Retirement Shouldn't Be a Spreadsheet: Building an AI-Powered Planning Tool for Property Investors

Picture this. A property investment advisor sits across from a couple in their early forties. The couple wants to retire at sixty with two million dollars. They own a home with a mortgage, one investment property, and a combined household income that looks decent on paper but gets eaten alive by living expenses and loan repayments.

The advisor opens a spreadsheet. Starts plugging in numbers. Changes a growth assumption. The formulas break. They rebuild the formula. Twenty minutes later, the couple is staring at a wall of cells and has no clearer idea whether their goal is realistic than when they walked in.

This is the state of retirement planning in property investment. And it's broken.

The Gap Between Ambition and Clarity

Here's what makes property-based retirement planning genuinely hard: it's not one calculation. It's dozens of interconnected ones. Mortgage amortisation schedules. Offset account interest reduction. Rental yields net of expenses. Capital growth projections across multiple properties. Australian tax implications at every layer. Superannuation comparison scenarios. And all of these need to compound over fifteen, twenty, twenty-five years into the future.

A spreadsheet can technically do this. But spreadsheets don't update in real-time when a client asks "what if I put an extra five hundred a month into the offset?" They don't generate multiple strategy paths and score each one for probability of success. And they certainly don't track whether a client is on pace three years later.

The numbers back this up. According to ASFA's 2025 Retirement Standard, Australians aged 55-59 have a median superannuation balance roughly $124,000 below what's recommended for a comfortable retirement. Nearly half of Gen-X worry about running out of money entirely (Property Update, 2025). The planning gap isn't just a mild inconvenience — it's a crisis of clarity that leads people to either overcommit to risky strategies or do nothing at all.

Designing for the Conversation, Not the Calculator

When I started building this tool — internally called Plannd — the brief wasn't "build a better spreadsheet." It was "change the conversation between advisor and client."

That distinction shaped every architectural decision. The entire experience lives on a single page. No navigation between screens, no losing context halfway through a session, no "let me go check that and get back to you." The advisor and client sit together, and everything unfolds progressively — complexity reveals itself only when it's relevant.

The system supports three distinct modes depending on the consultation context. There's the full analysis mode where existing client data pulls in automatically — financial profiles, property portfolios, mortgage details, the lot. There's a quick analysis mode for faster consultations where you input the essentials and get straight to scenarios. And then there's the discovery call mode, which is the one I'm most proud of.

Turning a Sales Call Into Structured Data

The discovery call mode is built around a conversational AI interface. An advisor on a phone call with a prospective client can type or speak naturally, and the system extracts structured financial data from the conversation in real-time.

Think about what normally happens on a discovery call. The advisor asks questions, scribbles notes on a pad, maybe types some numbers into a CRM field. After the call, someone has to compile those notes into something useful. Half the time, key details are missing.

The discovery mode flips this. As the conversation flows, the system tracks which critical data points have been captured — age, income, existing properties, mortgage details, retirement timeline — and shows completion progress in a side panel. It even suggests the next question to ask based on what's still missing. The client experiences a natural conversation. Behind the scenes, a structured financial profile is assembling itself.

By the time the call ends, the advisor has a complete enough picture to generate retirement scenarios immediately. No post-call data entry. No lost details.

The Confidence Score: Making Probability Tangible

The centrepiece of the system is the confidence score — a percentage that represents how likely a given strategy is to achieve the client's retirement goal.

When the advisor defines a retirement target (say, $2 million by age 60), the system generates three distinct scenarios: conservative, balanced, and aggressive. Each comes with its own confidence score. A conservative path that focuses on debt reduction and holding existing properties might score 72%. A balanced approach with one strategic acquisition might hit 81%. An aggressive portfolio expansion strategy might land at 65% with higher upside but more risk.

The scoring algorithm weighs multiple factors. It looks at the achievement ratio — can this strategy actually reach the target number given the timeframe? It penalises for years where cashflow goes negative. It adjusts for how aggressive the growth assumptions are. And it factors in market risk based on the number of planned property acquisitions.

What makes this powerful in a consultation is the real-time refinement. The advisor and client can adjust parameters using sliders — extra mortgage repayments, offset contributions, savings rate, property growth assumptions — and watch the confidence score update in under a second. All calculations run client-side. No server round-trips, no loading spinners. Just immediate feedback.

According to the 2025 Australian Financial Advice Landscape Report, 74% of Australian advice practices are now actively using or planning to use AI — up from just 45% the year prior. But most of that adoption is focused on admin tasks like file notes and meeting documentation. The real opportunity is in client-facing intelligence. That's exactly what the confidence score delivers — it takes a mountain of interconnected financial data and distils it into a single, intuitive number that both advisor and client can reason about together.

Progressive Strategy: Seeing the Impact of Every Decision

One of the more interesting UI patterns I built is the progressive strategy panel. Instead of presenting all strategy options at once (overwhelming), the system lets you toggle individual acceleration strategies on and off and see their cumulative impact.

Extra mortgage repayments — toggle it on, set the monthly amount, watch the confidence score shift. Offset account contributions — toggle, adjust, see the impact. Property equity extraction, portfolio cashflow redirection — each one stacks. When two or more strategies are active, the system shows the combined effect: total years saved on the mortgage, total interest avoided, the projected wealth gap between doing nothing and doing everything.

There's also a Self-Managed Super Fund comparison view that shows the leverage advantage of property investment through super versus traditional superannuation. It overlays traditional super growth against SMSF property equity accumulation — same starting balance, dramatically different trajectories when property leverage is involved. For advisors working with clients who have meaningful super balances, this visual alone changes the conversation.

Snapshot and Track: The Long Game

Most planning tools stop at the plan. Plannd doesn't.

When a plan is finalised, the system takes a complete snapshot — the client's financial position at that moment, the chosen strategy, the projected milestones, the confidence score. That snapshot becomes the baseline for ongoing progress tracking.

Months or years later, the advisor can pull up a client's plan and enter their actual current metrics. The system runs a variance analysis: are they ahead of projections, on track, building momentum, or falling behind? It uses a weighted scoring model — net worth carries the most weight, followed by portfolio value, mortgage balance, and savings.

This turns a one-time planning session into a long-term relationship anchor. The client has a reason to come back. The advisor has data to drive the next conversation. And the system does the heavy analytical lifting that would otherwise require someone manually maintaining spreadsheet models and producing comparison reports.

A stat that stuck with me: 92% of property investors in Australia never progress beyond their first or second investment property (Property Update, 2025). The reasons are complex, but one of the biggest is simply losing momentum — there's no system holding them accountable to the plan they started with. Progress tracking changes that equation.

By the Numbers

The system replaced what would have required two dedicated operational roles to deliver the same capability manually:

• Paraplanner (~$85,000 AUD/year, SEEK 2025) — building retirement scenario spreadsheets, compiling client financial snapshots, running projection models before each consultation
• Financial Planning Analyst (~$105,000 AUD/year, SEEK/Hays 2025) — maintaining ongoing progress tracking, running actual-vs-projected comparisons, updating client models as circumstances change, producing quarterly review reports

Operational savings breakdown:

• $190,000 AUD in annual savings from the two operational roles this platform replaces
• 400 dev hours avoided compared to what a traditional agency team would need to build a platform of this scope
• 70+ components across foundation, goal setting, AI analysis, interactive refinement, SMSF comparison, progress tracking, and plan persistence
• 2 FTEs replaced — ongoing, recurring savings from automated scenario generation, real-time calculations, and progress monitoring

The Takeaway

The most effective planning tools don't just produce better numbers. They change the quality of the conversation that happens around those numbers.

When an advisor can sit with a client, adjust a slider, and watch a confidence score move from 68% to 79% — both people in that room understand what just happened. The abstraction disappears. The strategy becomes tangible. And the client walks out with something they've never had before: a clear, visual, probability-weighted path to retirement that they helped build.

Retirement planning through property investment is complex. It always will be. But complexity doesn't have to mean confusion. Sometimes it just needs a better interface.