Know which actions will improve student outcomes, and by how much
Start with one decision, using data you already have. We show it first on synthetic data shaped like yours.
Schools and education partners we work with
Prediction shows risk. Causation shows the way forward.
Most early-warning tools tell you which students are at risk. CML+ estimates which support will actually change the outcome, and by how much.
Who is at risk?
- A risk flag for each student
- Little guidance on which support fits the cause
- Can't tell which interventions actually worked
What will help, and did it?
- Each student's likely root cause, and the expected effect of each support on your menu
- Every recommendation shows its basis: district, research or colleague judgement
- Measured impact, so time and budgets follow what works
How it compares with what you already have
CML+ builds on your dashboards and analysts rather than replacing them, starting from your SIS, MTSS, attendance, nurse and transportation data.
| Compared on | BI dashboards | In-house analysts | Consultants | CML+ |
|---|---|---|---|---|
| Answers | What happened | The questions they have time for | One question, once | Which support changes the outcome, for which students, by how much |
| Keeps learning | No | When staffed | Ends with the report | Each outcome updates the evidence for the next recommendation |
| Who can use it | Analysts and leaders | Analysts | Sponsors | Superintendents, assistant principals, counselors and nurses, each in their own terms |
| Works with | The reports you already build | The data they can reach | Extracts for one engagement | Your SIS, MTSS, attendance, nurse and transportation data |
BI dashboards
- Answers
- What happened
- Keeps learning
- No
- Who can use it
- Analysts and leaders
- Works with
- The reports you already build
In-house analysts
- Answers
- The questions they have time for
- Keeps learning
- When staffed
- Who can use it
- Analysts
- Works with
- The data they can reach
Consultants
- Answers
- One question, once
- Keeps learning
- Ends with the report
- Who can use it
- Sponsors
- Works with
- Extracts for one engagement
CML+
- Answers
- Which support changes the outcome, for which students, by how much
- Keeps learning
- Each outcome updates the evidence for the next recommendation
- Who can use it
- Superintendents, assistant principals, counselors and nurses, each in their own terms
- Works with
- Your SIS, MTSS, attendance, nurse and transportation data
From "who is at risk" to "what will help, and did it?"
Attendance and engagement
Student success and persistence
Literacy and learning programs
Staff capacity
K12 Education Products
Attendance AI carries the parts of attendance work that software can carry, and keeps the human parts with people.
Attendance AI
A causal AI attendance app that learns what brings students back: who to reach, why they are likely missing school, and which support is most likely to help.
- A daily ranked list against the district's own attendance checkpoints
- Likely root causes, and recommendations from the district's intervention menu
- Each recommendation shows its basis: district, research or colleague judgement
- Every outcome flows back, so recommendations shift toward what works
Built with East Central ISD. Screens are illustrative.
What each role gets
Superintendents and district leaders
Which supports and programs to keep funding, and which to stop. Impact estimates you can take to your board. Evidence from a pilot before any larger commitment.
Assistant principals, counselors, nurses (and families)
Why each student is flagged and why each option is recommended. Talking points for family conversations, and plain-language messages for families. Causal reasoning skills built through daily use, not separate training.
District IT, security and data governance
Built and demonstrated first on synthetic data that mirrors your district's structure, before any student data is shared. Works from SIS, MTSS and attendance extracts you control. Every recommendation can be overridden by staff.
How we handle securityWhere these applications are working
Each item is CML Insight's own work with the named partner. Status is shown as it stands today.
East Central ISD: a causal AI app that learns what brings students back
A San Antonio-area district lost its Programs Facilitator, the role that coordinated its response to chronic absenteeism. Together we built Attendance AI, an app that carries the parts of that role software can carry, keeps the human parts with people, and gets smarter every week.
- Challenge
Chronic absenteeism of 23.5% against a district goal of 20% (August 2026 baseline). With the facilitator role eliminated, identification, root-cause work, intervention matching and attendance-coding checks fell to assistant principals who already had two other jobs.
- Evidence
Every recommendation links enterprise evidence (the district's own attendance, discipline, grades, nurse visits and bus routes) with real-world evidence: a 30-study evidence library that includes null and negative findings. Each one shows its basis: District, Research or Colleague judgement.
- Approach
A daily ranked list against the district's own checkpoint table, likely root causes from 13 district categories, and recommendations from the district's closed intervention menu. Impact is measured with several causal methods and validated first on synthetic data with planted effects.
- Continuous learning
Interventions on the menu run as prospective multi-armed bandits: allocation shifts toward what works for each root cause while the district keeps learning, and each outcome updates the evidence for the next recommendation.
- Upskilling
Assistant principals, counselors and nurses see why each student is flagged and why each option is recommended, with talking points for family conversations. Staff build causal reasoning through daily use, not separate training.
Software carries the routine
Identification, root cause, intervention matching, attendance-coding accuracy checks and evaluation.
People keep the relationships
504 meetings, home visits and family conversations stay with staff, and every recommendation can be overridden.
Privacy first
Built and demonstrated on synthetic data that mirrors the district's structure before any real student data was shared.
KidsReadNow
+0.2 SD in 3rd-grade reading (about 7 percentile points) vs. comparison schools.
Challenge
Too little data to judge whether the program worked, and for whom.
Approach
Combined public and private data with fuzzy matching, then used causal AI to estimate impact across student groups and schools.
Result
A measurable reading gain, and staff now able to run their own experiments.
Bring one decision you want to get right.
Book a working sessionStart with a pilot, not a transformation
A low-lift pilot on one campus or intervention, using data you already have: evidence before any larger commitment.
Pick one decision
One campus, grade or intervention you need to get right, such as chronic absenteeism or a literacy program.
Use the data you have
SIS, MTSS and attendance extracts. We first show your views on synthetic data shaped like your district.
Decide on evidence
See which supports worked, for which students and by how much, then decide whether to scale.
Start with synthetic data shaped like your district
See your own dashboards, recommendations and evaluation views before any student data is shared.
Book a working sessionWhat happens next
- We reply within one business day.
- A 45-minute working session works through your decision with you.
- We show a prototype on synthetic data shaped like yours.