Case study format · honest empty state

The format first. The cases will follow.

Most early-stage products fake their case studies — stock photos, "Sarah K." testimonials nobody verified, vague percentages with no methodology. We're not going to do that. This page exists so when the first real case study lands, you know exactly what we'll publish and how we'll verify it.

We're 1 week into the public Orbit launch.

The full /orbit/ surface (this page, the live brain feed, the moat thesis, all 6 module pages, the security doc) shipped in the last 48 hours — Waves 1-5 from the changelog. We have 15 verticals deployed and the first paying customers in pilot, but we don't have a published, verified case study yet.

Our first case study publishes when our first pilot completes and signs off. Until then, the rest of this page is the format we'll use, the verification standard we'll hold ourselves to, and the placeholder cards showing where the cases will go.

▸ Counter at first publish: T-23 days

The format we'll publish — every field, every time.

Every Orbit case study uses the same structure. No mix-and-match, no "depending on the angle we want to play up." This is the same shape security questionnaires use — predictable for the reader, hard to fake for us.

Case study schema

Customer nameReal business name, used with explicit written permission. If the customer asks to be anonymized, we publish the vertical + region only and label it "anonymized at request."
VerticalOne of the 15 BizBot verticals (SiteLine, AgentEdge, etc.) so prospects can pattern-match.
Owner / point-of-contactReal name + role. Linked to LinkedIn or the customer's website where that person is publicly listed.
What they used beforeThe previous tool (Rosie / Smith.ai / Podium / nothing / a real receptionist), with the customer's words on why they switched.
Modules deployedExactly which Orbit modules they turned on (Voice / Recall / Stars / Confirm / Chat / Scout) and which they didn't.
Time-to-valueDays from signup to first measurable outcome (first booking from Recall, first Stars reply approved, first Voice call handled).
Measurable outcomeOne primary metric, with the methodology spelled out. "30% more bookings" is meaningless without "compared to last 30 days, same season, same staffing." We publish the comparison method.
Recorded demo linkA 60-90 second screen-recording of the brain doing the work, on the customer's actual data (with PII redacted). Hosted on YouTube / Loom; not behind a gate.
Owner quoteOne quote, < 50 words, in the owner's voice (we don't ghost-write). They sign off in writing before publish.
What didn't workEvery case study has a "what we'd do differently" section. If a customer says nothing fell short, the case study isn't honest enough to publish.
Independent verificationHow a prospect can verify the claims — usually a contact email for the customer who agreed to take 1-2 reference calls per quarter.

How we'll verify the numbers.

If a case study claims a metric, the methodology has to be checkable. Below is the standard we hold every published outcome to.

Verification checklist (every case study)

  • Baseline period named. Not just "before" — specific: "30 days prior to Orbit deployment, same staffing, same marketing spend."
  • Measurement window named. "First 30 days on Orbit" — not "in our best month."
  • Source-of-truth identified. Which system reported the metric — owner's calendar / Stripe billing / Twilio call log / etc.
  • Confounders acknowledged. If the customer also ran a Google Ads campaign during the window, we say so. We don't attribute lift to Orbit that we can't isolate.
  • Owner-signed. The customer reviews the final draft and signs off in writing before publish. They get veto power on any number we cite.
  • Reference call available. Prospect-to-customer reference call available on request. We never use a published case study from a customer who isn't willing to take a reference call.

Placeholder cards — what they'll look like when filled.

Below are the placeholder cards for the first three planned case studies. Each will be replaced with a fully-fleshed case study at publish time. Until then they show the vertical + format only.

Case 01 · SiteLine · Trades

Trades pilot

Will show: the answering service it replaces, the Voice + Recall + Confirm setup, and before/after call-capture metrics — owner-approved. Publishes when our first SiteLine pilot completes.

Case 02 · AgentEdge · Real estate

Solo agent

Will show: what it replaces (ad-hoc texting + missed-call voicemail), the Voice + Recall + Stars setup, and lead-response before/after. Publishes when our first AgentEdge pilot completes.

Case 03 · VitalCoach · Coaching

Boutique fitness studio

Will show: the booking flow it replaces, the Confirm + Chat setup, and no-show/rebooking before/after. Publishes when our first VitalCoach pilot completes.

Case 04 · CaseFile · Legal

Solo small-firm attorney

Will show: a greenfield Orbit intake deployment and the intake-to-consult metrics we'll track. Publishes when our first CaseFile pilot completes.

Want to be a case study yourself?

Pilot customers get the brain at 50% off for 12 months.

We're looking for our first 2-3 pilot customers across the 15 verticals before locking case study slots. In exchange for letting us publish the case (with your final-draft approval), you get half-off pricing for the first year and direct access to Steve.