🗞️ I fired Wix💥🚀

AI for eCommerce Newsletter - 70

In partnership with

If you’re new here, welcome! If you’ve been reading for a while, thank you for sticking around as we navigate this wild AI shaped shift happening across eCommerce. Each week I share what I’m experimenting with, what’s actually moving the needle, and the trends that deserve your attention before they hit your competitors’ playbooks.

A quick heads up. I’ve organized all previous editions into one searchable hub. If you want the full journey, it’s all here.

I Fired Wix

For 5 years, ppcninja.com ran on Wix, a popular hosted website builder.

It did its job, and it was ok. But as the business matured, the friction became harder to ignore. The editor was clunky. Responsive behavior was unpredictable. And every small change carried overhead. Find someone on Upwork, explain what should have taken minutes, pay $100 to $200, wait, and move on.

That model works at a certain stage. It just didn’t scale with how we operate now.

What pushed the decision wasn’t dissatisfaction. It was volume. We needed a lot of changes made, quickly. We wanted samples of the creative work we were doing uploaded fast, and we wanted to update blogposts on PPC topics each week.

And it was obvious the same process would repeat itself again. More updates = More coordination = More time spent managing the tool instead of improving the site.

So what had been background friction turned into a one day experiment. By the end of it, ppcninja.com was running a gorgeous, fully responsive static HTML website, hosted on AWS-S3, fronted by CloudFront, and Wix was gone.

Annual savings: about $700.
Time spent: one focused day.
Trade offs: fewer than you think.

Here’s what changed.

Instead of logging into a visual editor and nudging boxes around, I described the changes in plain English. A short conversation with Claude. What I wanted to change. How I wanted it to look. A few sentences at a time. Claude translated that into clean static HTML that I could review, tweak, and iterate on immediately.

Once the HTML was ready, the hosting part was almost boring. Claude showed me exactly how to do it. Upload to S3. Add a CloudFront distribution. Point the domain. Done. The whole thing felt less like “web development” and more like configuring infrastructure, which is exactly the point.

Wix sells convenience. And for a long time, that trade off made sense. But the moment AI can generate usable frontend code on demand, that value equation changes. Fast.

What surprised me most wasn’t the cost savings. It was the mental shift. The website stopped feeling like a product we rented and started feeling like an asset we owned.

No monthly platform tax.
No feature gating.
No fear of breaking things by clicking the wrong setting.

This is where the superpower actually shows up.

Claude didn’t just write copy. It didn’t just “help.” It replaced an entire category of SaaS. Not because it’s better at hosting, but because it collapsed the distance between idea and implementation.

Vibe coding with Claude

That’s the pattern to pay attention to.

AI doesn’t always replace tools directly. Sometimes it removes the reason the tool existed in the first place.

I’m not suggesting everyone should rip out their CMS tomorrow. But if you’re paying hundreds every year for something that boils down to static pages, it’s worth asking a different question now.

Think hard. What else are you renting that AI can help you own?

This is what I mean when I say superpowers don’t arrive as flashy features. They show up quietly, shave real costs, and give you back control.

Weavy AI

Weavy is a visual workflow canvas for building repeatable AI powered creative processes.

At its core, Weavy is a visual workflow canvas for creative AI. You build flows by connecting blocks, run the same prompt through multiple models, branch outputs, and layer in real editing steps like cropping, masking, resizing, compositing, or pulling frames from video. It feels less like prompting and more like building a repeatable machine.

The signal that this category is real: Figma acquired Weavy. That’s not a casual pickup. It’s a strong bet that visual, node based AI workflows are becoming part of mainstream design tooling, not just a playground for power users.

What makes Weavy especially fun is the model buffet. In a single canvas, you can route prompts through multiple image models side by side. Flux variants, Imagen, GPT Image, Higgsfield, Reeve, and more. Different aesthetics, different strengths, same input. Instead of debating which model is “best,” you just let them all compete.

On top of generation, Weavy includes a surprisingly deep toolbox. Levels, painter, crop, resize, blur, channels, mask extraction, mask by text, even video frame extraction. This is the stuff that usually forces you into Photoshop or After Effects. Here, it’s just another node in the flow.

The real unlock is reuse. Once you build a workflow you like, it becomes a template. You are no longer reinventing your process every time. You are refining a system.

If you care about AI generated visuals at any kind of scale, Weavy is worth a look. Not because it’s flashy, but because it turns experimentation into infrastructure.

How much you ask? There is a free plan with 150 credits followed by $19 a month for 1500 credits. Definitely worth a test!

PPC Ninja is helping brands future proof their listings for AI, helping you build stunning images and videos with AI. Reach out to [email protected] to explore how we can scale your content production across Social media, Amazon ads, Amazon Posts efficiently and affordably.

5 ways to move faster with AI

1. Double-tap the Control key to talk to your AI
On Mac, double-pressing the Control key triggers voice input, and it’s fully configurable in system settings. If you are on Windows, use a software like Willow Voice or any AI launcher, and you can start talking to your AI instantly. No typing. No context switching.

The real trick is not trying to sound polished. Speak messy. Half thoughts. Corrections mid sentence. AI is very good at cleaning up ideas. It’s terrible at inventing the ones you never said.

2. Use special-purpose markdown files as “first read” context
Files like AGENTS.mdPROJECT.md, or README.md are increasingly treated as first-stop context by modern LLM tools, especially in coding and repo-based workflows.

Use them to define what the model should never guess:

  • Project goals

  • Hard constraints

  • Style preferences

  • Tools or libraries to favor or avoid

Keep them short and opinionated. Think onboarding note, not documentation.

3. One project, one context file
If you work across multiple initiatives, create a small markdown file per project with:

  • Objective

  • Audience

  • Non-negotiables

  • Success criteria

Load it once at the start or let your tool auto-attach it. This prevents drift without forcing long prompts every time.

4. Tell the model what not to do
Negative constraints are underrated.

Examples:

  • “Do not explain basics.”

  • “Do not add marketing fluff.”

  • “Do not introduce new tools.”

This reduces exploration cost and keeps the model inside useful boundaries.

5. Snapshot good work before switching models
Before you jump from one model to another, pause and extract a clean snapshot:

  • Final prompt

  • Key assumptions

  • Best output so far

Paste that into the next model as starting context.

This avoids both extremes: overly long conversations and cold starts. You carry forward only what earned its place.

Super excited to be speaking at the Billion Dollar Seller Summit BDSS 13, and AI-focused event hosted by Kevin King. RSVP: https://www.bdss13.com

Join me for this incredible Hack Contest on Jan 28th organized by Smart Scout and Nick Penev. RSVP here.

We hope you liked this edition of the AI for E-Commerce Newsletter! Hit reply and let us know what you think! Thank you for being a subscriber! Know anyone who might be interested to receive this newsletter? Share it with them and they will thank you for it! 😃 Ritu

The Future of Shopping? AI + Actual Humans.

AI has changed how consumers shop by speeding up research. But one thing hasn’t changed: shoppers still trust people more than AI.

Levanta’s new Affiliate 3.0 Consumer Report reveals a major shift in how shoppers blend AI tools with human influence. Consumers use AI to explore options, but when it comes time to buy, they still turn to creators, communities, and real experiences to validate their decisions.

The data shows:

  • Only 10% of shoppers buy through AI-recommended links

  • 87% discover products through creators, blogs, or communities they trust

  • Human sources like reviews and creators rank higher in trust than AI recommendations

The most effective brands are combining AI discovery with authentic human influence to drive measurable conversions.

Affiliate marketing isn’t being replaced by AI, it’s being amplified by it.

Reply

or to participate.