A loyalty program does not stall for lack of projects. It stalls when nobody owns the question a project is supposed to answer.
The head of loyalty at a national retailer leaves the quarterly review with three questions penciled on his deck: why redemptions slowed, which partners to renew, and which of his best members went quiet. A week later, three proposals sit in his inbox. One pitches a new app, one a tier redesign, and one a partner marketplace. His budget covers one.
All three are sound projects, and each one answers somebody else’s question. None of them starts from the three he wrote down.
Loyalty teams show up with questions, and most of what reaches them arrives as a project.
That gap is what loyalty program management exists to close. The job is to ask the right questions on a schedule, with an owner for each, and let the answers pick the projects.
Loyalty program management is the work that starts after launch
Loyalty program management is running a loyalty program after launch: setting its goals, keeping members active, managing partners, owning the economics, and changing it on purpose.
Design builds the program. Points or tiers, earn and burn rules, and rewards for actions beyond purchase are all design choices, and the loyalty program design guide covers them. Management starts the day after launch and runs for as long as the program does.
A launch is an event. Management is a habit.
The first thing a head of loyalty designs after launch is the review calendar, and the second is the name of the person who owns each question on it.
A project earns its budget when a question picked it
Proposals come shaped as projects because a project can be scoped. A new app fits a statement of work, the contract that fixes a project’s deliverables and price, and so does a relaunch. A standing question about drifting members fits a calendar instead, which is why it rarely arrives in a proposal.
Research by Forrester Consulting for Zeta Global shows the reflex: 92% of the organizations it surveyed planned to invest in new technology or processes to use their loyalty data better. The same study named what stands in the way: data quality issues, fragmented systems, and organizational silos, where each team keeps its own data and its own goals.
Technology can connect a fragmented system, but it cannot decide who owns the answer when redemptions slow. A silo is a question with no owner, and no purchase order fixes that.
Members send mixed signals of their own. Bond’s 2026 research with Visa found engagement at peak levels, alongside a growing gap between the loyalty brands can measure and the loyalty that drives growth. Enrollment and earning show up on the dashboard. Preference, the choice of one brand over another, shows up later in the profit and loss statement, or P&L.
Projects still matter. A new app or a tier redesign can be exactly the right answer, once a diagnosis has asked for it. Software makes the questions answerable, but the operating model, meaning the reviews and owners around the program, decides which answers become projects, and that model is the head of loyalty’s to build.
Listed loyalty businesses answer their questions in public
For a picture of the discipline at full size, read an investor filing.
Qantas and IAG both report their loyalty businesses as segments of their own, the separate business lines a listed company reports results for, and each segment has a profit target attached. That is not a reporting detail. It is a statement about what loyalty has become: an enterprise growth engine that has to explain its numbers to the people who fund it.
IAG’s half-year report shows the habit at work. Avios issuance, the points IAG hands to members, outran redemptions, and the company explained the gap on the page: popular Middle East destinations were unavailable for members to redeem on. The slower number arrived with its causes attached, which is diagnosis done where investors can read it.
Best Buy shows the same discipline in a single design change. When it added reward points to its paid memberships, it pointed to what its customers want: to earn points when they shop there. The change started from something customers had already told the company, which is where a funded project should start.
This is why loyalty program management is a finance conversation as much as a marketing one. A program carries partner revenue and the cost of rewards. It also carries a liability for every point members have yet to redeem, because each of those points is a reward the company still owes. A team that manages by project hands the executive committee a string of launches. A team that manages by question hands it a cause and a decision, which is the report worth bringing to every quarterly review.
A managed program runs on a calendar
Every standing question needs a rhythm and an owner, and every answer should produce something a person acts on. Loyalty Juggernaut’s own list for airline programs shows what such questions sound like. It opens with which high-value members are showing the first signs of disengagement, and it closes with where program economics are deteriorating before the change reaches quarterly reporting.
Each week, loyalty operations checks whether enrollment, earning, or redemption moved outside its usual range, and which offers, partner feeds, or accounts behaved oddly. The output is an exceptions list, with an owner and a fix date for each item.
Each month, loyalty marketing and analytics ask which members are drifting and which offers changed behavior. They also ask which offers paid for purchases that would have happened anyway. The review ends in two or three diagnoses, and each one ends in an action or a decision to wait.
Each quarter, the head of loyalty sits down with finance. Together they ask which partners create value and what last quarter’s projects delivered against their business cases, the returns each project promised when it was funded. The output is a ranked project list and the program P&L. Once a year, the program owner and the executive sponsor ask whether the value proposition is still the reason members stay. They also ask whether the platform can answer the team’s questions, then set the plan and budget.
A question with an owner becomes a decision. A question with no owner becomes a slide.
The Qantas and IAG reports read like the quarterly and yearly reviews, written for investors. In any program, the weekly and monthly reviews are where numbers like these get made.
Each role in that calendar makes its own calls. Analytics designs the holdouts, groups of randomly selected members kept out of a change so their results show what would have happened without it, and decides whether a result is solid enough to act on. Finance owns the points liability and the breakage estimate behind it, and sets the reward budget the program can carry.
Five numbers carry most of these questions: the active member rate, the redemption rate, breakage, the points liability, and incremental lift, which is the difference between what members who got a change did and what the holdout did. The head of loyalty owns all five and decides which projects get funded, which is the decision the rest of the calendar exists to feed.
Loyalty management software keeps the program’s records
A loyalty management system is the software that runs a program day to day. It holds each member’s balance and status and applies the rules to every purchase and redemption.
A CRM, or customer relationship management system, does a different job. It keeps the record of the customer relationship, from contact details to service history. The loyalty system keeps the record of the program, from its rules to the points still owed, and the two work best connected.
The test for any platform is whether it can answer the calendar’s questions without a project. If every rule change needs a developer, the monthly review stalls. Infrastructure sets the pace too: WestJet’s year-end rollover on GRAVTY® takes 28 hours, down from 10 days on Siebel. The guide to loyalty management software sorts the main platforms by how they are built and who they serve. Before signing, a loyalty team should put its own standing questions to each shortlisted platform, because a system that cannot answer them turns each question into a project.
Five rules keep every project tied to a question
- Ask on a schedule. Put the standing questions on a weekly, monthly, quarterly, and yearly calendar, with one owner for each. A question asked only after a bad quarter gets a rushed answer, while a question asked every month builds a trend.
- Diagnose to a cause before choosing a fix. A slow redemption rate has a short list of usual causes, so name the cause first and then pick the fix. IAG named its causes in public, and a monthly review can do the same in private.
- Rank projects by the value at stake. Size each fix by the members, spend, and liability it touches, then fund from the top. Some fixes turn out to be rule changes or partner conversations, which cost little. The best pitch does not get a vote.
- Measure every change against a holdout. Hold out a randomly selected group of members from each change and compare the two groups over the same weeks. A spike in redemptions is not proof. The gap between the two groups is. The loyalty program ROI guide sets out the method.
- Give the program a P&L and one owner. Partner revenue, reward costs, the cost to run, and the liability for unredeemed points belong in one statement, owned by one executive with finance at the table. IAG defines its operating segments by how it treats its businesses and allocates resources, and IAG Loyalty is one of them. A program that reports like a business gets managed like one.
GRAVTY® puts the diagnosis ahead of the budget
In GRAVTY®, the retailer’s three questions become three investigations. Agentic AI Compass, launched in October 2025, works like a team of expert analysts. Its specialized AI agents cover program performance, sentiment, anomalies, benchmarks, and offer tests, and they run the three investigations in parallel.
Say the redemption slowdown traces to one reward category, where a partner’s catalog ran thin. The fix is not a project. It is a conversation with that partner.
The partner comparison comes back next: two partners send members who redeem and return, and one sends points that sit unused, which settles the renewal list. Patented multi-dimensional behavior tracking supplies the evidence: first-party data, collected directly from members, on what each one did before and after going quiet. The team asks Compass why activity fell among the program’s best members and compares them with the members who stayed. Compass then runs a win-back offer for the quiet ones through a simulation before any member sees it.
When a fix is a rule change, GRAVTY Visual Rules, the patented no-code rules engine, lets the loyalty team make it directly. AI-Sense monitors the program and alerts the team to anomalies, which feeds the weekly exceptions review.
His budget still covers one project. He funds the fix the diagnosis supports: a win-back program for the members who went quiet, measured against a holdout from its first week. All three proposals wait for a question of their own.
Members feel a well-run program as attention
None of this machinery is visible to the member.
She never sees the quarterly review. What she notices is that the rewards she wanted are back in the category she shops, and that an offer arrives while her habit is still warm. She earns with a partner that sends her somewhere she wants to go.
To her, the program seems to notice. That feeling comes from a calendar with an owner for every question, and building that calendar is the part of the program a loyalty team controls.
Members do not reward a program for being busy. They reward it for paying attention.
The numbers behind loyalty program management
These are the figures and formulas behind this essay. Each figure comes from the company’s own release or report, or from the study’s published findings.
Research on loyalty data and member engagement
| Study | What it found | What it means for a loyalty team |
|---|---|---|
| Forrester Consulting, commissioned by software vendor Zeta Global, May 2026 | More than 300 marketing and loyalty decision-makers in B2C industries. 92% planned to invest in new technology or processes | The named barriers include organizational silos, which an owner fixes and a purchase does not |
| The Bond Loyalty Report, with Visa, June 2026 | Engagement at peak levels; 85% of consumers more likely to keep doing business with a brand that has a loyalty program, and 73% spending more | Participation reads strong on the dashboard; the report points to a growing gap between the loyalty brands can measure and the loyalty that drives growth |
Disclosures from listed loyalty businesses
| Company | What it disclosed | What a loyalty team can benchmark |
|---|---|---|
| Qantas Loyalty, full-year results, August 27, 2026 | Underlying EBIT up 12% to A$625 million. Active members up 6%. Members earning in two or more categories up 8%. Uber the fastest-growing partner, with more than 1 million members earning points on rides and deliveries. On track for a 2030 target of A$800 million to A$1.0 billion | A loyalty segment with its own profit target, with growth reported through active members and multi-category earners |
| IAG Loyalty, half-year report, July 31, 2026 | Avios issuance up 15%, driven mainly by non-airline partners, especially financial services. Redemptions up 6%, helped by more Reward flights. Drag from unavailable Middle East destinations. Objective of €1 billion in operating profit in the medium term | Issuance and redemption reported side by side, with the causes of the gap named |
| Best Buy, reward points for paid members, June 2026 | Design change: reward points added to the paid My Best Buy Plus and Total memberships, at 1% back in rewards on eligible purchases, or 6% with the My Best Buy Credit Card. Stated reason: more than 80% of its customers want to earn points when they shop there | A design change tied to a stated customer demand, with the earn rate set by payment method |
The management calendar by rhythm
| Rhythm | The questions | Owner | Output |
|---|---|---|---|
| Weekly | Did enrollment, earning, or redemption move outside its usual weekly range? Which offers, partner feeds, or member accounts behaved oddly since the last review? | Loyalty operations | An exceptions list, with an owner and a fix date for each item |
| Monthly | Which members are drifting? Which offers changed behavior, and which paid for purchases that would have happened anyway? | Loyalty marketing, with analytics | Two or three diagnoses, each ending in an action or a decision to wait |
| Quarterly | Which partners create value? What did last quarter’s projects deliver against their business cases? | Head of loyalty, with finance | A ranked project list and the program P&L |
| Yearly | Is the value proposition still the reason members stay? Can the platform answer the questions the team asks? | Program owner, with the executive sponsor | The program plan and budget, including any redesign |
Each role’s review and decisions
| Role | What it reviews | What it decides |
|---|---|---|
| Head of loyalty | The program P&L and last quarter’s projects against their business cases | Which projects get funded, and which questions go to the executive sponsor |
| Loyalty marketing | Member activity by segment and offer results against holdouts | Which offers run and which drifting members get a win-back |
| Analytics | Holdout design and the diagnosis behind every number that moved | Whether a result is solid enough to act on |
| Partner manager | Points earned and redeemed by partner, and each partner’s billings | Which partners to grow and which to renegotiate or exit |
| Finance | The points liability and the breakage estimate behind it | The liability assumptions and the reward budget the program can carry |
| Loyalty operations | The weekly exceptions and fraud flags | Which rule fixes ship and who owns each exception |
Five numbers and their formulas
- Active member rate: active members divided by enrolled members over a set window, such as twelve months.
- Redemption rate: points redeemed divided by points issued over the same period. A falling rate is the first question in the retailer’s deck.
- Breakage: the share of issued points that members never redeem. Over a program’s life it is the mirror image of the redemption rate.
- Points liability: points outstanding, times the value of a point, times the share members are expected to redeem. It sits on the balance sheet, and finance owns the estimate.
- Incremental lift: what members who got a change did, minus what a randomly selected holdout did over the same weeks. It is the number that shows what a change caused.
Checks before choosing a loyalty management system
- Fit to the program: a program with partners needs partner earning and settlement built in.
- Rules the team can change: if every rule change needs a developer, the monthly review stalls.
- Answers at the level of the question: reports by segment, offer, and partner, with holdout groups built in.
- Records finance can use: every earn, burn, and expiry kept at the transaction level, so the liability starts from real data.
- Integration and migration: the work to connect point of sale, app, CRM, and partners, and the cost of moving members off the current platform.
Frequently asked questions
What is loyalty program management?
Loyalty program management is the ongoing work of running a loyalty program after launch. The team sets goals and tracks members and partners. It answers for the program’s economics and changes the program through measured projects. Management runs it on a weekly, monthly, quarterly, and yearly calendar, with an owner for each question.
Is a loyalty program a CRM?
No, a loyalty program is not a CRM. A CRM is the system of record for the customer relationship, from contact details to service history. A loyalty program is a value exchange with its own currency, rules, and partners, and it carries a liability for every point outstanding. The loyalty management system that runs it keeps each member’s balance and status and applies the program’s rules to every transaction. The two systems work best side by side.
What does a loyalty program manager do?
A loyalty program manager runs the program between launches and owns its results. She sets goals with the business and watches member and partner performance every week. When a number moves, she diagnoses the cause before anyone picks a fix. Then she ranks the projects that follow and measures each one against a holdout group.
Who should own a loyalty program’s P&L?
One executive should own the loyalty P&L, such as the head of loyalty, with finance as a standing partner. The statement puts partner revenue and reward costs beside the cost to run the program and the liability for unredeemed points. Qantas and IAG report their loyalty businesses as segments with profit targets. Qantas Loyalty grew Underlying EBIT 12% to A$625 million in fiscal 2026 and targets A$800 million to A$1.0 billion by 2030. IAG Loyalty aims for €1 billion in operating profit.
How do you know if a loyalty program is working?
A loyalty program is working when members who get a change do better than a randomly selected holdout that went without it. Compare the two groups over the same period. The difference is what that change caused. The holdout shows what would have happened anyway. The loyalty program ROI guide walks through the method.
Why do loyalty redemption rates fall?
Redemption rates fall when the rewards members want run short or their points buy less than before. IAG Loyalty’s half-year report to June 30, 2026 shows the first cause. Avios issuance rose 15%, while redemptions grew 6%. IAG said redemptions were held back by the absence of popular Middle East destinations. For a loyalty team, a falling rate belongs in the monthly review: find which reward categories or partners ran short before anyone proposes a fix.
What software do you need to manage a loyalty program?
Managing a loyalty program takes a loyalty management system: the software that holds each member’s balance and status, applies the program’s rules to every purchase and redemption, and reports on activity and cost. It connects to the point of sale, the app, the CRM, and any partners. Before choosing one, check that the team can change rules without a developer, that reports reach the level of the segment, offer, and partner, and that the transaction records support the liability finance has to book.
