Miss Blue Research Preprint · August 22, 2026 iMessage vs SMS response rates # Do blue bubbles *get more replies?* A transparent look at Miss Blue beta-customer observations and the broader reported evidence on delivery, response, revenue, and cost per engagement. [Download the paper](https://missblue.dev/research/miss-blue-imessage-response-rate-study.pdf) [Read the methodology](#methodology) First-party observations ## What participating Miss Blue beta customers reported. These figures compare those customers’ iMessage outreach with their SMS outreach. Revenue means revenue attributed by participating customers to the measured conversations. 01 **80%** ### higher response rates Observed across participating beta customers’ own channel comparisons. 02 **37%** ### more attributed revenue Revenue attribution followed participating customers’ existing measurement practices. **Observational—not causal.** These internal beta results have not been independently audited and do not guarantee an outcome. Audience, consent, timing, message, offer, sales coverage, and attribution method vary. Multi-source synthesis ## The published ranges point in the same direction. Their precision does not. The Miss Blue preprint normalizes reported figures from benchmark compilations, vendor split tests, migration reports, practitioner observations, and a single longitudinal case. It reports ranges instead of pooling incompatible raw data. Reported metric **iMessage** **A2P SMS** How to read it **Delivery rate** **\>94%** **≈68%** Reported benchmark range; SMS comparison assumes registered A2P traffic. **Open rate** **95–98%** **18–22%** Definitions differ materially across sources; SMS opens are commonly inferred. **Reply rate** **25–35%** **2–5%** Warm, consented, iOS-prevalent audiences represented in the reviewed reports. **Cost per send** **$0.02–$0.04** **$0.01–$0.015** Historical reported ranges, not Miss Blue plan pricing or a current carrier quote. Illustrative benchmark model ## Follow the losses through the funnel. This model applies the preprint’s reported midpoints to 10,000 sends. It is an illustration of the published ranges—not a forecast, a randomized Miss Blue result, or a claim that every business should expect these counts. Delivered per 10,000 sends **9,400** *iMessage* **6,800** *SMS* Opened modeled **9,212** *iMessage* **1,360** *SMS* Replied modeled **2,764** *iMessage* **68** *SMS* Reported field comparisons ## Hold the result beside the study design. The comparison gets more credible when audience, creative, and timing stay constant. Missing sample sizes, uncontrolled pre/post periods, and vendor publication bias weaken several rows. | Setting | Design | Sample | iMessage outcome | SMS outcome | Reported difference | | --- | --- | --- | --- | --- | --- | | Retail promotion | Same-offer split test | 10,000/channel | 97% opens | 19% opens | **5× reported open rate** | | B2B SaaS demo invites | Split test | Not reported | 31% replies; 12% booked | 3% replies; 0.8% booked | **≈10× reported replies** | | Auto appointment reminders | Dual-channel campaign | 5,000/channel | 1,410 confirmations | 102 confirmations | **14× reported confirmations** | | Migration cohort | Pre/post channel switch | 14 teams | Example: 19.5% replies | Example: 10.8% replies | **+8–9 percentage points** | Methodology ## A structured qualitative synthesis—not a pooled experiment. The source studies do not expose shared raw microdata or common sampling frames. The report therefore preserves source-level definitions, uses ranges, and treats cross-source directional consistency as evidence rather than calculating a false pooled estimate. 1. 01 ### Benchmark compilations Vendor and industry analyses aggregating campaign telemetry provide the headline delivery, open, and reply ranges. 2. 02 ### Vendor split tests Reported comparisons that hold message or offer constant and vary the delivery channel provide the strongest causal clues in the corpus. 3. 03 ### Migration and practitioner reports Pre/post cohorts and operator observations add real-world context but often lack concurrent controls, raw samples, and complete definitions. 4. 04 ### Miss Blue beta observations Participating customers compared iMessage and SMS outcomes using their own audiences and attribution practices, producing the 80% response and 37% revenue observations. Candidate mechanisms ## Why might the channel change the answer? The evidence identifies plausible contributors, but it does not isolate their individual causal weight. 01 ### Transport and filtering iMessage uses Apple’s data network rather than carrier SMS infrastructure. The reviewed reports attribute part of the observed delivery gap to differences in filtering and routing. 02 ### Primary conversation placement A native thread can be easier to notice and continue than business traffic presented in an unknown-sender or promotional context. 03 ### Familiarity and trust Several sources propose that blue-bubble presentation acts as a familiarity heuristic. The synthesis treats this as a candidate mechanism—not a directly proven psychological cause. 04 ### Richer reply affordances Read state, reactions, media, link previews, and an existing two-way thread may reduce the effort required to understand and answer a useful message. Threats to validity ## Read the caveats before the headline. A useful research page should make the evidence easier to challenge, reproduce, and improve. - Most quantitative inputs come from messaging vendors or practitioner reports with commercial incentives and incomplete sampling disclosure. - The Miss Blue beta figures are observational customer comparisons. They are not a randomized experiment, universal benchmark, or promise of future performance. - Open rate, reply rate, qualified response, delivery, and revenue attribution are not defined consistently across the source corpus. - The evidence concerns warm, consented audiences that skew toward Apple devices. It does not support extrapolation to scraped, purchased, cold, or non-consented lists. - Audience, offer, copy, timing, reply coverage, novelty, device mix, platform policy, and attribution rules can all change the observed effect. Replicate it yourself ## Run a controlled pilot before forecasting the lift. The preprint recommends testing 100–500 warm, consented contacts before migrating a larger workflow. [Start with a sandbox or shared line](https://missblue.dev/pricing) 1. 01 ### Define one eligible cohort Use contacts with the same relationship, consent standard, lead source, device eligibility, and measurement window. 2. 02 ### Hold the workflow constant Keep copy, offer, timing, follow-up count, routing, and human reply coverage as similar as operations allow. 3. 03 ### Choose one primary outcome Predefine unique reply rate, qualified reply rate, booking, resolution, or attributed revenue before inspecting results. 4. 04 ### Report the denominator Preserve eligible recipients, delivered messages, replies, conversions, opt-outs, complaints, missing data, and attribution rules. Paper and source trail ## Read the evidence behind the page. The external links below are vendor-published sources, not independent validation. Their claims are summarized and challenged in the Miss Blue preprint. [**Miss Blue preprint** Eight-page PDF · August 22, 2026](https://missblue.dev/research/miss-blue-imessage-response-rate-study.pdf) [**Tuco AI iMessage Benchmarks 2026** Vendor-reported campaign benchmarks and methodology](https://tuco.ai/imessage-benchmarks) [**Sendblue business messaging API comparison** Vendor comparison of iMessage, RCS, SMS, and WhatsApp](https://www.sendblue.com/blog/business-messaging-api-comparison) [**VoidFix iMessage business communication guide** Vendor discussion of channel characteristics and use cases](https://gateway.voidfix.com/blogs/imessage-business-communication) Explore the platform ## Put the research into a measurable workflow. [For developers ### iMessage API Build two-way blue-bubble conversations into your product or agent. Explore](https://missblue.dev/imessage-api) [For teams ### Message Center Use a complete shared conversation platform without writing code. Explore](https://missblue.dev/features/message-center) [For Apple contacts ### FaceTime Audio calling Call from your blue line without carrier Spam Likely labels. Explore](https://missblue.dev/features/facetime-audio-calling) [For every phone ### Outbound calling Call any dialable number from your Miss Blue business line. Explore](https://missblue.dev/features/outbound-calling) Test the result on your audience ## Start small. Measure honestly. *Keep the reply.* [Create your account](https://missblue.dev/signup) [Explore use cases](https://missblue.dev/use-cases)