How do I install CustomerGenius in my Shopify store?

Click Install app in the Shopify App Store and approve the requested permissions. Our onboarding wizard opens automatically inside Shopify Admin, guiding you through setting your fraud detection thresholds and running a one-time data sync across your last 60 days of orders. Most merchants are fully configured in under 10 minutes.

Does CustomerGenius start refunding orders the moment I install it?

No — auto-refund is disabled by default, so you're always in control. During onboarding you set the fraud score threshold at which orders are flagged. You can choose to manually review flagged orders for a period to build confidence in the scoring, or enable automatic refunds right away. Either way, you'll receive an email notification each time a suspicious order is detected.

What signals feed the fraud score?

CustomerGenius checks every new order against all previous orders that used the same discount code. It compares email addresses, phone numbers, billing addresses, shipping addresses (both exact and fuzzy matching), and customer names. Each field that matches a prior order adds points to the fraud score. Fuzzy matching catches small variations like transposed letters or different address formats that a simple exact match would miss.

Can CustomerGenius detect abuse across a family of similar codes (e.g. SAVE10, SAVE20, SAVE30)?

Yes — this is what cross-evaluation discount groups solve. Add codes to the monitored list individually, or use a wildcard like SAVE* to group an entire family of codes into a single eligibility pool. CustomerGenius then scores incoming orders using any code in the group against prior orders using any other code in the group. It's the right setting when customers rotate through different codes from the same campaign to dodge per-code limits.

Does CustomerGenius work with order tags as well as discount codes?

Yes. Some promotions don't flow through a discount code — gift-with-purchase, BFCM offers tagged at the order level, influencer flags, and so on. Add the tag to the monitored list in settings and CustomerGenius applies the same identity-based abuse detection it uses for discount codes. Wildcard prefixes work for tags too, so you can group whole families of campaign tags together.

Can different discount codes have different fraud thresholds or actions?

Yes — custom automation rules let you set per-code (or per-tag) fraud-score thresholds and choose whether to refund, place the order on fulfillment hold, or take no automatic action. Rules are matched in priority order, so you can override the default behavior for high-value or sensitive codes without changing how everything else is treated. If no custom rule matches, the default threshold from your settings applies.

Can the app observe abuse for a while before refunding automatically?

Yes. Every automation rule has an occurrence setting from 2 to 10. Occurrence 2 means act the second time an identity hits the rule; occurrence 3 means wait until the third; and so on, up to 10. Many merchants start at 3 or 4 to confirm patterns are real before going aggressive, then tighten the threshold once they trust the matches. It's the safest way to roll out auto-refunds on a new code.

Can I adjust how much weight each signal carries in the fraud score?

Yes. In the Fraud Score Rules tab you can tune the points awarded for each signal — exact shipping address, billing address, fuzzy address, phone, email, full name, first name, last name — and the threshold at which a customer is considered suspicious. The defaults work well for most stores; adjust only if you're seeing too many false positives or want stricter scoring on a specific signal.

What exactly happens when an order is refunded?

CustomerGenius issues a full refund through the Shopify API and optionally sends an automated email to you and the customer explaining the cancellation. If you have Recharge connected, any associated subscription is cancelled there as well. If you use ShipStation, the order is cancelled before it ships. All refund activity is logged in your CustomerGenius dashboard for monthly reporting.

Can I run a discount that's used once on my online store but applies on every subscription renewal?

Yes. Enable the Skip recurring subscription orders toggle in settings and renewal orders from Recharge and other supported subscription apps are excluded from fraud evaluation entirely. That means a discount code intended to be used only once on online store orders can keep auto-applying on every recurring renewal without each renewal triggering a duplicate-use refund — fraud checks run on the original online order, not on the renewals. As a side benefit, your existing subscribers also never get flagged for placing legitimate repeat orders that share the same address, email, or phone.

Can I block specific customers from ordering at all, regardless of discount use?

Yes — this is a separate feature from discount monitoring. The block list is for customers you've decided not to serve, like repeat chargeback filers or people who've abused your support team. Add a tag to a customer in Shopify and CustomerGenius will refund any future order from anyone matching their identity — even if they sign up with a new email, phone, or address. The block follows the person, not just the email.

Will CustomerGenius slow down my checkout?

No — CustomerGenius has zero impact on checkout speed. All fraud detection happens post-checkout via Shopify's order webhooks. Your storefront, cart, and payment flow are completely unaffected. Customers experience no difference at checkout regardless of whether their order is later flagged.

Is there a free trial for CustomerGenius?

Yes. All three pricing tiers include a 14-day free trial with full access to every feature — no credit card required to start. There is no permanent free tier, but the trial period is designed to give you enough data to see exactly how much discount fraud exists in your store and whether the subscription pays for itself (on average, merchants save 11x their monthly cost in recovered discounts).

What happens to customers who are caught abusing discounts?

Their order is refunded and they receive an optional notification email. Interestingly, based on aggregate data from CustomerGenius merchants, 38% of customers who are caught repurchase at full price — meaning catching fraud often converts an abusive customer into a profitable one.

How do I get help if something looks wrong?

Email our support team directly at support@customergenius.io. We're run by Shopify merchants ourselves and respond quickly to questions about fraud scoring, threshold tuning, or integration issues.

Comparison

CustomerGenius vs. Shopify Flow

Shopify Flow is a free, powerful workflow automation tool that lets you trigger actions on store events. When Shopify retired the Fraud Filter app in January 2025, Flow was the official recommended migration path — which has left many merchants asking whether Flow alone can handle discount abuse detection. The short answer: Flow is great at what it does, but it can't do cross-order identity matching, which is the entire problem.

TL;DR

Shopify Flow is a workflow automation tool — it lets you build if-then rules that fire on store events like 'order created' or 'customer tagged'. It's free, well-documented, and excellent for use cases like 'tag customers who spend over $500' or 'send a Slack notification when a high-AOV order comes in.' Shopify recommended Flow as the migration path when Fraud Filter was sunset on January 31, 2025, but Flow cannot do identity matching across orders: comparing every new order against every prior order on five signals (email, phone, name, shipping address with fuzzy matching, billing address) and scoring them for the likelihood of being the same person under a different identity. That's a different category of tool, and it's what CustomerGenius is built for.

Feature-by-feature comparison

Feature	CustomerGenius	Shopify Flow	Notes
Identity matching across orders	✓	—	Flow can't compare orders to each other
Fuzzy address matching	✓	—	Flow has no address normalization
Detects same person across multiple emails	✓	—
Customizable fraud scoring across 5 signals	✓	—
Cross-evaluation discount groups	✓	—
Per-code custom automation rules	✓	Partial	Flow can branch on code but not score identity
Occurrence-based thresholds (act on 3rd attempt etc.)	✓	—
Automatic refund/cancel via Shopify API	✓	✓	Flow can call admin APIs
Identity-based block list	✓	Partial	Flow can match a customer tag; CG matches identity across new emails
General-purpose workflow automation	—	✓	Flow's strength
Pricing	$9.99–$29.99/mo	Free	Flow is included with Shopify

What Shopify Flow does well

Shopify Flow is a no-code workflow builder. You define a trigger (an event like 'order placed', 'customer created', 'fulfillment shipped'), optionally add conditions ('if order total greater than $500'), and connect to actions ('add tag', 'send email', 'call Admin API'). It's powerful for the use cases it's designed for: customer segmentation, automated tagging, integrations between Shopify and other systems, and simple conditional logic at order time. Every Shopify store gets it for free. For straightforward if-then operations, it's hard to beat.

Why Flow can't detect discount abuse

Discount abuse detection isn't an if-then problem — it's a similarity problem. To know whether the new order under email NEWUSER123@gmail.com is actually the same person who previously ordered under email JOHNSMITH@gmail.com, you have to: pull every prior order in your store, match on email (exact), phone (normalized via libphonenumber), name (with first/last swap tolerance), shipping address (with fuzzy matching for typos and format variations), and billing address — then score the matches across all five signals and decide whether the score exceeds your suspicion threshold. Shopify Flow has no primitives for any of this. It can branch on the current order's properties, but it can't compare the current order to every other order, and it has no fuzzy-matching or scoring logic.

The 'I'll just check the email' trap

Some merchants try to approximate abuse detection in Flow with a rule like 'if the email has been seen before, flag the order'. That doesn't catch any real abuse, because abusers create new emails for each order — that's the entire point of the abuse pattern. A rule that only fires on duplicate emails will never trigger for a determined abuser. To catch the actual pattern, you need to match on address and phone (which the abuser typically can't change as easily), which Flow has no way to do because it doesn't compare orders to each other.

Where Flow and CustomerGenius complement each other

Flow is a great companion to CustomerGenius — they're not competitive. Flow handles general workflow automation. CustomerGenius handles cross-order identity scoring. A common setup: CustomerGenius detects discount abuse and tags the customer; Flow takes over from there to add the customer to a segment, kick off a Klaviyo flow, send a Slack alert, or update a Google Sheet. CustomerGenius does the hard part (identity matching); Flow does the easy part (downstream automation).

Cost vs. capability

Flow is free; CustomerGenius starts at $9.99/month. If your fraud detection needs are simple enough that Flow can handle them — for example, 'tag orders above $1000 as VIP and notify me' — there's no reason to buy a paid tool. If the question is 'is this customer using my first-time discount under a fresh email even though they're at the same address as a prior order?', Flow can't answer that, and CustomerGenius can. The pricing difference is the cost of having a tool that's purpose-built for that specific question.

Which one should I use?

Use Flow for general workflow automation. Use CustomerGenius for discount abuse detection. They're not alternatives — they're complementary tools, and most stores using CustomerGenius are also using Flow for other things. If you're trying to detect discount stacking with Flow alone, you'll catch the obvious (same email twice) and miss everything that actually matters. The free trial of CustomerGenius scans your last 60 days of orders during onboarding so you can see whether any abuse exists before paying anything.

Start the 14-day free trial See setup recipes

Other comparisons