Case Study
·Food & Beverage · Single Location
Straws Smoothie Bar
A single-location smoothie and juice bar with ~3,800 customers and no systematic retention strategy. Here's what happened after connecting ReturnFlowHQ.
[X]%
Revenue lift (attributed)
[X]%
Lapsed customer reactivation
[X]x
ROI on program spend
Before ReturnFlowHQ
The starting point
Straws had a healthy Square customer list (~3,800 records) and solid foot traffic, but no structured way to bring lapsed customers back. There was no outbound messaging program — customers who drifted away simply stayed away.
~3,800
Customer list
[X]%
Lapsed customers (60+ days)
$[X]
Baseline monthly revenue
Intervention
What ReturnFlowHQ did
Square POS connected in under 5 minutes
OAuth connection imported all 3,800+ customers and full order history automatically. No manual data entry, no CSV exports.
AI segmented lapsed customers
The platform identified customers who had not visited in 60+ days and built a win-back segment automatically based on purchase history and visit patterns.
Personalized campaigns launched
Daily AI-written messages went out to the win-back segment and active customers on the optimal send schedule. Each message was personalized using the customer's actual purchase history.
Revenue attributed using 7-day window
Every Square order from a customer who received a message in the prior 7 days was tracked and credited to the campaign. See our full attribution methodology.
Results
Month-by-month
Month 1
Square sync & baseline
Connected Square POS. ~3,800 customers imported. First 30 days used to establish baseline attribution metrics without active campaigns.
Month 2
First campaigns launch
[PLACEHOLDER] First win-back campaigns sent to lapsed segment (60+ days inactive). [X] messages sent, [X]% open-rate equivalent measured via attributed orders.
Month 3
Steady-state results
[PLACEHOLDER] Campaigns running on full auto-schedule. Attributed revenue: $[X]. Campaign cost: $[X]. Net return: $[X].
Ongoing
Continuous optimization
AI learning layer refines send timing and message strategy per customer cohort. Results tracked monthly against the 7-day attribution window.
ROI Breakdown
The math, shown openly
All numbers use the ReturnFlowHQ attribution methodology: completed Square orders within 7 days of an outbound SMS, deduplicated, with refunds removed.
Avg monthly attributed orders: [X]
Avg order value: $[X]
Attributed revenue/mo: $[X]
Plan + SMS cost/mo: $[X]
Net return/mo: $[X]
ROI: [X]%
[PLACEHOLDER — replace all [X] values with real verified numbers before publishing]
Skeptic Test
What a skeptic would ask
“Wouldn't they have come in anyway?”
Some, yes. The 7-day attribution window does not claim 100% causation — it captures customers who visited shortly after a message. The control question is: were these customers trending toward a visit before the message, or were they in a lapsed state? For the 60+ day segment, most had not visited in 2+ months before the campaign.
“Is $[X]/month sustainable, or was this a launch spike?”
[PLACEHOLDER — add context on month-over-month consistency once 3+ months of data are available]
“How do we know the phone matching is accurate?”
ReturnFlowHQ matches on normalized E.164 phone numbers from Square's customer records. The same number that is in the order record is the number the message was sent to. There is no probabilistic matching.
“[Owner quote — pending sign-off]”
Want results like this for your location?
Every location gets the same Square sync, AI segmentation, and attribution dashboard. Setup takes under 5 minutes.