LoyaltyLion: What Actually Moves Retention, Not the Vendor

LoyaltyLion is a loyalty and rewards platform that a lot of $3M-$30M brands shortlist early, usually because a competitor uses it or because it turns up first in a comparison search. That is a reasonable way to build a shortlist and a poor way to make the decision, because the platform you pick determines less about your loyalty programme’s outcome than the design decisions you make inside whichever platform you choose. This article works through what actually moves retention, the programme design choices that matter more than the vendor name, and the real cost of moving an existing points balance if you decide to switch.

What actually determines whether a loyalty platform works?

A loyalty platform works when its earn rate, redemption threshold, tier logic and expiry policy are calibrated to your margin and your customers’ actual buying rhythm, not when it has the most features on a comparison page. Two brands running the identical platform, LoyaltyLion or any competitor, can see completely different results if one has set an earn rate that quietly erodes margin on every order and the other has priced points against a redemption rate it actually measured.

Vendor comparison pages rarely spend much time on this part, because it is not a feature the vendor controls. A platform can offer tiers, referral rewards, birthday bonuses and VIP perks, and every one of those mechanics can still fail to move a single customer’s behaviour if the thresholds are set somewhere your customers never reach, or the reward is worth less than the friction of claiming it. The mechanism that makes a loyalty programme work is simple to state and easy to get wrong in practice: a customer needs to notice the reward, believe it is reachable soon, and value it enough to come back for it. Every configuration decision either supports that chain or breaks it somewhere.

Why is comparing loyalty platforms on features the wrong first step?

Comparing loyalty platforms on features first is the wrong step because most platforms in this category, including LoyaltyLion, support the same broad set of mechanics: points for purchase, points for engagement actions, tiers, referrals and redemption at checkout. The differences that show up on a feature comparison are mostly differences in how deep the configuration goes for each mechanic, not whether the mechanic exists at all. A brand that has not yet decided what earn rate or tier threshold it needs cannot meaningfully evaluate configuration depth, because it does not know which knobs it will actually turn.

The better first step is specifying the programme on paper before touching a vendor’s settings screen: what earn rate the margin supports, how many tiers make sense given your actual repeat purchase distribution, what a redemption should feel like at checkout, and how long points should live before they expire. Once that specification exists, evaluating a specific platform, LoyaltyLion included, becomes a question of whether its configuration options can express what you have already decided, rather than a question of which vendor’s marketing page sounds more complete. Confirm LoyaltyLion’s current configuration limits and pricing directly with the vendor, since plan structure and feature gating change over time and are not something worth guessing at here.

What programme design decisions actually move retention?

Four design decisions carry most of the weight in a loyalty programme, and each one has a specific, checkable failure mode if set wrong.

Earn rate. The rate at which purchases convert to points determines your cost per dollar of loyalty-driven revenue. Set it by starting from gross margin: work out what a point costs you at the moment a customer redeems it, multiply by the share of issued points you actually expect to be redeemed rather than assuming full redemption, and compare the resulting cost against what you already pay to acquire a comparable repeat order through paid channels. An earn rate set as a round, easy-to-market number without that arithmetic tends to either erode margin on your best customers or reward so little that nobody notices it exists.

Redemption threshold and friction. A reward that requires an unreachable number of points reads as a programme built to look generous without costing anything, and customers who do the mental maths notice. A reward with almost no threshold devalues itself the same way a coupon code that appears on every visit does. The friction of redeeming matters as much as the threshold: a redemption buried three menus deep at checkout loses far more customers to abandonment than a threshold set slightly too high.

Tier thresholds. Tiers work when they are set against your own order frequency distribution, not a vendor’s suggested default or a round number that sounds aspirational. If fewer than a meaningful share of your active customers can reasonably reach a second tier within a normal purchasing cycle, that tier is invisible to almost everyone it was built for, and the programme reads as designed for a customer segment that does not actually exist in your file.

Expiry policy. Points that expire faster than your typical repurchase cycle punish a normal customer for shopping at their normal pace, which is the opposite of what a retention mechanic should do. Points that never expire remove any urgency but also become a growing balance sheet obligation with no natural ceiling. The right expiry window sits somewhere past your average time between orders, long enough that a normal customer never loses points through no fault of their own, short enough that the outstanding liability does not compound indefinitely.

How much does a customer’s points balance actually cost to carry?

A points balance costs whatever it would take to fulfil every point currently outstanding if every eligible customer redeemed at once, discounted by a realistic estimate of how many actually will. That framing matters because it is easy to treat unredeemed points as free marketing rather than as deferred spend the business has already promised. Outstanding points that customers can redeem for value represent an obligation the business has not yet fulfilled, and that is generally the right way to think about them on your books rather than as a cost that only materialises at redemption. This is not accounting advice; how your specific programme should be recognised and reported is a question for your accountant, not a blog article, but the operating discipline of tracking the balance as a real number starts with your finance team, not with the loyalty platform’s dashboard.

In practice this means someone should be able to answer, at any point, what the current outstanding points balance is worth at your set earn rate, and what a reasonable redemption rate assumption implies for near-term liability. A platform that shows you enrolled members and points issued but does not make it easy to see the outstanding balance in dollar terms is not giving you a complete picture, whatever else it does well.

What does it cost to run a loyalty programme beyond the points liability?

Running a loyalty programme costs more than the platform’s subscription fee, though that fee is the only line item most budgets account for up front. Confirm LoyaltyLion’s current pricing structure directly with the vendor, since plans and how they scale with order volume or active members change over time. Beyond the subscription, factor in the ongoing staff time to monitor programme performance, adjust tier thresholds as your customer base grows, and handle the support tickets a badly calibrated redemption flow generates when a customer cannot work out how to spend their points. None of that shows up on a vendor’s pricing page, and none of it is optional if the programme is still supposed to make sense in a year rather than quietly decaying into a line item nobody reviews.

Redemption itself carries a cost too. Processing a reward at checkout, whether that is a discount code, free shipping, or a physical gift, has an operational cost before you even count the margin given up on the discount itself: someone has to fulfil it, someone has to handle the edge case where a customer tries to redeem twice, and someone has to reconcile the redemption against the liability it just reduced. A programme that is cheap to configure but expensive to redeem against at scale can still cost more per retained customer than a pricier platform with tighter redemption controls.

What does migrating an existing points balance actually involve?

Migrating an existing points balance involves four things that are each easy to underestimate individually: exporting the balance accurately, deciding a conversion rate if the two platforms value a point differently, handling the gap between export and import, and telling customers what happened without it reading as their points were taken away.

The export itself is usually the easiest part. Most loyalty platforms can produce a per-customer balance, though the completeness of what comes with it, tier history, anniversary dates, referral chains, varies, and not everything maps cleanly onto a new platform’s data model. Confirm what LoyaltyLion’s or your current platform’s export actually contains before assuming it is a complete record; a bare points total without the context that earned it is common, and it is worth knowing that in advance rather than discovering it mid-migration.

The conversion rate question only comes up if the old and new platforms priced a point differently, which is common rather than rare, since earn rates and redemption values are configuration choices, not platform standards. If 100 points were worth one dollar on the old platform and the new platform’s redemption value implies 100 points are worth eighty cents, a straight one-to-one transfer either shortchanges the customer or costs you more than the old programme did, depending on which direction the rate moved. Decide the conversion rate deliberately, document the reasoning, and be ready to explain it if a customer asks, because some will.

The gap between export and import is where points earned during the cutover window go missing if nobody accounts for them. Freezing new points issuance during the cutover, or explicitly capturing and applying orders placed in that window after the new platform is live, avoids a customer’s most recent purchase silently not counting. Test the whole sequence in a sandbox or staging environment against a handful of real customer records before running it on the full base, and reconcile the totals before and after rather than trusting that the import ran without checking the numbers on the other side.

Customer communication is the part most teams treat as an afterthought and should not. A points balance that changes value, even for a defensible reason, reads to a customer as the brand quietly taking something away unless you tell them first, explain the reason in plain terms, and give them a way to ask questions. Sending that message after the migration rather than before invites exactly the complaint you were trying to avoid.

What breaks during a loyalty platform migration?

The most common break is a reconciliation gap: the sum of points imported into the new platform does not match the sum exported from the old one, and nobody notices until a customer’s balance looks wrong to them specifically. This usually traces back to points earned in the cutover window, a rounding difference in the conversion rate applied inconsistently across the file, or records for customers who existed in one platform’s export but were filtered out of the import for a reason nobody flagged, such as a missing email match.

Tier status is the second common break. A customer’s tier is often a function of cumulative lifetime spend or a rolling window of recent orders, and if the new platform calculates tier eligibility differently, on a rolling twelve months instead of lifetime spend, for example, customers can lose a tier they had held for a long time even though their points balance transferred correctly. That is arguably a bigger trust issue than a points discrepancy, because status, unlike a points number, is visible every time the customer logs in.

Referral chains and VIP overrides, if your programme uses them, are the third common break, because they depend on relationships between customer records rather than a single number per customer, and relationship data is exactly what a straightforward export tends to flatten or drop. If referral tracking matters to your programme, confirm explicitly, before migration, whether the new platform can import existing referral relationships or whether that history effectively resets.

How do you evaluate LoyaltyLion specifically for a $3M-$30M brand?

Evaluate LoyaltyLion the same way you would evaluate any platform at this stage: against the programme specification you have already written, not against a generic feature list. Ask specifically whether its tier configuration can express the thresholds your order frequency data supports, whether its redemption flow at checkout matches the friction level you want, and whether its expiry settings can be set past your typical repurchase window rather than defaulting to something shorter. Confirm current pricing, plan limits and integration coverage directly with LoyaltyLion, since all three change over time and a competitor’s comparison page is rarely current by the time you read it.

Where LoyaltyLion, like most standalone loyalty platforms, tends to differ from a loyalty module bundled inside a reviews or subscription platform is configuration depth: more granular tier rules, more control over VIP segmentation, more flexibility in how points are earned beyond straight purchase value. Whether that depth is worth the switching cost depends entirely on whether your programme specification actually needs it. A brand that wants a simple points-per-dollar programme with one or two tiers may find that depth adds configuration overhead without changing the outcome, while a brand running a genuinely tiered VIP structure with multiple earn actions may find the same depth is the entire point.

One thing worth naming plainly: how a loyalty platform connects to your email tool, helpdesk, subscription platform and product feed is a separate integration question, with its own failure modes around identity matching and event timing, distinct from the programme design question this article focuses on.

Who should not switch loyalty platforms right now?

A brand that has not yet measured whether its current programme changes behaviour, repeat purchase rate and average order value for enrolled members against a comparable non-member group, should not switch platforms yet, because a platform migration does not fix a programme design problem, and a badly designed programme will underperform on any platform it runs on. Switching first and diagnosing the design second usually means repeating the same mistakes on new infrastructure, at real migration cost.

This article is written for brands doing $3M to $30M in revenue, running Shopify Plus or a paid subscription platform, with an existing points balance large enough that a careless migration would be noticed by real customers. A brand below that floor, or one still deciding whether to run a loyalty programme at all, does not face the migration cost question yet and should focus on the design decisions in the earn rate, tier and expiry sections above before it becomes relevant. If your churn is driven mostly by payment failure or unclear renewal timing rather than a weak loyalty mechanic, the subscription churn dtc consumables benchmark and a subscription churn calculator are more useful starting points than a loyalty platform comparison.

Loyalty and subscription retention sit closer together than most teams treat them: a points balance that keeps growing in value the longer a customer stays subscribed is one of the few mechanics that gives a subscriber a reason to stay past the point where the product itself would have kept them. Pointerflow’s subscription retention work treats programme design decisions like these as retention infrastructure, not a reviews-and-rewards add-on, because getting the earn rate, tier threshold and expiry policy wrong costs more in quiet churn than any platform’s monthly fee.