# How to build the business case for an educational AI platform

> Learn how to present the business case for an educational AI platform to leadership, with costs, benefits, risks, indicators, and implementation models.

- Site: Genialoh (https://genialoh.org)
- Language: en
- Category: Institutional strategy
- Reading time: 20 min
- HTML version: https://genialoh.org/#/en/blog/business-case-educational-ai-platform

---
A proposal to implement artificial intelligence in a university should not begin with a list of features.

Leadership, general management, and finance need to understand what institutional problem will be solved, what it will cost, what resources it will require, what risks it will introduce, and how it will be determined whether to expand the project.

A business case turns a technological idea into an institutional decision. It should answer at least these questions:

- What problem do we want to solve?
- Why must we act now?
- What would happen if we do nothing?
- What alternatives exist?
- What difference does an institutional platform make?
- Who will use the solution?
- What investment is needed?
- What costs are included?
- What benefits do we expect?
- How will we measure those benefits?
- What risks must be controlled?
- Who will be responsible?
- What will the first phase be?
- What decision must be made afterward?

The goal is not to promise immediate transformation or to present AI as an automatic solution. The goal is to demonstrate that there is a concrete institutional opportunity, a controlled implementation model, and a reasonable way to evaluate value before scaling.

## What is a business case for an educational AI platform?

It is a document that justifies why an institution should adopt an AI platform and under what conditions. It integrates four perspectives.

| Perspective | What it analyzes |
| --- | --- |
| Strategic | Educational project, innovation, student experience, value proposition, positioning, digital transformation, competitiveness, and growth |
| Academic | Student guidance, personalized learning, practice, tutors by degree or grade, teacher work, activity design, curricular integration, and responsible adoption |
| Operational | Responsible parties, documents, users, initial programs, training, support, administration, follow-up, and schedule |
| Financial | Licenses, implementation, customization, training, support, internal costs, commercial models, potential savings, possible revenue, and financial risk |

A convincing business case integrates the four perspectives. A purely academic proposal may not answer finance's questions. A purely economic proposal may ignore pedagogical requirements.

## Step 1. Define the institutional problem

The business case should start with an observable situation. It should not start with "We want to implement AI because it's the future". That phrase expresses interest but does not define a need.

- Teachers and students use different tools without common guidelines.
- The university has no visibility into how AI is used.
- Current solutions do not reflect institutional identity.
- Teachers need training to incorporate AI in their courses.
- Students lack additional support outside class hours.
- The institution wants to contextualize tutors with its own programs.
- Leadership needs to evaluate a real experience before authorizing implementation.
- The university does not have budget to buy licenses for the whole community.
- Generic tools are not aligned with institutional methodology.
- Individual adoption is growing without a common strategy.

> **Template to write the problem**
> 
> Currently, [affected group] faces [specific situation], which produces [institutional consequence]. Current solutions do not adequately solve it because [limitation]. The institution needs to evaluate [required capability] before [decision or relevant date].

## Step 2. Explain why it must be addressed now

Urgency should not be built on fear or unprovable claims. It should relate to institutional facts.

- Students already use generative tools.
- Teachers are requesting guidelines.
- There are programs that want to incorporate use cases.
- Budget decisions are approaching.
- The university is reviewing its digital proposal.
- Other tools will be renewed soon.
- A new school year is being prepared.
- Leadership needs a proposal for the strategic plan.
- The institution wants to differentiate its offering.
- There is interest in offering an institutional alternative.

### Cost of not acting

- Fragmented adoption.
- Personal accounts without coordination.
- Duplicated subscriptions.
- Lack of training.
- Contradictory rules.
- Privacy risks.
- Confusion about academic integrity.
- Dependence on external platforms.
- No reports.
- Missed opportunity to create an own experience.

## Step 3. Define the project's objective

| Weak objective | More precise objective |
| --- | --- |
| Implement AI to transform the university. | Evaluate in a first phase whether an institutional platform can offer contextualized tutors for two degrees, support their teachers, and provide enough information to decide on broader implementation. |

> It is best to limit the first phase to one to three objectives.

## Step 4. Identify beneficiaries

| Audience | Value received |
| --- | --- |
| Students | Tutors by degree or grade, explanations at different levels, practice, guided questions, out-of-class support, institution-recognized platform, and clearer rules |
| Teachers | Support for planning, exploring explanations, preparing examples, drafting activities, practice questions, organizing materials, and interpreting adoption |
| Academic coordinators | Organize use cases, accompany teachers, review needs, identify candidate programs, and share practices |
| Academic direction | Common strategy, curricular contextualization, centralized training, information for decisions, coherence across programs |
| Leadership and general management | Institutional differentiation, financial viability, scalability, strategic alignment, risks, commercial model, expected return |
| Technology | Common platform, less fragmentation, defined responsibilities, formal process, support channels, institutional administration |

## Step 5. Describe the proposed solution

> An institutional AI experience with the university's name, logo, and colors, contextualized with its values, methodology, and academic programs, organized through tutors by degree or grade and accompanied by training, implementation, support, and analytics.

| Block | Capabilities |
| --- | --- |
| Institutional identity | Name, logo, colors, brand, interface, avatars |
| Academic context | Values, methodology, curricula, programs, offering, authorized materials |
| Student experience | Tutors by degree or grade, explanations, practice, questions, personalized learning, guidance |
| Teacher experience | Planning support, initial activity creation, examples, materials organization, exploration of explanations |
| Institutional information | Reports by student, reports by group, analytics, adoption tracking |
| Services | Diagnosis, configuration, training, implementation, support, follow-up |

## Step 6. Compare alternatives

| Alternative | Advantages | Risks |
| --- | --- | --- |
| Do nothing | No immediate investment; no internal project; no short-term changes | Disorganized adoption; no institutional experience; no visibility; policies disconnected from practice |
| Generic tools | Quick start; known products; less initial configuration; individual accounts | Provider brand; general experience; limited curricular context; scattered administration; unused licenses |
| Build from scratch | High technical control; specific functions; custom integrations | High initial investment; development time; specialists; maintenance, security, and support |
| White-label platform | Institutional identity; shorter preparation; contextualization; tutors by program; training, support, and analytics; gradual scaling | Provider dependence; customization limits; contracts and documentation; need for internal governance |

## Step 7. Define initial scope

- Two degrees.
- Two grades.
- Two areas.
- Two programs.
- Up to five institutional evaluators.
- Defined use cases.
- An evaluation period.
- An executive presentation.
- Decision criteria.

## Step 8. Explain the implementation model

| Stage | What it includes |
| --- | --- |
| 1. Diagnosis | Objectives, programs, users, use cases, responsible parties, documents, risks, indicators |
| 2. Customization | Brand, interface, avatars, values, methodology, degrees or grades, tutors, examples |
| 3. Training | Capabilities, limits, use cases, verification, academic integrity, privacy, activity design, support |
| 4. Rollout and follow-up | User onboarding, communication, launch, monitoring, support, adoption review, adjustments, evaluation |

## Step 9. Identify required resources

| Role | Main responsibility |
| --- | --- |
| Executive sponsor | Approve priority, facilitate collaboration, remove obstacles, receive results, make final decision |
| Operational lead | Coordinate meetings, documents, users, calendar, training, feedback, follow-up |
| Academic lead | Validate programs, use cases, tutors, language, level, methodology, materials |
| Technical lead | Review access, requirements, administration, security, support, integrations |
| Legal / data protection | Assess documentation, conditions, access, retention, deletion, privacy, responsibilities |
| Participating teachers | Test the experience and document usefulness, errors, use cases, and needs |

## Step 10. Calculate total cost

> **General formula**
> 
> Total cost = licenses + configuration + implementation + training + support + integrations + additional consumption + internal resources

| Item | What to ask |
| --- | --- |
| Licenses | Charged per active or registered user? Minimum? Annual? Consumption limits? Both students and teachers? |
| Customization | Included: identity, brand, avatars, programs, tutors, methodology, use cases? |
| Implementation | Includes diagnosis, configuration, testing, onboarding, launch, follow-up? |
| Training | Sessions, participants, materials, follow-up, training by discipline, new teachers? |
| Support | Channels, hours, response times, scope, period, included requests? |
| Internal resources | Time from academic direction, technology, coordinators, teachers, legal, communication, finance |

## Step 11. Analyze commercial models

| Model | Description | Includes |
| --- | --- | --- |
| Institutional licensing | The institution pays directly for the platform. Priced by size, scope, users, programs, and needs. | White-label platform, implementation, training, support, and analytics within the agreed scope. |
| Shared model (revenue share) | Students pay directly for access. No direct institutional investment. | Platform with institutional identity, implementation, training, support, and 10 % of student billing. |

> **How to present the shared model**
> 
> The platform should not be described as free. The correct framing is: the university can implement the platform without direct institutional investment. Students who contract the service pay for access and the institution receives 10 % of the corresponding billing.

## Step 12. Identify economic benefits

- Avoid a massive initial purchase before knowing adoption.
- Reduce duplicated tools by consolidating scattered subscriptions.
- Link cost with use: the shared model ties revenue to real adoption.
- Generate economic participation: 10 % in the shared model.
- Reduce separate implementation costs: training, implementation, and support included.
- Avoid building from scratch: no need to fund technology construction and maintenance.

> These benefits must be presented as opportunities, not guaranteed savings.

## Step 13. Identify academic benefits

- Tutors by program.
- Out-of-class support.
- Additional practice.
- Explanations at different levels.
- Use cases for teachers.
- Curricular integration.
- Common rules.
- Teacher training.
- Greater institutional consistency.

> **Do not promise**
> 
> Guaranteed better grades, elimination of learning gaps, perfect personalization, teacher replacement, or immediate scientific results. Analytics can show patterns but does not prove causality by itself.

## Step 14. Identify strategic benefits

- Institutional differentiation.
- Visible innovation.
- Own digital experience.
- Brand strengthening.
- Communication with families and prospects.
- Teacher preparation.
- Internal capability development.
- Positioning as an institution that adopts AI responsibly.

## Step 15. Identify operational benefits

- Common platform.
- Defined responsibilities.
- Centralized training.
- Support channels.
- Reports.
- Institutional administration.
- Process to incorporate programs.
- Less dispersion.
- Adoption tracking.
- Coordinated adjustments.

## Step 16. Define indicators

| Category | Indicators |
| --- | --- |
| Preparation | Named lead, selected programs, documents received, defined use cases, guidelines, participating teachers |
| Implementation | Configuration time, tutors ready, enabled users, training sessions, incidents, response time, adjustments |
| Adoption | Invited users, active users, frequency, use by program, participating teachers, use cases |
| Usefulness | Student and teacher perception, identified applications, valued explanations, designed activities, needs covered |
| Economic | Institutional investment, student billing, 10 % share, cost per active user, avoided costs, renewals, paid adoption |
| Decision | Interest in continuing, additional programs, requested proposal, approved budget, next-step owner, implementation date |

## Step 17. Define how return will be measured

| Dimension | How it is observed |
| --- | --- |
| Financial | Economic participation, lower initial investment, reduction of duplicated tools, lower development cost, included services |
| Academic | Use cases, trained teachers, designed activities, tutors used, perceived usefulness, guidance |
| Operational | Less fragmentation, common process, centralized support, reports, scaling capacity, implementation time |
| Strategic | Differentiation, institutional experience, digital readiness, perception of innovation, capability development |

> **Financial return formula**
> 
> ROI = (financial benefit − total investment) ÷ total investment × 100. Useful with direct investment. In a shared model, measure received economic participation, internal costs, adoption rate, net income, academic and operational value, and sustainability.

## Step 18. Analyze risks

| Risk | Possible consequence | Mitigation |
| --- | --- | --- |
| Low adoption | The platform doesn't generate enough use or value | Concrete cases, training, teacher involvement, communication, follow-up, adjustments |
| Academic | Wrong answers, unauthorized uses, or excessive dependence | Verification, tutors with rules, guidelines, activity design, teacher supervision, error reporting |
| Privacy | Sharing personal information or sensitive materials | Data minimization, authorized tools, training, limited access, retention rules, no sensitive data |
| Financial | Higher-than-expected costs or low student contracting | Gradual model, cost comparison, clear conditions, follow-up, limits, periodic review |
| Reputation | Excessive promises or unreliable experience | Responsible communication, no guaranteed results, academic validation, support, transparency |
| Operational | Lack of leads, incomplete documents, or delays | Operational lead, schedule, checklist, follow-up meetings, limited initial scope |

## Step 19. Define decision criteria

| Decision | When it applies |
| --- | --- |
| Move forward | The institution approves implementation |
| Request adjustments | The solution is relevant but needs changes |
| Expand evaluation | More information is needed before deciding |
| Postpone | Interesting but not a priority |
| Do not continue | The platform doesn't respond to current needs |

### Decision matrix

| Criterion | Suggested weight |
| --- | --- |
| Academic alignment | 25 % |
| Student and teacher experience | 15 % |
| Financial viability | 20 % |
| Implementation and support | 15 % |
| Privacy and governance | 10 % |
| Analytics and follow-up | 5 % |
| Scalability | 10 % |

## Step 20. Prepare the executive summary

| Section | Content |
| --- | --- |
| Problem | The university uses AI tools in a fragmented way and lacks a contextualized institutional experience |
| Opportunity | Create a platform with its own identity, tutors by program, teacher support, and analytics |
| Proposed solution | Evaluate Genialoh with two degrees or programs |
| Initial scope | Two programs, academic and technical leads, personalized demo, up to five evaluators, evaluation period |
| Economic model | Custom licensing or revenue share with no direct institutional investment (10 % share); implementation, training, and support included |
| Benefits | Identity, curricular context, teacher support, tutors, lower financial barrier, information to decide |
| Risks | Adoption, privacy, response quality, academic integrity, accessibility, provider dependence |
| Indicators | Users, use cases, teachers, usefulness, incidents, adoption, billing, final decision |
| Requested decision | Authorize the first meeting and preparation of a personalized demo |

## 30-day plan to build the business case

| Days | Activities |
| --- | --- |
| 1 to 5 | Diagnosis: define the problem, interview leads, review existing tools, identify current costs, document risks, select objectives |
| 6 to 10 | Alternatives: evaluate do nothing, generic tools, own development, and white-label; compare commercial models |
| 11 to 15 | Scope and costs: select two programs, identify users, calculate costs, review internal resources, analyze licensing vs. revenue share |
| 16 to 20 | Benefits and indicators: define benefits, assign indicators, identify risks, prepare mitigations, set decision criteria |
| 21 to 25 | Validation with academic direction, technology, finance, and legal; incorporate changes |
| 26 to 30 | Presentation: executive summary, deck, requested decision, owner, next step |

## Common mistakes when building the business case

- Starting with features: they don't explain what problem is being solved.
- Promising academic results: a platform doesn't guarantee better grades.
- Not presenting alternatives: the committee needs to understand why the route was chosen.
- Showing only the price: analyze total cost.
- Ignoring internal costs: coordination also requires resources.
- Not involving teachers: the proposal may lack pedagogical viability.
- Not defining indicators: without metrics there's no clear way to decide.
- Presenting revenue share as free: students do contract the service.
- Guaranteeing revenue: 10 % depends on real billing and adoption.
- Not analyzing risks: proposals showing only benefits lose credibility.
- Requesting full implementation from the start: a controlled phase reduces uncertainty.
- Not asking for a concrete decision: end with authorization, owner, and date.

## Frequently asked questions

### Who should prepare the business case?

Educational innovation, academic direction, or digital transformation can coordinate it, with participation from leadership, technology, finance, legal, and teachers.

### How long should it be?

The executive summary can be one or two pages. The full document may include financial, technical, and academic annexes.

### Should it include an ROI figure?

Only when benefits and investments can be estimated responsibly. Academic, operational, and strategic indicators can also be used.

### How is total cost calculated?

Add licenses, customization, implementation, training, support, integrations, consumption, and internal resources.

### Which Genialoh model avoids institutional investment?

The shared or revenue-share model. Students pay for access and the institution receives 10 % of billing.

### What does the model include?

The white-label platform, implementation, training, and support within agreed conditions.

### How many programs should be included initially?

Two degrees, grades, or programs allow a representative, manageable experience.

### How long does the demo take?

Genialoh can prepare a functional demo in less than 14 days after receiving complete documentation, based on the agreed scope.

### Is a budget required before the first meeting?

No. The initial meeting serves to learn about the proposal and decide whether to move forward.

### What documents are needed?

Visual identity, values, methodology, curricula, programs, representative materials, use cases, and objectives.

### Are student data needed?

Not to prepare the demo. Names, records, individual grades, or other sensitive data must not be sent.

### How is academic value demonstrated?

Through use cases, teacher and student perception, tutor relevance, activities developed, adoption, and qualitative evidence.

### Does higher use prove better outcomes?

No. Use can show adoption, but any relationship with performance must be analyzed with methodology and limitations.

## Turn the AI proposal into an institutional decision

A good business case doesn't try to convince leadership with technological enthusiasm. It presents a concrete problem, compares alternatives, acknowledges risks, and proposes a first phase that allows learning before scaling.

Genialoh allows evaluating a platform with institutional name, logo, and colors; values and methodology; context based on curricula and programs; tutors by degree, grade, or program; teacher support; personalized learning; reports by student and group; institutional analytics; training; implementation; support; and adoption tracking.

The institution can choose between custom institutional licensing or a shared model with no direct institutional investment. In the shared model, students pay for access and the institution receives 10 % of billing.

> Book a first 30-minute meeting. During the conversation you will learn the proposal, review available models, and select two programs if your institution decides to move forward.

> Once we receive complete documentation, Genialoh will prepare a functional demo with the institution's identity and academic context in less than 14 days. After the presentation, up to five authorized people can explore it for 30 calendar days.