Pilot structure, data governance, teacher visibility, institutional fit, and what you need to evaluate Guided Scholar for your school or district.
A pilot is a structured, limited deployment, typically one or two classrooms over a defined period, most often a semester or a targeted instructional unit. It is designed to answer a specific question: does Guided Scholar improve writing quality, revision behavior, and student engagement with feedback in your school's context?
In a school-account pilot, Guided Scholar operates as teacher-connected instructional support. Student activity is class-linked, visible to the responsible teacher, and bounded by the assignment context. Students are not accessing a general-purpose AI tool, they are working within a structured feedback workflow that the teacher can see in real time.
The pilot requires one participating teacher, student access to the system during class or at home, and a brief check-in process at the midpoint and end. There is no dedicated IT infrastructure required for the current phase. The system runs through a web browser.
Participating schools need a willing teacher, students with reliable internet access, and a designated point of contact for the evaluation period. No additional software, licensing, or infrastructure is required during the pilot phase.
Formal data processing agreements and compliance documentation are in development alongside the full teacher and administrator account system. Schools with specific requirements (FERPA notices, data processing agreements, or board-level disclosure requirements) should contact us before the pilot begins so those can be addressed directly.
The recommended pilot window is one grading period, typically six to nine weeks, long enough to observe meaningful revision behavior and writing improvement across multiple assignment cycles. A shorter pilot can demonstrate how the system works, but it is rarely enough time to observe the revision patterns that distinguish students who engage with feedback from those who don't.
Pilots tied to ACT preparation can be structured more tightly around a specific testing window.
A successful pilot produces enough evidence to make an informed decision about broader deployment. That evidence should include teacher observations, student revision data from the dashboard, and if the pilot is tied to ACT preparation, pre- and post-practice score comparisons.
At the conclusion of the pilot, we provide a structured debrief covering what the data shows, what teachers and students reported, and what a broader rollout would require. The goal is a decision you can defend, not a sales conversation.
Pilots are conducted without cost. The purpose of a pilot is mutual, allowing the school to evaluate Guided Scholar under real classroom conditions, and it provides us with the observational data needed to refine the system before broader deployment.
Pricing for post-pilot deployment is being finalized and will be structured at the school or district level. Submit a pilot request → or contact us directly to discuss what a sustainable cost structure would look like for your context.
Guided Scholar collects a display name and grade level to support the student account. It collects the text of student submissions and the feedback generated in response. It does not collect home addresses, precise location data, demographic information beyond grade level, or any data unrelated to instructional use.
No student data is sold, shared with third parties, or used for advertising. Student submission data is retained for six months in active storage, archived for a further six months, and then permanently deleted.
No. Student submissions are used only to generate feedback for that student in that session. They are not used to train, fine-tune, or retrain any AI model. Rubric and scoring calibration is conducted using purpose-built benchmark essays created for that purpose, not student work.
AI processing occurs through a controlled API operated by Brau Consulting LLC. Student submission text is transmitted to generate a feedback response and is not stored, retained, or shared by any third-party AI provider beyond that single exchange. The API provider does not retain student data or use it for model training.
Full technical documentation on data flow and API provider agreements will be published prior to any non-pilot deployment.
Guided Scholar is designed with FERPA requirements in mind. The system collects only what is needed for instructional purposes, does not share student data with unauthorized parties, and supports teacher oversight and visibility into student work.
Formal FERPA compliance documentation (including a complete privacy policy, student data agreement, and FERPA/COPPA alignment overview) is being developed alongside the full account and teacher management system. These documents will be published prior to any non-pilot deployment.
Districts with specific FERPA or COPPA requirements should contact us directly before beginning a pilot. We are committed to working through district-specific requirements before any student data is transmitted.
Prior to any non-pilot deployment, Guided Scholar will publish a formal Privacy Policy, Terms of Use, Student Data Policy, and a FERPA/COPPA Alignment Overview. These documents are in development and will be shared with any school or district beginning a formal evaluation.
Schools that need these documents before beginning a pilot can request draft versions directly.
No. Guided Scholar is structured feedback and revision infrastructure, not a teaching replacement and not an autonomous grading system. Teachers remain the instructional leaders. The system surfaces information (draft history, revision behavior, rubric scores) that helps teachers make better use of limited time. It does not make instructional decisions, assign final grades, or communicate with students outside of feedback on their submissions. Teacher judgment governs how the feedback is applied and how student work is evaluated.
The teacher dashboard is designed to extend teacher visibility, not to substitute for it. A teacher who uses Guided Scholar can see what every student submitted, when they revised, and whether their revision addressed the feedback, information that would otherwise require reading every draft manually.
Guided Scholar is designed to support academic integrity rather than undermine it. The system cannot generate student work. Every word in a student's submission was written by the student. The system responds to what they submitted, it does not produce assignments, complete paragraphs, or write sentences on the student's behalf.
In school-account deployments, student practice inside Guided Scholar is not private, unmonitored AI access. It is a governed, class-linked practice activity visible to the responsible teacher and controlled by the school deployment context. Teachers can see every submission and every revision. That visibility is the accountability mechanism.
The one exception worth noting directly: each feedback response includes a single example showing the student's original sentence alongside a stronger version. This is a teaching feature, not content generation. The example is scoped to one sentence and is visible to the teacher in the dashboard. Whether students are expected to incorporate it or treat it as a model is a classroom policy decision.
Coach Me (the mode for submitting complete independent work) contains no pre-writing assistance by design. Students enter their finished draft; the system evaluates it.
Guided Scholar's general writing rubric (Organization, Clarity of Ideas, Supporting Details and Evidence, Grammar and Mechanics, and Overall Effectiveness) maps to the writing standards present in most ELA frameworks at the middle and high school level. It is not built around any single state's standards document, but the skills it evaluates are consistent with what most writing instruction targets.
For ACT preparation specifically, the system scores against the official ACT four-domain rubric. That alignment has been validated against human-scored essays and produces scores within one point of human graders across a 66-essay benchmark set.
Teachers using Guided Scholar with specific standards alignment requirements can contact us. The rubric categories and feedback prompts can be discussed in the context of your curriculum framework.
Guided Scholar is designed for grades 4 through 12. The feedback language and rubric scaffolding are calibrated for middle and high school students. The system has been developed primarily with high school contexts in mind, particularly schools where ACT performance is an instructional priority.
Elementary use at grades 4 and 5 is possible for basic essay and short response templates, but the feedback complexity is better suited to grade 6 and above. This is an area of ongoing refinement.
There is no contractual obligation beyond the pilot agreement. If the pilot does not produce evidence that warrants continued use, there is nothing to unwind. Student data is retained according to the standard retention schedule (six months active, six months archived, then permanently deleted) regardless of whether the school continues.
We would ask for honest feedback on what didn't work. That is more valuable to us than a school that continues out of inertia.
Guided Scholar is a new system. There is not yet a body of independent peer-reviewed research on its effectiveness. What exists is the ACT scoring validation: 66 essays scored by the system against human-graded results using official ACT criteria, with the system scoring within one point of the human score on all 66.
The broader evidence base for the instructional approach, including immediate feedback, revision cycles, and rubric transparency, is substantial in the writing instruction literature. Guided Scholar is built on those principles, not on claims beyond what the system can currently demonstrate.
The pilot process is designed partly to generate that evidence. Schools that participate are contributing to an honest evaluation of what the system does and doesn't improve.
The pilot debrief covers three areas. First, behavioral data from the teacher dashboard: how many students revised after receiving feedback, how substantial those revisions were, and whether revision behavior improved over the course of the pilot. Second, teacher observations: whether the system fit into instruction, whether feedback quality was useful, and whether it changed how students engaged with writing. Third, if applicable, pre- and post-practice ACT score comparisons.
We will not claim the pilot worked if the data doesn't support it. The purpose of measuring is to find out what is actually true, not to confirm a predetermined conclusion.
The strongest case is a narrow one: this is a targeted intervention in a high-stakes area, piloted in a limited classroom context, with defined evaluation criteria and no long-term obligation. It does not require replacing existing tools, changing teacher assignments, or committing budget before evidence exists.
For schools where ACT performance is an institutional priority, the connection is direct: students need more practice cycles with feedback tied to the actual scoring criteria, and the school day cannot provide enough of them. Guided Scholar adds capacity without adding staff, and does so within a teacher-visible, school-governed structure, not as unlimited AI help operating outside school oversight.
A one-page summary suitable for a board or leadership presentation is available to download. Download the one-pager (PDF) → When you are ready to start the conversation, submit a pilot request and we will follow up within a few business days.