Methodology: IMS Missed-Call Study (2026)
This document explains the sampling, data reconstruction, response classifications, callback attribution and limitations behind IMS Research Dataset v1.0.
Design and sample
Observational mystery-shop exercise involving 222 unique U.S. med-spa and aesthetic-wellness businesses. A trained IMS representative placed 467 outbound attempts between July 27 and July 29, 2026, during business hours. The business, not the call attempt, was the unit of analysis. Inbound responses were tracked through August 5, 2026. The sample was not selected to be nationally representative.
Source records and reconstruction
Close CRM activity history, caller notes, call dispositions and inbound SMS/call timestamps were reviewed read-only. Dataset v1.0 was frozen August 21, 2026. The calling line did not record audio; zero of the 467 attempts have a recording. The team reviewed notes across all attempts to a business, not merely the longest call.
Live-answer rules
A live human answer required an explicit note documenting a conversation. Explicit no-answer or busy notes, IVR, auto-attendant and voicemail transfers were classified as no documented human answer. A technical “answered” disposition alone was not proof of a person. Four businesses with no useful notes and ambiguous durations remained indeterminate.
| Classification | Businesses |
|---|---|
| Confirmed human conversation | 81 |
| No documented human answer | 137 |
| Indeterminate | 4 |
| Total | 222 |
The 137 no-human classifications represent 61.7% of all 222 businesses, or 62.8% of the 218 determinately classified businesses. The publication uses the full-cohort percentage unless otherwise noted.
Voicemail and confounding
Calls were intended to be made without leaving voicemail. When a voicemail was left or a live conversation occurred earlier, later responses were flagged as confounded rather than counted as clean missed-call-text callbacks. Such records remain in the evidence package.
SMS classification
Twenty unique businesses generated an inbound SMS following a missed or unanswered call, deduplicated to one per business. Eighteen appeared in the July 27–29 call window and two in the August 3–4 follow-up. Nine first texts were clearly automated based on arrival timing (roughly 21–95 seconds) or content. The content categories were: eight invitations to text, five acknowledgements, three generic automated replies, one booking link, and three unclear.
Callback attribution
Three of the 20 SMS businesses had an inbound callback candidate. One qualified as clean because there was no prior live conversation or voicemail confound. Two were preserved as confounded and excluded from the clean count. In the clean case, the missed call was logged July 28 at 19:16, the SMS around 19:17, and the callback July 29 at 15:34, approximately 20.3 hours later. The timestamps come from the CRM record; no typical callback time is inferred from one case.
Reconciliation of field notes
The original calling report summarized approximately 220 businesses, 48 human answers, 25 missed-call SMS responses and two callbacks. Read-only CRM reconstruction established 222, 81, 20 and one clean callback respectively. Five field-reported texts had no locatable CRM record. The field report remains documented as historical evidence but is not the source of the published statistics.
Known limitations
- No audio recordings to independently establish human answers.
- Four answer classifications remain indeterminate.
- Five reported texts could not be located in Close CRM.
- Two callback candidates had prior-contact or voicemail confounds.
- Short calling period and nonrepresentative sample.
- No measured consumer buying intent, missed revenue, sales conversion or causal effects.
Dataset governance
IMS Research Dataset v1.0 was frozen August 21, 2026, with 222 business-level records and 467 call attempts. Future corrections should be documented in a versioned changelog. The underlying business-level cohort and personally identifiable contact records are not publicly released on this page.
Publication note: This methodology describes the completed frozen audit. It does not claim that all med spas behave the same way, that SMS follow-up is ineffective in general, or that observed missed calls caused lost revenue.