IVF Performance Outcome Dashboard
Multi-center fertility analytics for clinical performance benchmarking
An interactive analytics dashboard designed to monitor and compare IVF treatment outcomes across multiple centers, enabling data-driven clinical decisions and performance optimization.
Simulated Data for Demonstration
All clinical data, performance metrics, and outcomes shown are synthetic and created for portfolio demonstration purposes.
At a Glance
The Situation
A fertility group operating 4 UAE centers had a 7.4% cycle cancellation rate. but no way to identify where in the patient journey cancellations were happening or which patients were at risk before treatment began.
What We Built
An interactive multi-center analytics dashboard combined with a machine learning cancellation risk model. giving clinical teams both a rear-view mirror (what happened) and a windscreen (what is about to happen).
What Changed
Center-level performance gaps of 5.5 percentage points were identified and attributed. The ML model detected 82% of at-risk cycles before treatment began. enabling protocol adjustment before cancellation occurred.
The Situation
A fertility group operating four centers across the UAE. Abu Dhabi, Dubai, Sharjah, and Al Ain. was running approximately 687 IVF cycles per year. Each center generated its own cycle data, but there was no centralised system to compare performance, track cancellations, or identify which protocols were driving outcomes and which were not.
The medical director knew that performance varied across centers. What she didn't know was by how much, why, and. most critically. which patients were at highest risk of cycle cancellation before treatment even began.
Two questions sat at the centre of this engagement:
- Where in the patient journey are cycles being lost. and what is causing it?
- Can we identify high-risk patients before the cycle starts, so we can intervene rather than react?
The Business Problem
A 7.4% cycle cancellation rate sounds like a clinical statistic. It is also a business problem.
Every cancelled cycle represents medication costs already incurred, monitoring appointments already used, clinical staff time already spent. and a patient who leaves without the outcome they came for. At a clinic running 687 cycles per year, a 7.4% cancellation rate means approximately 51 cancelled cycles annually.
The deeper problem was that nobody knew where in the cycle those cancellations were happening. Without that information:
- Clinical teams couldn't identify whether the issue was patient selection, stimulation protocols, lab performance, or a combination
- Protocol adjustments were made retrospectively. after the cancellation. rather than proactively
- Center-level performance differences were visible only as headline numbers, not as actionable clinical intelligence
- High-risk patients received the same consultation and the same protocol as low-risk patients, because there was no system to distinguish them
The core insight that drove this engagement:
A cancellation rate is a lagging indicator. By the time it appears in a report, the cycle has already been lost. What the clinic needed was a leading indicator. a way to identify which patients were heading toward cancellation before it happened, so the clinical team could change course.
Where in the Journey Were Cycles Being Lost?
The first step was to stop treating cancellation as a single event and start mapping it to the specific stage of the cycle where it occurred. An IVF cycle can be cancelled at four distinct points. each with a different cause, a different clinical fix, and a different cost.
| Cancellation Stage | What Happened | What the Data Showed | Clinical Implication |
|---|---|---|---|
| During Stimulation | Poor ovarian response. too few follicles developing to proceed to egg retrieval | AMH and AFC were the strongest predictors of poor stimulation response. Patients with AMH below 1.0 ng/mL had dramatically higher cancellation risk. identifiable at first consultation | Protocol adjustment before stimulation begins. not after poor response is observed mid-cycle. AMH-guided dosing decisions at first consultation change outcomes before any medication is administered |
| Pre-Retrieval | Hyperstimulation risk (OHSS). cycle stopped to protect the patient before egg retrieval | Younger patients with high AFC and high AMH were disproportionately represented in this cancellation type. a high-responder profile that can be identified before stimulation | OHSS risk identification at stimulation planning stage. modified protocols and trigger decisions for high-responder patient profiles reduce this category of cancellation |
| Post-Retrieval | No viable embryos available for transfer. fertilisation failure or embryo quality failure | Fertilisation rate (89.7%), blastulation rate (70.7%), and maturation rate (80.5%) tracked at embryologist level revealed performance variation that was not previously benchmarked | Embryology lab quality monitoring and blastulation rate benchmarking. identifying protocol or equipment gaps at the lab level rather than attributing all outcomes to patient factors |
| Administrative | Cycle stopped for non-clinical reasons. scheduling conflicts, financial decisions, or patient choice to withdraw | Tracked separately from clinical cancellations. lower percentage but a distinct pattern that varied by center and consultation process | Operational and patient communication review. this category requires a different intervention from clinical cancellation and should not be aggregated with it in reporting |
Stage-level cancellation breakdown is the analytical output this dashboard is designed to produce. Identifying which stage cancellations are occurring at is the prerequisite for any meaningful reduction strategy.
PHASE 1 OF 2
Business Problem
Fertility clinics operating across multiple centers face challenges in monitoring and comparing IVF treatment outcomes. Without centralized analytics, it becomes difficult to:
- Identify underperforming centers or treatment protocols
- Benchmark doctor-level performance objectively
- Track temporal trends and seasonal variations
- Understand which patient and clinical factors drive success
- Make data-informed decisions to improve success rates
Project Objective
Compare Center Performance
Benchmark clinical pregnancy rates across all locations
Evaluate Doctor Outcomes
Provide transparent, data-driven performance insights
Track Temporal Trends
Monitor success rates by year, month, and season
Segment Clinical Data
Analyze outcomes by age, embryo quality, AMH, and cycle type
Dataset Description
Sample Size: 700 simulated IVF treatment cycles
Time Period: January 2024 – December 2025
Key Variables
Clinical
Center, Doctor, Cycle Date, Year, Month
Patient Demographics
Patient Age, Age Group, BMI
Clinical Markers
AMH, AFC, Previous IVF Attempts
Treatment Details
Embryo Quality, Cycle Type, Fresh/Frozen
Outcomes
Transfer Status, Pregnancy Test, Clinical Pregnancy
Final Result
Outcome (Positive, Negative, Ongoing, Cancelled)
Dashboard Preview
Interactive analytics interface providing real-time insights into IVF performance metrics
Interactive Filters
IVF Performance Dashboard
Multi-Center Fertility Analytics Report
Center-wise Clinical Pregnancy Rate
Success Rate by Age Group
Treatment Outcome Distribution
Embryologist Performance Metrics
Fertilisation Rate
Oocytes successfully fertilised
Blastulation Rate
Embryos reaching blastocyst stage
Maturation Rate
Oocytes reaching maturity
Based on 687 cycles
Disclaimer: All data is simulated for demonstration purposes only. Does not represent actual patient information.
Generated from IVF Multi-Center Performance Analytics Dashboard
Key Insights
- Abu Dhabi center showed the highest clinical pregnancy rate at 46.8% across 206 cycles
- Age group analysis revealed a clear inverse relationship between maternal age and success rates
- Embryo quality emerged as a critical success factor, with 'Good' quality embryos showing 58% higher success rates
- Monthly trend analysis identified seasonal variations in clinic volumes and outcomes
- Doctor-level performance benchmarking revealed opportunities for knowledge sharing across centers
Business Impact
This dashboard delivers measurable value by transforming raw clinical data into actionable intelligence:
Improved Clinical Outcomes
Real-time monitoring enables rapid identification of underperforming protocols and immediate corrective actions
Performance Benchmarking
Center and doctor-level comparisons facilitate best practice sharing and continuous improvement
Resource Optimization
Cycle volume and capacity planning data supports efficient resource allocation across centers
Data-Driven Decisions
Evidence-based insights support clinical decision-making and strategic planning for management
Key Findings From the Dashboard
Center Performance Gap. 5.5 Percentage Points
Clinical pregnancy rates ranged from 41.1% at Al Ain to 46.6% at Abu Dhabi. a 5.5 percentage point difference across the same clinic group using the same broad protocols. This gap was not visible before centralised analytics existed. Once identified, Abu Dhabi's clinical and embryology practices became the reference standard for improvement planning across all other centers.
Cancellation Rate Surfaced as a Trackable KPI for the First Time
A 7.4% cancellation rate. approximately 51 cycles per year at this volume. had not previously been tracked as a performance KPI. The dashboard broke this down by center, protocol, age group, embryo quality, and doctor, making it possible for the first time to identify patterns rather than just react to a headline number.
Embryo Quality as the Strongest Outcome Predictor
Embryo quality emerged as the highest-impact clinical variable across all analysis. 'Good' quality embryos showed 58% higher success rates than lower quality embryos. This directed quality improvement focus toward embryology lab protocols specifically rather than stimulation protocols broadly.
Seasonal Variation Identified and Now Actioned
Monthly trend analysis revealed significant seasonal variation in both cycle volume and clinical outcomes. a pattern that had not previously been captured or acted on. This now informs capacity planning, staffing decisions, and protocol review cycles across all four centers.
This Is Part 1 of 2
The dashboard shows where cycles are being lost across centers, protocols, and patient profiles. Part 2 takes the next step. a machine learning model that predicts which patients are heading toward cancellation before treatment even begins, using data collected at first consultation.
Methodology
Data Aggregation & KPI Calculation
Raw cycle-level data was aggregated by center, doctor, time period, and patient cohort. Key performance indicators (KPIs) were calculated including clinical pregnancy rate, positive test rate, and cancellation rate.
Cohort Filtering & Segmentation
Interactive filters enable dynamic cohort selection by center, doctor, year, cycle type, embryo quality, and fresh/frozen status. Segmentation analysis reveals performance patterns across age groups, AMH categories, and treatment protocols.
Monitoring & Benchmarking Framework
The dashboard serves as a continuous monitoring tool for clinical operations and a benchmarking platform for comparative performance analysis across centers and practitioners.
Tools & Skills Used
What This Project Demonstrates
Does Your Clinic Know Where Its Cycles Are Being Lost?
If you have cycle-level data and a cancellation rate that is not broken down by stage, patient profile, or protocol. you have the same blind spot this clinic had.
We start with a free 30-minute consultation: we look at what data you have, identify where the gaps are, and show you what is possible with what already exists.