The app itself installs from the Shopify App Store and configures no-code. The paid work is the strategy and the storefront integration — picking the right rewards economy for your margin and getting the widget to read as part of the brand rather than a bolt-on.
Point ratio against margin. The first strategy call is how many points a dollar of spend earns and how many points a dollar of discount costs. The usual starting point is 1 point per $1 spent, 100 points for $5 off — a 5% effective rebate before redemption rate adjusts it. Your gross margin on the typical order sets the ceiling. The expert work is the model, not the setting itself.
Earning rules beyond purchases. Alpha Loyalty gives points for sign-ups, newsletter opt-ins, birthdays, reviews and social follows. Each has to be priced sensibly: a review that earns 500 points is worth $25 off at a 1-point-per-$1 ratio, which is more than most reviews return in incremental revenue. The setup work is picking a short list of actions and point values that reward real behaviour without creating arbitrage.
Bonus points and retroactive points. Points can be boosted on key products and collections, and the app can award retroactive points for past orders. The retroactive piece matters on launch day — existing customers with historic spend should land in the program with a balance, not at zero. The expert work is deciding the retroactive window and running the import cleanly.
VIP tiers that shoppers can actually reach. VIP tiers unlock exclusive perks by points earned or money spent. The common failure mode is setting the first tier at a threshold most of a store's customer base will never hit — the tiers then look aspirational to nobody. The expert work is reading the customer file first, picking three to four thresholds that bracket real spend distributions, and defining perks that cost less than the retention they drive.
Multiple programs by customer tag. Alpha Loyalty runs several programs at once by customer tag — wholesale, retail, B2B — so a wholesale customer can earn and redeem on a different ratio from a retail customer. For stores with mixed channels, this is where the setup earns its complexity: tag hygiene has to be clean, or customers will land in the wrong program.
Theme block and loyalty page. The widget and the dedicated loyalty page carry the brand. Alpha Loyalty's editor covers colours and copy; getting typography, button styles and the tier badges aligned with the storefront theme is theme-level work. Points on product and cart pages need placement that reads as part of the page, not an overlay.
POS setup (optional). For stores with physical retail, Alpha Loyalty integrates with Shopify POS so customers earn and redeem points in-store the same way they do online. The expert work is confirming the POS extension is enabled on the right locations, training staff on the redemption flow, and testing that an online balance and an in-store balance stay synced on the same customer record.
Interaction with existing discounts. If the store already runs a welcome code, a free-shipping threshold or a standing promotion, the loyalty redemption has to coexist cleanly. The setup work is deciding stacking rules — can a VIP tier perk apply on top of a welcome discount, can a redeemed-points reward combine with a sale — and confirming at the checkout level that the stacking behaves.
Migration from another loyalty app. Alphalogic offers free migration from other loyalty apps. The expert work is pulling the balances cleanly from Smile, Rivo, Yotpo Loyalty, LoyaltyLion or wherever the merchant is leaving, mapping tiers across the two schemes if they differ, and running a communications pass so customers know their balance carried over.
Launch and ongoing analytics. A proper launch run-through: a test purchase and redemption, a tier upgrade confirmed, a POS test if that's in scope, and the first-week numbers reviewed against the model. The retainer version is reading enrolment rate, redemption rate and incremental AOV each month and tuning the ratios where real behaviour disagrees with the model.
