Shopify Sidekick and Shopify Magic are the AI features built into the Shopify admin. Shopify Magic is the set of generative tools that write product descriptions, email copy, FAQ answers and image edits inside the screens merchants already use. Sidekick is the conversational assistant that answers questions about the store, pulls reports and guides or performs admin tasks. Both are useful for small teams, as long as a person still checks facts, tone and settings.
What do Shopify Magic and Sidekick each do?
Shopify Magic shows up as small buttons across the admin. In a product page, it drafts a description from a few keywords. In Shopify Email it suggests subject lines and body text. In the Shopify Inbox it suggests replies to customer chats. In the media editor it can remove or replace image backgrounds. It is included in Shopify plans at no extra cost at the time of writing, though availability can vary by language and region.
Sidekick is a chat panel in the admin. Merchants ask it questions in plain language, such as “which products sold most last month” or “how do I set up a discount for returning customers”. It can read store data, explain settings and in many cases prepare the change for approval. Shopify has been widening what Sidekick can do, so the exact list of actions is worth checking in the current help centre.
Which store tasks benefit most?
The gain is largest for tasks that are repetitive and low risk. Good examples:
- First drafts of product descriptions for large catalogs, edited for brand voice afterwards.
- Background cleanup for product photos so the collection page looks consistent.
- Quick sales questions answered without building a report from scratch.
- Step-by-step help with settings the team rarely touches, such as shipping zones or tax setup.
- Draft replies to common chat questions in Shopify Inbox.
A homeware store adding fifty new SKUs before a season, for example, can get serviceable drafts for each product in an afternoon, then spend its real effort on the ten hero products that drive most revenue.
Product descriptions that do not sound generic
Magic writes better when it gets better input. Give it the material, dimensions, care details and one honest selling point, plus the tone you want. Then edit: cut adjectives, add a concrete detail a customer would ask about, and check measurements against the supplier sheet. Duplicate-sounding descriptions across a catalog are a common SEO weakness, so vary the structure by category.
Where should merchants be careful?
AI drafts can state things that are not true. A description might claim a fabric is waterproof when it is only water-resistant, or invent a dimension. That creates returns and, in some markets, consumer protection risk. Image edits carry a similar risk: a replaced background is fine, but altering the colour or shape of the product itself misleads buyers. Every factual claim should be checked against the product data.
Settings changes deserve the same care. If Sidekick prepares a discount or a shipping rule, read the summary before saving. A wrong discount combination can stack with other offers and erode margin over a weekend before anyone notices.
Which settings and habits keep output on-brand?
A few habits make Magic and Sidekick output more reliable over time:
- Keep a one-page tone guide and paste its key lines into the input each time.
- Write a list of banned claims, such as “best in the world” or medical promises, and check drafts against it.
- Use the same structure per category, for example a short intro, a features list and care notes.
- Ask Sidekick to explain how it got a number before using it in a report.
- Review generated chat replies weekly and correct the source FAQ when answers drift.
These steps take little time and stop small errors from repeating across hundreds of products.
How do these tools compare with third-party apps?
Shopify’s App Store has many AI copy and image apps. Built-in tools win on convenience, since there is nothing to install and data stays in Shopify. Third-party apps can win on features such as bulk generation with templates, multilingual output or brand voice training. Some merchants use Magic for one-off edits and a dedicated app for bulk catalog work. Install apps sparingly, because each one adds scripts and permissions to the store. Extra scripts can slow product pages, and page speed affects both conversion and search visibility, so remove apps that no longer earn their place.
What should you measure after using them?
Track time saved and quality, not just output volume. Useful signals include hours spent per new product listing, organic sessions to product pages after descriptions are rewritten, conversion rate on edited pages compared with untouched ones, and return reasons that mention a mismatch with the description. If returns citing “not as described” rise after a batch of AI copy, the review step needs tightening.
How Alien Road applies Shopify Sidekick and Shopify Magic
In a typical engagement we start with a catalog review: how many products lack descriptions, which have thin or copied supplier text, and which drive the most traffic. That review sets priorities so effort goes to products that matter.
We then write a short brand voice guide and an input template for each product category, covering the facts Magic needs to draft well. The client’s team or ours generates drafts, and an editor checks claims against product data before publishing. For store admin, we train the team on Sidekick for reports and settings questions, with a rule that any change touching prices, discounts or shipping is reviewed by a second person. Reporting compares organic traffic and conversion on rewritten product pages against a control group, plus hours saved per listing.
Where to start this week
Pick ten products that get traffic but have weak descriptions. Use Shopify Magic to draft new versions from a proper input of facts and tone, then edit each one by hand. Publish them, note the date, and compare their conversion and organic sessions after four weeks with ten similar products you left alone.