From Search to Shore: Buying a Profitable Beach Rental with AI

eLearning Portfolio Case Study

Project Overview

Course Title: From Search to Shore: Buying a Profitable Beach Rental with AI

Authoring Tool: Articulate Rise 360

Duration: 45-60 minutes

Target Audience: First-time short-term rental investors, ages 28–55

Development Time: 14 hours

Modules: 7 modules + welcome and completion

The Problem

First-time short-term rental investors consistently make three high-stakes mistakes: they choose markets based on gut feel rather than data, they fall in love with properties before validating the financials, and they underestimate the operational burden of managing a rental after closing.

The result is predictable. Investors close on properties that generate negative cash flow, spend eight or more hours per week on management tasks that could be automated, and expose their personal assets to liability by never forming an LLC. These are not failures of effort or intent. They are failures of framework.

At the same time, AI tools capable of solving each of these problems — Claude for financial modeling and legal drafting, ChatGPT for market research and listing copy, Gemini for real-time regulation lookup, Canva for branding — are freely available but widely underused by this audience. The gap is not access to tools. It is knowing how to use them.

This course was designed to close that gap — giving first-time investors a structured, AI-assisted framework for every phase of the rental journey, from market selection through monthly profitability analysis.

The Solution

From Search to Shore is a 45–60 minute self-paced microlearning course built in Articulate Rise 360. It teaches learners to use four AI tools across five phases of the rental lifecycle through a constructivist approach that prioritizes immediate application over information delivery.

The course moves learners through a deliberate arc:

Identify the right market → Evaluate the right property → Validate the financials → Close and launch → Manage and optimize

Every module was designed so that learners leave with something immediately usable — a copy-paste AI prompt, a weighted scoring matrix, a pro forma model, or a guest communication template.

My Role

I served as sole project manager, instructional designer, and developer on this project, managing the full engagement from formal training needs analysis through published course output and this complete project management documentation suite.

Responsibilities included:

• Conducting a formal training needs analysis identifying six discrete performance gaps across knowledge, skill, and process domains

• Developing a complete ADDIE instructional design document covering all five phases from analysis through evaluation planning

• Writing a full screen-by-screen storyboard across all 8 modules and 46 screens, including narration scripts, interaction specifications, and visual direction for 20 Canva graphic assets

• Engineering a detailed AI generation prompt that produced a structurally accurate Rise course in a single pass, then refining all content against the storyboard

• Building branching decision scenarios with realistic financial data and explanatory feedback on every answer path

• Designing and applying a Kirkpatrick Levels 1–4 evaluation framework with specific measurement methods and targets at each level

• Managing scope, schedule, budget, risk, and project closure using PMI PMBOK methodology throughout the development lifecycle

Instructional Design Approach

Learning theory

The course follows a constructivist approach — learners build understanding by applying AI tools to authentic real-estate scenarios rather than passively receiving information about them. Every module moves from concept introduction to guided practice to independent application, with a copy-paste AI prompt at each stage that bridges the gap between knowing and doing.

Bloom's taxonomy alignment

Content was deliberately sequenced to reach Bloom's levels 3 and 4 — application and analysis — rather than stopping at recall. Learners are not asked to remember what cash-on-cash return is. They are asked to calculate it from real inputs, interpret whether a deal meets the threshold, and decide what to do when it doesn't.

Scenario-based learning

Two branching decision scenarios place learners inside real investment moments with genuine financial stakes. In Module 1, learners choose between two real-feeling markets — one with strong nightly rates but an active STR permit freeze, one with lower rates but no regulatory risk — and must weigh data rather than instinct. In Module 3, learners receive a completed pro forma showing a slightly negative cash flow and must decide whether to buy anyway, walk away, or use Claude to calculate the offer price that makes the deal work. Both scenarios use realistic numbers, and feedback on every answer path — not just the correct one — explains the logic behind the decision.

Transfer design

The course was designed with post-course application as the primary goal, not course completion. Every module produces something the learner can use beyond the course window:

• A weighted property scoring matrix generated by Claude

• A complete rental income pro forma with three revenue scenarios

• A 12-item due diligence checklist organized by category

• A 5-message guest communication sequence template

• A downloadable AI Prompt Cheat Sheet covering all 10 core prompts across every phase of the rental journey

Technical Highlights

AI generation prompt engineering

The most significant upfront investment in this build was the Rise AI generation prompt — a detailed, structured specification that defined module sequence, Rise block types for every screen, interaction formats, question types, correct answers, feedback language, branching scenario logic, tone, color palette, and navigation settings. The prompt produced a structurally accurate course in a single generation pass, saving an estimated 15–20 hours compared to a manual screen-by-screen build. All prompt card content was then replaced manually against the storyboard, as AI-generated text paraphrased rather than reproduced the exact prompt wording — a known limitation documented in the project lessons learned.

Branching scenario architecture

Both branching scenarios required manual build and testing outside the AI generation workflow. Rise's scenario block does not reliably configure merge points through AI generation — both scenarios produced dead-end paths on the initial build that were identified during QA and rebuilt manually. Each scenario was tested on every branch path before the module was marked complete.

Legal disclaimer design

Module 4 uses Claude to generate an LLC operating agreement outline — a content type that carries real liability risk if learners treat AI output as finished legal advice. The disclaimer was designed at the storyboard level as a mandatory visual element, not added during build. On the initial Rise AI build, the disclaimer rendered as unstyled body text and was easy to overlook. It was reformatted as a prominent styled callout block and verified visible on both desktop and mobile before the module passed QA.

Accessibility Approach

From Search to Shore was designed with accessibility as a build-time consideration, not a post-publish checklist.

Key accessibility decisions:

• Alt text applied to all graphic assets, including annotated mockups and labeled graphics

• Color contrast verified across all palette combinations — teal on white and dark text on light backgrounds all meet WCAG 2.1 AA requirements

• No information conveyed by color alone — all feedback states use text labels in addition to color indicators

• Mobile layout verified in Rise's mobile preview mode across all 46 screens

• All AI output mockups labeled as illustrative with the disclaimer "Sample AI output for illustration. Your results will vary" — ensuring learners understand the content is representative, not prescriptive

• Financial benchmarks presented in green/yellow/red gauge graphics with numeric ranges labeled in text, so the data is accessible independent of color perception

Results and Reflection

What this project demonstrates

Instructional design thinking: The course reflects deliberate decisions about sequencing, cognitive load, transfer design, and assessment type at every stage. The needs analysis drove the module structure. The performance gaps drove the interaction choices. The financial analysis module is the longest in the course because the needs analysis identified it as the highest-stakes skill gap — not because it had the most content.

AI fluency as a design competency: AI tools were used at every stage of this project — Claude for financial analysis content and legal drafting, ChatGPT for listing copy, Gemini for regulation research, Canva for all graphic assets, and Rise's AI generation for the initial course scaffold. The result demonstrates that AI-assisted development is not a shortcut. It is a multiplier that requires equally rigorous design upfront and equally rigorous QA after.

Project management discipline: This project was managed using PMI PMBOK methodology from initiation through closure, with a full documentation suite including business case, project charter, WBS, schedule, budget, risk register, issue log, weekly status reports, lessons learned register, and final project report. The discipline of treating PM documentation as a first-class deliverable — not an afterthought — produced artifacts that are accurate, traceable, and genuinely useful.

Scope management under real conditions: This project was managed using PMI PMBOK methodology from initiation through closure, with a full documentation suite including business case, project charter, WBS, schedule, budget, risk register, issue log, weekly status reports, lessons learned register, and final project report. The discipline of treating PM documentation as a first-class deliverable — not an afterthought — produced artifacts that are accurate, traceable, and genuinely useful. On a solo project, that discipline is entirely self-imposed — which makes it more meaningful, not less.

What I would do differently

With additional time I would conduct a formal pilot with five to ten target learners before final publish to validate Level 1 satisfaction and Level 2 learning outcomes against the Kirkpatrick targets established in the needs analysis, and I would build a live connection between the monthly P&L prompt in Module 7 and a downloadable spreadsheet template that learners could populate with their own property data immediately after completing the course.

Skills Demonstrated

Articulate Rise 360 · Claude Sonnet 4.6 · ChatGPT · Gemini · Canva · AI prompt engineering · Rise AI generation · Branching scenario design · Constructivist learning design · Bloom's taxonomy alignment · Kirkpatrick evaluation planning · ADDIE methodology · Training needs analysis · Performance support design · WCAG 2.1 accessibility · PMI PMBOK project management · WBS and schedule development · Risk register management · Scope control · Lessons learned documentation