Product Strategy for Amplify Assessments

[Your one-paragraph thesis. For example: trustworthy measurement is the foundation, but assessment only creates value when it helps a teacher understand what a student needs and confidently decide what to do next.]

ForProduct Manager, Assessment

AuthorJovanay Carter

Date[Date]

StatusDraft

[Amplify strategy walkthrough video, to record]

01Where Amplify is today

18M+students and teachers across all 50 states

1 platformAmplify Classroom brings curriculum, assessment and intervention together

mCLASSliteracy and math assessment, math now K–8

Bounded AIspeech recognition and analysis of student thinking, with the teacher in charge

[Your read of the strategy: Amplify is moving from selling separate products to running one connected instructional system, where assessment is a capability many products use, not a standalone test.]

Source: Amplify public materials, September 2026. [Add links]

02The loop assessment has to serve

Teach → Observe → Assess → Understand → Intervene → Teach.

Traditional assessment stops at a score. The direction here is question → response → interpretation of thinking → instructional recommendation → intervention. [Where you think the loop breaks today.]

03Six customers, one result

This is not normal SaaS. Every assessment decision has to hold up for all of them at once.

Student: an accessible, fair experience
Teacher: clear, actionable next steps
School leader: implementation and reporting
District: interoperability, reliability, outcomes
State: standards, validity, screening compliance
Learning scientists: psychometric validity

04My strategy: three bets

Bet 01: [Bet title]

Problem: [who struggles, and with what]
What I'd build: [the feature or capability]
How we'd know: [the metric that moves]

Bet 02: [Bet title]

Problem: [who struggles, and with what]
What I'd build: [the feature or capability]
How we'd know: [the metric that moves]

Bet 03: [Bet title]

Problem: [who struggles, and with what]
What I'd build: [the feature or capability]
How we'd know: [the metric that moves]

My first 90 days

30[Listen: teachers, learning science, engineering, the backlog]

60[Diagnose: where the loop breaks, and the first bet to test]

90[Ship: first release, the OKR it moves, what comes next]

05How I think about assessment

01Name the construct. Decide exactly what we are measuring before we write a single question.

02Rubrics are product. A rubric turns a messy human response into a result people can act on.

03AI proposes, people decide. Set quality thresholds and validate before broad release. Humans stay accountable.

04Feedback should teach. A score ends a conversation. A next step starts one.

05Protect integrity. If a result can be gamed, nobody can trust it, least of all the learner.

The loop I design for: question → response → interpretation of thinking → feedback → next practice.

06Same shape, different classroom

Amplify: district → teacher → student
The Dev Difference: university → career program → student

I've already built the core of this problem: turning a complex human response into trustworthy feedback, with AI where it helps and people where they're accountable.

I'd love to talk through this with the Assessment team.

← Back to home Read case study 01 →