Barkbus
First Appointment Cancelled

Keep the charm. Remove the friction.

The current email has strong brand voice and persuasive proof. The opportunity is to acknowledge the cancellation, respond while intent is still fresh, and test whether customers need more persuasion or simply an easier path back to the calendar.

01 · Current email

The starting point

This is the existing “First Appointment Cancelled” email sent three days after cancellation. The review below keeps its strongest brand elements while improving relevance, timing, and the path back to booking.

Original Barkbus First Appointment Cancelled email
Original creative

Strong personality, but the cancellation moment is missing

The visual identity, testimonials, and multiple response options are all useful. The main opportunity is to make the message feel triggered by what just happened—not like a general brand introduction.

Current setup: one email · sent three days after cancellation · no additional touches in the flow
02 · The review

What works, and what I would change

I would preserve the personality and social proof rather than redesigning for redesign’s sake. The biggest gains should come from relevance, timing, and less booking friction.

Keep

What is already working

  • Warm, distinctive Barkbus voice that does not feel like a generic win-back.
  • Strong first-appointment testimonials that reduce risk and support premium positioning.
  • The right value propositions: one-on-one attention, no cages, and grooming at the customer’s door.
  • Multiple ways to respond, including booking online, calling, and texting.
  • Clean single-column structure with an obvious primary action.
Improve

Make the message match the moment

  • Acknowledge the cancellation directly: “Life happens. Let’s find a better time.”
  • Lead with the dog, not Barkbus, and personalize with the pet’s name from the cancellation event.
  • Move the first touch earlier while the customer’s intent and context are fresh.
  • Surface the value props visually instead of burying them in a paragraph.
  • Exit anyone who rebooks and separate customer cancellations from Barkbus-initiated cancellations.
03 · Timing experiment

Test the delay before adding more complexity

The current three-day delay is a hypothesis, not a fact. Keep the email identical and isolate timing as the only variable.

flowchart LR T(["Appointment cancelled"]) --> S{{"Split evenly
Same email, delay is the only change"}} S --> A["4 hours"] S --> B["24 hours"] S --> C["3 days
Current control"] S --> D["7 days"] A --> M B --> M C --> M D --> M M["Measure appointment rebooking
within 30 days"] --> W(["Winning delay becomes
the new default"]) classDef start fill:#2a5bf6,color:#fff,stroke:#2a5bf6; classDef node fill:#fff,color:#202330,stroke:#d9dce6; classDef decision fill:#f2f5ff,color:#202330,stroke:#2a5bf6; class T,W start; class A,B,C,D,M node; class S decision;

Volume guardrail: if the flow does not receive enough monthly entries for four arms, run two rounds: 24 hours versus three days, followed by the winner versus four hours.

04 · Flow design

Add touches only when the data supports them

The expanded journey should respond to who cancelled, why they cancelled, whether they can receive SMS, and whether they have already rebooked. Every message needs a clear exit rule.

flowchart TD C([Appointment Cancelled event]) --> V{Required event properties present?} V -->|No| Q[Send to data QA queue
Do not enter automated flow] V -->|Yes| R{Who initiated cancellation?} R -->|Barkbus| BA[Apology + priority rebooking
Send within 1 hour] BA --> BX{Rebooked?} BX -->|Yes| E([Exit flow]) BX -->|No| BS[Personal support follow-up] R -->|Customer| X{Booked again since cancellation?} X -->|Yes| E X -->|No| Y{Cancellation reason} Y -->|Scheduling / unknown| EM1[Email 1
4–24 hours
Easy rebook + reply for help] Y -->|Price| PV[Email 1 variant
Explain value before discounting] Y -->|Dog anxiety / service concern| SC[Service reassurance
What to expect + human support] EM1 --> RC{Rebooked?} PV --> RC SC --> RC RC -->|Yes| E RC -->|No| SMS{SMS consent?} SMS -->|Yes| SM[Day 3 SMS
Two-way help + booking link] SMS -->|No| D2[Wait until day 5–7] SM --> R2{Rebooked?} R2 -->|Yes| E R2 -->|No| D2 D2 --> EM2[Email 2
Social proof + first-visit reassurance] EM2 --> R3{Rebooked?} R3 -->|Yes| E R3 -->|No| F[Day 12–14 final test] F --> AB{A/B test} AB -->|A| OFF[Soft incentive or waived fee] AB -->|B| ND[No-discount urgency
Local availability / founder note] OFF --> R4{Rebooked?} ND --> R4 R4 -->|Yes| E R4 -->|No| N[Move to broader nurture / win-back] T[(Track: cancellation source,
reason, dog name, city, service,
appointment value, SMS consent,
rebook timestamp, message attribution)] C -. required tracking .-> T EM1 -. performance .-> T SM -. performance .-> T EM2 -. performance .-> T F -. performance .-> T classDef start fill:#2a5bf6,color:#fff,stroke:#2a5bf6; classDef decision fill:#f2f5ff,color:#202330,stroke:#2a5bf6; classDef action fill:#fff,color:#202330,stroke:#d9dce6; classDef data fill:#f7f8fb,color:#555b6e,stroke:#858b9b,stroke-dasharray:4 3; class C,E start; class V,R,X,Y,RC,SMS,R2,R3,AB,R4,BX decision; class Q,BA,BS,EM1,PV,SC,SM,D2,EM2,F,OFF,ND,N action; class T data;

Core safeguards: suppress anyone who rebooks, separate Barkbus-initiated cancellations from customer cancellations, honor SMS consent and quiet hours, and use Smart Sending so email and SMS do not stack unnecessarily.

05 · Creative experiment

Persuasion versus convenience

Once timing is established, test two genuinely different ideas. Variant A builds confidence in the service. Variant B removes steps by putting available appointment windows directly in the email.

06 · Measurement

Judge the flow on bookings, not opens

The primary outcome is a completed appointment after cancellation. Engagement metrics are diagnostic, but they are not the business result.

30-day rebooking rate
Median time to rebook
Revenue per flow recipient
Conversion by delay and variant
SMS-assisted bookings
Unsubscribe and opt-out rate