Katalyst demos
CT-204 All steps
STEP 3 · OPTIMIZE & MONITOR

Your recruitment, live — and always improving

We don't hand you a dashboard and walk away. Our team watches every conversation and continually tunes the messaging, the model, and the workflow to lift conversion — right through to close-out. Live · updated 2 min ago

Patients contacted
0
▲ 186 this week
Reply rate
0
▲ 21% vs. site baseline
Median response
0
to an inbound reply
Qualified
0
▲ 17 this week
Randomized
0
3.5% of contacted

Recruitment funnel · CT-204

This study, to date
Contacted1,240100%
Engaged / replied50841%
Pre-screened31225%
Qualified12810.3%
Randomized443.5%

Channel performance

Reply rate · qualified
SMS46%
Voice AI38%
Email22%
🤝 Warm transfer71%

Conversation library

3,180 analyzed
PatientChannelOutcomePre-screenHandled
Jordan EllisSMS → VoiceQualified · warm-transferredEligibleAI + coordinator
M. OkaforVoiceCallback requested (ES)EligibleAI
D. NguyenSMSScreening bookedEligibleAI
AnonymizedSafetyEscalated to clinicianHuman
R. AlvarezEmailNot eligible · nurturingNot elig.AI

Human-in-the-loop inbox

2 need you
MO
M. Okafor · callback (ES)Voice · qualified, prefers Spanish coordinator
Action
Safety follow-upEscalated 41m ago · clinician reviewing
Review
JE
Jordan EllisWarm-transferred to Dana R. · synced to CTMS
Done
DN
D. NguyenScreening visit booked Thu 10:30
Done
Open the full communication log

What we've optimized

Since go-live
Reworked intro SMS for digital-ad leads +9% replyCold leads needed more context up front; warm leads needed less.
Added evening send window (6–9pm) +14% connectThis cohort engages most after work hours.
Tuned pre-screen question phrasing for clarity −18% drop-offFewer patients abandoning mid-screen.
Model update · adverse-event detection 100% recallEvery safety scenario caught in weekly QA.
Bottom line

Katalyst reached 1,240 patients for CT-204 and handed 128 qualified candidates to the coordinator team — with a +21% reply lift over the site's baseline and a human on every safety-critical conversation.

That's the loop

Design & test → launch → optimize until close-out. Want this running for your study?

Book a demo