Shopify Just Released 2,000 AI Shoppers to Test Your Store — Here's How to Use SimGym Before Your Competitors Figure It Out

Shopify SimGym AI store simulation dashboard showing AI shopper navigation patterns and add-to-cart analysis

You changed your homepage hero image last month. Sales dipped 8% the next week. Was it the image? A seasonal trend? Bad luck? You have no idea — because you didn't test it first. Shopify SimGym exists to kill that guessing game.

That guessing game is what kills most Shopify stores' conversion rates. You make changes based on gut feeling, wait weeks for enough traffic to notice a pattern, and by then you've already lost hundreds of sales you'll never get back. Shopify's new SimGym tool changes this entirely — it sends hundreds of AI-powered shoppers through your store and tells you what's working and what's broken in minutes, not weeks. It launched in the Winter 2026 Edition, it's in Research Preview right now with free credits, and most merchants haven't heard of it yet.

What SimGym Actually Does (And What It Can't)

SimGym is a collaboration between Shopify and NVIDIA. It creates AI shoppers — each with a unique persona, budget, and shopping intent — and sends them through your store in a contained environment. These aren't random bots clicking links. They're trained on real commerce sessions across Shopify's platform, meaning they browse, evaluate, and make purchase decisions the way actual humans do.

You get two types of simulations:

  • Theme comparison — pits your live theme against another installed theme. The AI shoppers split between both and the winner is determined by add-to-cart rate.
  • Single theme analysis — examines one theme and delivers a breakdown of strengths, weaknesses, and prioritized recommendations ranked by conversion impact.

The results include add-to-cart rate, cart value, navigation patterns, and qualitative feedback about the shopping experience. That last part matters — SimGym doesn't just tell you what happened, it tells you why. "Navigation was confusing" or "product descriptions didn't answer the shopper's question" is the kind of feedback you'd normally need user testing interviews to get.

The limitation worth knowing: SimGym shoppers mimic real behavior, but they aren't real customers spending real money. Shopify says it directly — if you need to know whether changes actually increase revenue, you still need real A/B testing. SimGym is for catching problems and comparing options before you put changes in front of real customers. Think of it as a dress rehearsal, not opening night.

How to Run Your First Simulation in 10 Minutes

SimGym is available as a free app in the Shopify App Store. Install it, and you'll find it under Apps in your admin. During Research Preview, you get free simulation credits — each simulation costs one credit.

Before you start, make sure your store has at least one product listed and uses a Liquid storefront. For theme comparisons, you'll need two themes installed.

  1. Open the SimGym app and click Create simulation.
  2. Name your simulation something descriptive — "Homepage redesign test March" beats "Test 1."
  3. Choose your simulation type. For your first run, single theme analysis on your live theme gives you the most actionable feedback with the least setup.
  4. Optionally set a Focus area — Home page, Products, Collections, Cart, or Search. If your bounce rate is high, focus on the homepage. If add-to-cart is low, focus on Products.
  5. Click Start simulation. You can watch AI shoppers interact with your store in real time as it runs.

The simulation queue typically completes within minutes. When it's done, you get a results dashboard with metrics and written recommendations sorted by impact.

The 4 Store Changes Worth Testing First

Don't waste your free credits on trivial experiments. These four areas have the highest conversion impact based on what SimGym's AI shoppers are trained to evaluate:

Product page layout. This is where most conversions die. SimGym's shoppers evaluate whether they can quickly find pricing, understand what the product does, and locate the add-to-cart button. If your product page buries the CTA below three paragraphs of description, the simulation will catch it. Run a single theme analysis with Focus area set to Products.

Homepage hero section. Your hero has a few seconds to communicate what you sell and why someone should care. SimGym measures how AI shoppers navigate after landing — if they bounce to collections immediately or scroll past the hero without engaging, your messaging isn't landing. Compare two hero variants by duplicating your theme, changing the hero on the copy, and running a theme comparison.

Navigation and collection structure. "I couldn't find what I was looking for" is the silent conversion killer you never hear about because those customers just leave. SimGym tracks navigation paths and flags when shoppers loop through pages without finding products. If your categories are organized by your internal product taxonomy instead of how customers actually shop, the simulation exposes it.

Cart and checkout flow. Set Focus area to Cart and see where friction hits. SimGym evaluates checkout experience and flags confusion points. If shoppers are adding to cart but not completing purchases, the qualitative feedback tells you why — surprise shipping costs, unclear delivery timelines, or too many form fields. (If checkout friction is your main issue, our guide to Shopify checkout customization covers what the new customization tools actually let you fix.)

How to Read SimGym Results Without a Data Degree

SimGym delivers two types of output. For theme comparisons, the dashboard shows a winner based on the higher add-to-cart rate. Simple. If scores tie at zero, the result is inconclusive — your two themes perform identically for the metric that matters, so the difference isn't worth worrying about.

For single theme analysis, you get something more useful: a prioritized recommendation list. Each item includes what's working, what's not, and specific changes ranked by expected conversion impact. Start at the top of the list. Ignore the bottom. Shopify already did the prioritization work for you.

The qualitative feedback is where the real value hides. When an AI shopper says "I couldn't tell if shipping was free" or "the product images didn't show the size," that's the same feedback a real customer would give — except real customers don't give feedback. They just leave.

One pattern to watch for: if add-to-cart rate is reasonable but cart value is low, your product pages are doing their job but your cross-sell and upsell strategy isn't. That's a separate problem from layout or navigation — it means customers find what they want but never discover complementary products. Tools like EasySell can add quantity offers and upsells directly on the product page to close that gap. For more on this, see our post-purchase upsell guide.

What Should You Use When SimGym Isn't Enough?

SimGym handles the "will this break something?" and "which of these two options is better?" questions. It doesn't handle the "did this actually increase my revenue?" question. For that, you need real traffic A/B testing.

Shopify's new Rollouts feature — also launched in Winter 2026 — lets you control what percentage of real visitors see a theme change. It's designed primarily for safe deployments, but it gives you basic split-testing capability at no cost. Push a new theme to 20% of traffic, watch your conversion metrics for a full business cycle (at least one week, ideally two), and compare.

For stores doing under 1,000 sessions/week, real A/B testing takes too long to reach statistical significance. That's exactly where SimGym fills the gap — you get directional feedback in minutes instead of waiting 6 weeks for inconclusive data. Use SimGym to narrow your options, then validate the winner with real traffic through Rollouts.

For more advanced testing — price testing, content variations within the same theme, product page element experiments — third-party tools like ShopLift or Intelligems offer deeper CRO capabilities. But don't start there. Most merchants haven't even run their first simulation yet.

The Playbook: SimGym + Rollouts in Practice

Here's the workflow that gets the most out of both free tools:

  1. Audit first. Run a single theme analysis on your live store with no focus area. Read every recommendation.
  2. Fix the obvious stuff. SimGym's top recommendations are usually low-effort fixes — button placement, missing product info, unclear navigation labels. Make those changes on a duplicate theme.
  3. Simulate again. Run a theme comparison between your original and updated theme. Confirm the changes improved the add-to-cart rate in simulation.
  4. Deploy carefully. Use Rollouts to push the updated theme to 10-20% of real traffic. Monitor for one business cycle.
  5. Scale or revert. If real metrics match the simulation prediction, roll out to 100%. If they don't, you only risked a small slice of traffic.

This cycle takes days, not months. And it costs nothing except the simulation credits you already got for free during Research Preview.

Your store is getting real traffic right now — and every unoptimized page is costing you sales you can't measure because you never tested the alternative. Install SimGym from the Shopify App Store, run your first single theme analysis today, and read the top three recommendations. You'll know more about your store's conversion problems in 10 minutes than you've learned from guessing over the last year.