up partnership automations. In practice that often means building an email campaign and an affiliate funnel that contacts influencers, gives them free products, and tracks resulting sales. Many influencer‑marketing services are generic and lack personalization; you can improve them by scraping a creator’s recent post, paraphrasing it as an icebreaker, and offering a simple promo or swag kit. You can also layer AI into newsletters or marketing sequences—using survey data, product‑history, or other signals to personalize outreach. For coaches who feel stuck on choosing a niche or services, the real issue is a lack of confidence; pick three niches you feel about 70 % confident in and start acting (LinkedIn cold outreach, short videos, Upwork applications) rather than waiting for certainty.
The biggest lesson I learned is that the actual product or service you sell doesn’t matter much—AI automation is just a fulfillment mechanism. What really counts are the core business systems you build around it: lead handling, sales scripts, onboarding, admin, hiring, marketing, taxes, etc. Those pieces stay the same across industries; only the specific service you deliver changes. Once you have a solid business foundation, you can swap in any offering and still succeed. I’ve seen the same principles work when I sold videography, local SEO services, courses, and even events—so it’s less an ‘AI journey’ and more a general business journey.
If I were you and thinking long‑term, I’d ask where everything is heading. Everything is moving toward AI and automation over the next five years; what happens after that is uncertain, but my honest advice is to choose the AI automation business model. I’ve found the most success with it and have run it four times. The main reason is leverage: AI and automation offer far more leverage than traditional marketing agencies. Traditional marketing agencies are declining—this model peaked around 2017‑2019 and has been falling since, partly due to market saturation and the rise of SaaS tools that automate tasks like scheduled posting, brand strategy, and marketing strategy. In contrast, the AI automation model is still early; it really took off around 2021‑2022 with ChatGPT, and I predict it will continue growing until about 2027‑2028, when AI agents could double the market and demand. After that, demand may dip slightly as agents become more advanced. If you're in Saudi Arabia, you're essentially seeing a delayed version of this curve, giving you a few extra years to ride the upward trend. So, if I were in your place, I’d go all‑in on AI automation.