Why wallet load cadence predicts repeat visits.

Executive summary.

Here is what the headline numbers say: the global loyalty wallet market for restaurants is valued at $6.8 billion in 2025 and projected to reach $19.2 billion by 2033, compounding at 12.3% annually. Quick-service restaurants already command 35.2% of that market. McDonald's has nearly 210 million 90-day active loyalty users across 70 markets and drove nearly $37 billion in systemwide sales to those members in 2025. In the U.S., the average customer visited 10.5 times per year before joining MyMcDonald's Rewards. After joining, that same customer visited 26 times per year. On paper, the industry has figured out that digital wallets and stored value are the engine of customer engagement, and a lot of corporate boardrooms are reading those numbers and feeling confident about their loyalty roadmap.

Here is what the headline numbers do not say: 78.8% of restaurant guests are gone after one year. The average first-time visitor has a 70% chance of never returning. Customer acquisition costs now range between $27 and $83 per new guest, and those costs are rising 25% to 40% year over year by digital channel. The loyalty wallet market is growing, but the vast majority of QSR operators are still treating stored value as a payment convenience rather than a predictive behavioral signal. They are collecting the float. They are not reading the cadence.

The disconnect between wallet adoption and wallet intelligence is not a technology problem. It is a data literacy problem. And it is costing the industry hundreds of millions in preventable churn. The operators who understand that wallet load frequency is a leading indicator of visit frequency are building compounding customer equity. The ones who do not are simply holding other people's money until those customers disappear.

The growth story corporate is telling.

The macro numbers look extraordinary.

The global loyalty wallets for restaurants market was valued at $6.8 billion in 2025. It is projected to reach $19.2 billion by 2033, growing at 12.3% CAGR. North America leads with 38.9% revenue share, driven by 73.4% cashless payment adoption across restaurant transactions. The QSR segment alone commands $2.39 billion of that market, expanding at 14.2% CAGR. Mobile payments functionality captures 42.1% of wallet feature adoption. These are genuinely extraordinary numbers, and assuming that they reflect underlying operator sophistication would be a reasonable mistake. Except it would be wrong.

The loyalty investment pipeline reflects some of that ambition. McDonald's reported nearly 210 million 90-day active loyalty users at year-end 2025, with systemwide sales to loyalty members reaching nearly $37 billion for the full year. In the U.S., the loyalty effect is measurable and stark: customers visited 10.5 times per year before joining MyMcDonald's Rewards, and 26 times per year after joining. Chipotle has grown its Rewards program to more than 21 million active members, with digital sales representing 36.7% of total food and beverage revenue and approximately 90% of those digital transactions linked to a rewards account. Burger King expanded its Royal Perks loyalty database by 500% in under eight months, with app downloads and signups rising 900% and 10 million diners engaged. These expansion metrics are real, they are funded, and they are being marketed to investors, prospective franchisees, and the financial press.

But here is the tension that almost nobody in a corporate boardroom wants to say out loud: adoption is not the same as optimization. The expansion of wallet infrastructure is a corporate technology story. The behavioral signal inside that infrastructure is a franchisee profitability story. And right now those two things are moving at different speeds. The brands that succeed in actually converting wallet adoption into predictable repeat revenue will be the ones that treat load cadence as a predictive metric, not a passive liability entry.

What is actually happening at the unit level.

The retention engine is under serious stress.

Bloom Intelligence's 2025 State of Restaurant Guest Retention analysis found that 78.8% of restaurant guests churn within one year. For a typical 100-location operation, that represents 548 lost guests per location annually, costing $375,380 in unrealized lifetime value per location. The average guest visits only 1.23 times per month. First-time visitors return within 90 days only 25% of the time. And the guests who do return once show 13x the lifetime value of one-time visitors: $345 versus $26.

The stress is showing up concretely in the economics. Customer acquisition costs now range between $27 and $83 per new guest depending on the channel and market, and they have been rising consistently as digital advertising costs climb and organic social reach contracts. But acquisition cost is the wrong number to be focused on if the retention rate is broken, and in QSR right now the retention rate is broken.

The cost structure of customer loss is the root cause. A franchise operator running a 5% net margin who loses 548 guests per location annually is not just losing revenue. They are losing the compounding lifetime value that would have funded marketing, labor stability, and expansion. When 70% of first-time visitors never return, every dollar spent on acquisition is partially a donation to the churn rate. The operator is not building a customer base. They are renting one.

What is happening as a result is a consolidation dynamic that sounds healthy from a distance but represents real risk to the growth story. Stronger operators with capital and mature digital infrastructure are retaining customers at higher rates. Weaker operators without the tools to read wallet behavior are exiting customers faster than they can replace them. The franchise system is concentrating into fewer, larger hands while the mid-tier operator, the one who historically drives steady location growth, gets thinner. Right now, that environment is prioritizing customer-level survival over expansion targets, and that is a direct constraint on corporate's global development pipeline.

The third-party trap is eating what is left.

Delivery apps are not a revenue channel. They are a margin tax.

On top of the existing retention pressure, QSR franchisees are running a channel that is quietly extracting whatever margin remains. Third-party delivery platforms charge 15 to 30% commission on every order they route, and when you factor in packaging, mandatory promotional discounts, paid visibility fees, and service add-ons, the effective cost per order routinely exceeds 40% of the order value. The customer on DoorDash is not your customer. They are DoorDash's customer. You do not have their wallet data. You cannot see their load cadence. You cannot predict their return.

The math on this is not subtle. A franchise operator running a 5% net margin on a $15 delivery order earns $0.75. A 20% platform commission on that same order costs $3.00. The platform is not reducing margin on that transaction. It is inverting it entirely. The operator loses money. And they are doing it at scale, on every order routed through the platform, often because the alternative is losing the customer to a competitor who is listed right next to them on the same app. It is the definition of a coercive dependency, and it is structural, not tactical.

The industry is aware of this in aggregate. 55% of QSR brands in Qu Beyond's 2025 State of Digital Report identified first-party digital ordering as their single largest revenue growth opportunity, specifically to eliminate the intermediary. Mobile app sales grew 57.2% year over year in 2024 versus 29.8% for third-party delivery in the same period. The brands building owned digital channels are pulling away from the ones that are not, and they are doing it in unit economics, not just revenue.

But here is the corporate dimension of this problem that gets missed: the franchisees being most damaged by third-party dependency are the ones who did not receive adequate digital infrastructure from their franchisor to build a first-party alternative. The operator who does not have a branded mobile app with real stored value capability, real loyalty mechanics, and real attribution is not making a bad business decision by defaulting to DoorDash. They are making the only decision available to them with the tools they were given. Corporate provided the brand, the supply chain, the training. In too many cases, corporate did not provide the technology stack that would have let operators actually own the customer. And that omission is showing up in the P&L of operators who cannot see whether their customers are reloading or drifting.

What this means for corporate's expansion map.

The global growth ambition is running on broken retention economics.

The relationship between customer retention and global expansion is not complicated. Franchisors grow by attracting new operators to sign agreements and open locations. New operators make that decision based primarily on the expected return on their investment, which they evaluate by looking at existing operator performance data, talking to current franchisees, and modeling the unit economics. When the customer base churns at 78.8% annually and the operator cannot predict which customers will return, the conversion rate on prospective franchisees drops. The pipeline gets thinner. The expansion map shrinks.

McDonald's U.S. data shows what is possible when the loyalty infrastructure works. A customer who joins MyMcDonald's Rewards increases their visit frequency from 10.5 times per year to 26 times per year. That is not a marginal improvement. That is a fundamental rewiring of the customer relationship from episodic to habitual. The brands that are genuinely executing on customer retention right now share a characteristic that is easy to overlook: they can see the behavioral signal inside the wallet data. They know which customers are reloading on a schedule and which are letting balances sit dormant. They can intervene before the churn happens.

The brands that are not there yet are the ones treating loyalty as a marketing program instead of a behavioral data system. Circana's 2025 data shows that loyalty members now represent 39% of total restaurant visits, a figure that has roughly doubled since 2019. In 2025, loyalty traffic grew 5% while total restaurant traffic declined 2%. The gap between loyalty-driven growth and non-loyalty decline is widening. The brands that can read wallet load cadence are capturing that growth. The brands that cannot are watching their customer base erode while their competitors compound.

So the corporate question is not whether to expand globally. The corporate question is whether the customer retention economics of the existing system are healthy enough to support the expansion roadmap. Right now, for a lot of brands, the honest answer is no, and the gap between the roadmap and the reality is widening every year that the retention problem goes unaddressed at the system level.

The loyalty and digital data gap.

Franchisees cannot build what corporate has not provided.

One of the most consistent findings across QSR loyalty research right now is that the brands with mature loyalty programs are dramatically outperforming those without them, and the performance gap is widening. The National Restaurant Association's 2025 data shows that more than three-quarters of restaurant customers say they are more likely to visit a restaurant where they can earn points, even if it is not as convenient. The NRA's off-premises research found that 65% of drive-thru users and over 60% of takeout and delivery users say membership affects where they order. Quick-service restaurants generate roughly 71% of sales from repeat customers. Loyalty is not a nice-to-have feature of the customer experience. It is a structural profitability advantage at the unit level, and right now the access to that advantage is wildly uneven across QSR franchise systems.

The gap exists because loyalty infrastructure that actually changes behavior requires a branded mobile app integrated with POS, that runs real-time personalization, that ties in-store and in-app transactions to individual customers, and that generates the data necessary to make intelligent marketing decisions. 70% of QSR loyalty users engage through mobile apps. The app is not a feature of the loyalty program. It is the loyalty program. And building that correctly requires investment, integration expertise, and ongoing management that most individual franchise operators simply cannot do on their own.

This is a system-level responsibility, not an operator-level one. Corporate holds the brand, controls the technology standards, and sets the expectations for the customer experience across every location in the system. If corporate has not deployed the digital infrastructure that lets franchisees build genuine customer relationships and read wallet load cadence as a predictive signal, it is not just leaving operator profitability on the table. It is leaving its own global expansion viability on the table, because the profitability of the franchise system is the foundation of the entire growth model.

Float money, the revenue nobody built into the system.

The cadence is the signal. The balance is just the receipt.

Here is a financial mechanic that the QSR industry's most sophisticated operators figured out years ago and that most franchise systems have still not deployed at scale: stored value float. When a customer loads money onto a restaurant's app, whether through a gift card, an auto-reload feature, or a promotional credit, that money sits on the balance sheet as a liability until it is spent. In the interim it is an interest-free loan. And when some portion of that balance is never redeemed, the brand recognizes it as direct revenue. This is called breakage. It is real, it is consistent, and at the scale of a large franchise system it is worth a lot of money.

But here is what most operators miss: the frequency with which a customer reloads that wallet is one of the strongest leading indicators of whether that customer will return, how soon they will return, and how much they will spend when they do. The balance amount tells you about past behavior. The cadence tells you about future behavior.

A customer with a $100 balance who loaded it once six months ago is a churn risk. A customer with a $15 balance who reloads $20 every ten days is a regular. The high-balance, low-frequency customer should trigger a win-back campaign. The low-balance, high-frequency customer should trigger a frequency bonus or tier upgrade, because they are demonstrating habitual commitment at a level that predicts long-term retention.

The behavioral mechanics behind this are well-documented. When a customer loads funds onto a restaurant app, they have made a financial pre-commitment. The balance creates what behavioral economists call a commitment device: a self-imposed mechanism that locks in future behavior. The customer who has $18.40 sitting in your app is psychologically more likely to visit you than the customer who has to decide, at the moment of hunger, whether to choose you or a competitor.

A customer with a $100 balance who loaded it once six months ago is a churn risk. A customer with a $15 balance who reloads $20 every ten days is a regular. Balance is a lagging indicator. Cadence is a leading one. Dual A wallet intelligence analysis

The most rigorous academic research on this topic comes from a 2026 NBER Working Paper analyzing over 4 million consumers in a large-scale restaurant Pay Now, Buy Later program. The study found that joining a prepaid program roughly doubles customer spending at the merchant over subsequent months. Post-prepayment, customers visit and spend about twice as often as before. But the loyalty effect decays within 6 to 9 months. Approximately 40% of prepaid value is never used. The median firm earns roughly $5.50 in breakage profit for every $1 of bonus credit issued.

What this means practically is that the loyalty effect from prepayment is real but fragile. It doubles visit frequency for active users, but it decays. The operators who win are the ones who do not just collect the prepayment and hope for breakage. They are the ones who monitor load cadence and intervene before the decay sets in. The reload is the signal. The visit is the outcome. If the reload stops, the visit will stop soon after.

Right now, most QSR franchise systems are not deliberately engineering this dynamic. The technology infrastructure has not been deployed at the system level to create the auto-reload mechanics, the gift card integration, the stored value reporting, or the breakage recognition that would let franchisors and their operators capture this revenue and read the behavioral signal inside it. That is not an operator failure. It is a system design gap. And closing it would generate meaningful, structurally recurring revenue for franchisees without adding a single new menu item, marketing campaign, or labor hour.

The attribution problem corporate cannot ignore.

Marketing budgets are being spent without accountability, and finance teams know it.

QSR brands collectively spend hundreds of millions annually on marketing: digital advertising, social media, SMS campaigns, in-app promotions, loyalty incentives, out-of-home placements. And the vast majority of that spend is being evaluated against metrics that are either incomplete, lagging, or entirely disconnected from the actual revenue outcomes they are supposed to produce. This is not a new observation. It is a documented, researched, consistently identified gap that the industry has not solved.

Forrester's 2024 research found that 63% of food and beverage companies cannot link their digital marketing activity to their offline sales. Multi-touch attribution, the practice of crediting every touchpoint in a customer's journey proportionally before a purchase, was classified by Forrester as still an extended use case for most retail and QSR advertisers. That means more than six in ten operators are making budget allocation decisions without knowing which channels are actually driving revenue.

The consequence compounds fast. As of Q1 2025, customer acquisition costs across QSR digital channels had risen 25 to 40% by channel. When budgets are under pressure and attribution is broken, the natural response is to optimize toward the channels that show the best reported performance. But if the attribution is crediting last-touch digital interactions while ignoring the paid social awareness campaign that actually created the intent, the optimization is chasing a measurement artifact, not actual customer behavior. Corporate marketing teams are spending more to learn the wrong lesson, and the budget cycle repeats.

For corporate decision-makers specifically, this is not just a marketing efficiency problem. It is a governance problem. Finance teams at QSR corporate offices are being asked to approve marketing budgets that cannot be traced to specific revenue outcomes. They are approving those budgets because the business requires marketing spend, not because the ROI case is solid. And right now, in an environment where franchisee profitability is already under pressure and expansion targets require every dollar to work, unaccountable marketing spend is a liability, not just an inefficiency.

The technical problem underneath this is structural. QSR customers move between digital ads, SMS, push notifications, in-app offers, and in-store transactions in ways that collapse the attribution window. A customer sees a paid social ad, gets a push notification three days later, and pulls into the drive-thru that afternoon. Most systems attribute the sale to the push notification. The social campaign gets zero credit. The team optimizes toward push, underinvests in awareness, and in six months wonders why new customer acquisition is declining while retargeting performance stays flat. The answer is that they are squeezing an audience that social stopped refilling. Right now this is happening at scale across the industry.

Retention economics matter more than acquisition.

QSRs are paying more to fill a leaking bucket.

Restaurant customer acquisition costs now range between $27 and $83 per new guest depending on the channel and market, and they have been rising consistently as digital advertising costs climb and organic social reach contracts. But acquisition cost is the wrong number to be focused on if the retention rate is broken, and in QSR right now the retention rate is broken.

70% of first-time diners at any given QSR never return. The average restaurant retains only about 55% of its customers overall, well below the 75% benchmark across service industries. QSRs generate approximately 71% of their sales from repeat customers, which means the entire revenue base of the business rests on a pool of loyal regulars that is constantly at risk of attrition, while the system continuously spends $27 to $83 to acquire new customers who have a 70% chance of never coming back. This is not a customer experience problem. It is a structural economic hemorrhage.

The retention math is well established. According to Harvard Business Review research, retaining customers costs 5 to 25 times less than acquiring new ones. A 5% increase in customer retention generates 25% to 95% in profit improvement depending on the model. 72% of QSR customers are more likely to return when personalized offers are deployed. 45% explicitly expect personalization based on order history. None of this is possible without the customer data infrastructure to identify who the customer is, track what they ordered, predict when they are at risk of churning, and deliver a relevant intervention at the right moment through the right channel.

At the corporate level, this retention problem is directly connected to the expansion challenge. Every operator whose customer base churns faster than it grows is running a declining business regardless of what the top-line sales look like. And a declining unit is not an operator who is going to sign an agreement for a second or third location. The prospective franchisees who fuel expansion targets make their investment decision looking at system-wide operator health data. A system with high customer churn, even if it is not visible in aggregate sales numbers, communicates something that experienced investors and prospective franchisees eventually pick up on.

What corporate must build.

This problem is not solvable at the operator level alone.

The challenges described in this report — third-party dependency, loyalty infrastructure gaps, stored value unrealized, attribution blindness, customer retention failure, and the failure to read wallet load cadence as a predictive signal — are not problems that individual franchise operators can solve independently. An operator with three locations does not have the engineering team to build a multi-touch attribution system. They do not have the product team to design an auto-reload stored value mechanic. They do not have the data science capacity to build a churn prediction model based on reload patterns. These are system-level capabilities that need to be deployed by corporate or by a technology partner operating at the system level, then activated across the franchise network.

The operational path forward is clear, and the research on each component is unambiguous:

1. Deploy a first-party digital ordering channel. A branded mobile app with real loyalty and stored value mechanics, as a system-wide infrastructure investment, not a brand-optional add-on. Mobile app sales grew 57.2% in 2024 and that growth is going to the brands that own the channel. Franchisees who do not have this tool are permanently disadvantaged in unit economics.
2. Build loyalty infrastructure integrated with POS and ordering across the entire system. Loyalty programs that cannot identify and reward customers at point of sale are not capturing the behavioral data that predicts churn, enables personalization, or justifies the 20% more spend and 20% more visits that loyalty members reliably generate. The in-restaurant gap is critical: Chipotle links 90% of digital transactions to rewards but only about 20% of in-restaurant business. That 80% invisible majority is where the retention battle is being lost.
3. Engineer stored value mechanics at the system level. Auto-reload, gift card programs, prepaid balance incentives. These are not operator-level features. They require system-wide deployment, financial reporting integration, and breakage recognition methodology that only makes sense designed at the corporate level and activated across the network. The NBER research shows that the loyalty effect from prepayment decays within 6 to 9 months. Auto-reload extends that window by removing the friction of manual decision-making.
4. Implement multi-touch attribution that closes the loop between corporate marketing spend and franchise-level transaction outcomes. Right now, corporate is spending millions on marketing campaigns whose impact on individual franchise unit revenue is essentially invisible. That is not acceptable governance, and it is not acceptable to the finance teams who are approving those budgets.
5. Measure franchise system health in customer retention terms, not just comp sales and net unit count. Retention rate by franchise cohort, customer acquisition cost by marketing channel, loyalty penetration by location, wallet load cadence by market, and stored value balance by customer segment are the leading indicators that predict whether the expansion pipeline will actually hold. Net unit count is a lagging indicator. By the time it moves, the underlying damage has been building for years.

Where Dual A fits.

A platform built for the system, not the operator.

The infrastructure described in the previous section — first-party digital ordering, integrated loyalty, stored value mechanics with auto-reload, multi-touch attribution, franchise-level wallet cadence visibility, and customer data infrastructure that reads load frequency as a leading indicator of visit behavior — is exactly the architecture that Dual A Solutions is built to deliver. And the distinction that matters here is the one between an operator-level tool and a system-level platform. Dual A is designed to be deployed at the corporate or franchisor level, configured to brand standards, and activated across a franchise network, which is the only deployment model that actually solves the problem at the scale where it exists.

For corporate marketing teams, Dual A closes the attribution gap that Forrester identifies as still missing for 63% of food and beverage operators. Multi-touch attribution connecting paid media, push notifications, SMS, in-app promotions, and email to verified transaction outcomes including in-store purchases tied through loyalty membership. For the first time, corporate can answer which channel combination produces the highest-quality customers at the lowest acquisition cost across the franchise system, and allocate budgets accordingly. That is not just marketing efficiency. It is governance.

For corporate finance teams and franchise development leadership, Dual A surfaces the revenue visibility layer that fragmented technology stacks currently obscure: stored value balances and breakage projections at the system level, customer lifetime value by franchise cohort, wallet load cadence by market, retention rates by acquisition channel, loyalty penetration by location, and the financial impact of digital infrastructure investment on unit-level EBITDA. These are the metrics that let corporate set and track financial goals tied to franchisee health, not just aggregate system sales, and that let franchise development teams make evidence-based decisions about where to accelerate and where to stabilize.

For strategic partners evaluating the QSR technology landscape, the thesis here is straightforward: the brands that solve the franchisee profitability problem at the system level through owned digital infrastructure will be the ones that actually execute their global expansion roadmaps. The ones that do not will find the roadmap increasingly theoretical as their operator base contracts, consolidates, and struggles to attract the new franchisees needed to fill the location pipeline. Dual A is positioned exactly at the intersection of corporate strategy and unit economics, and right now that intersection is where the most consequential decisions in QSR are being made.

The QSR brands that execute their global expansion plans in the next decade will be the ones that solved the franchisee profitability problem at the system level today. The technology to do that exists. The decision is strategic. Dual A Solutions · August 2026

Conclusion.

The QSR industry's wallet adoption story is real. A $6.8 billion market, 12.3% annual growth, 210 million McDonald's loyalty users, $37 billion in systemwide loyalty sales, 57.2% mobile app sales growth. The ambition is genuine and the market opportunity is massive. But the growth machine runs on customer retention, and right now customer retention is broken.

78.8% of guests churn within a year. 70% of first-time visitors never return. Acquisition costs are rising while third-party platforms extract the margin that would fund first-party alternatives. The NBER research shows that prepayment doubles visit frequency but the effect decays in 6 to 9 months. The operators who are winning are not the ones with the biggest wallet balances. They are the ones reading the cadence behind those balances and acting on it before the churn happens.

The tools to fix this exist. First-party digital infrastructure that gives franchisees owned customer relationships. Loyalty mechanics that convert that relationship into measurable profit improvement. Stored value programs that generate float revenue and breakage income at the system level. Auto-reload mechanics that extend the loyalty effect beyond the decay window. Multi-touch attribution that tells the truth about which marketing spend is working. Customer data infrastructure that lets corporate see wallet load cadence in real time and intervene before customers reach the point of dormancy. None of this is speculative. All of it is documented, researched, and being executed by the brands that are actually growing their franchise systems in a way that holds. The question for corporate decision-makers is not whether to build it. It is whether to build it before the retention gap gets any wider.

A few things worth clarifying.

Direct answers to the questions operators and corporate teams ask most about wallet load cadence and stored value intelligence.