Frustrated by Template Portfolios: AI-Driven Services for Complex Financial Lives

Robo-advisors slot you into one of 5–10 pre-built portfolios based on a questionnaire. Our models incorporate business cash flow data, multi-account tax optimization, family dynamics, and asset-specific inputs like fleet depreciation or lease schedules. A trucking company owner whose income swings $400,000 between quarters needs a fundamentally different analytical framework than a salaried professional contributing monthly. We serve situations too complex for templates — and we've been building these custom models since 2013.

We can analyze your existing holdings, identify gaps and redundancies, and either restructure them or build a complementary allocation. We don't require you to move everything — though most clients eventually do once they see the consolidated picture. Our case study for Gill Demolition & Recycling is a good example: seven accounts across three institutions, none of them coordinated, costing $41,000 annually in overlapping fees.

Quarterly as a baseline, with regime-detection algorithms triggering interim adjustments when market conditions structurally shift. You approve every change before it's executed. Our market intelligence page explains how our regime-detection models work — identifying when markets shift from low-volatility trending to high-volatility mean-reverting states so we can act before backward-looking rebalancing rules would trigger.

A CFA or CIM charterholder reviews every AI-generated recommendation before it reaches you. For most clients, that's Naveen Kaur, CFA or Marcus Redfield, CIM — the same person you spoke with during onboarding. We don't hand you off to a junior analyst after the sale. The person who builds your model is the person who explains it to you.

That's precisely what it's designed for. The model uses time-series decomposition to separate seasonal patterns, trend components, and irregular shocks. It retrains quarterly as new data arrives, continuously improving its ability to distinguish signal from noise. For Pacific Prairie Grain Corp., Dr. Cheung's team trained the model on 11 years of actual transaction history — learning the cash flow cadence so precisely that the portfolio dynamically shifts allocation month-by-month around procurement and settlement cycles.

Ideally 3–5 years of bank statements and accounting records. We handle the digitization and feature engineering — you don't need to format anything. Tomasz Wójcik, our quantitative developer, manages the data ingestion pipeline. The more history, the better the model's ability to distinguish genuine seasonal patterns from one-off anomalies. Even irregular, messy data is useful — the model thrives on complexity.

The cash flow model feeds directly into portfolio timing decisions — when to deploy surplus capital into equities or fixed income, when to hold elevated cash reserves ahead of anticipated expenses, and when to liquidate positions ahead of predictable outflows. Your portfolio stops assuming you're a salaried employee. The practical impact: Pacific Prairie Grain Corp.'s Sharpe ratio improved from 0.42 to 0.91, and they avoided two forced liquidation events the old schedule would have triggered. Full details are on our strategies page.

We prefer it. The cash flow model generates outputs that are directly useful for tax planning — identifying optimal deployment windows that align with CCA timing, fiscal year-end planning, and estimated instalment schedules. We routinely share model outputs with clients' CPAs and coordinate on timing decisions that cross the line between investment management and tax strategy.

Not necessarily. Our consolidation engine maps holdings across institutions without requiring transfers. We recommend the optimal structure — sometimes that means fewer accounts, sometimes it means keeping accounts where they are but changing what's in them. The goal is coherence and cost reduction, not consolidation for its own sake. For Gill Demolition & Recycling, we reduced seven accounts to four — but the real win was eliminating the $41,000 in annual overlapping fees, not the account count.

Across our client base since 2018, we've identified $2.3M in total fee savings. Individual results range from $8,000 to $52,000 annually depending on account complexity and existing MER loads. The most common source of waste: holding actively managed mutual funds with MERs of 2.0–2.5% in multiple accounts when a coordinated structure using lower-cost instruments would deliver equivalent or better exposure at a fraction of the cost. Our research article on fragmented accounts breaks down the five most common hidden costs in detail.

Typically 2–3 weeks from receiving all account statements to delivering the consolidation report. Tomasz Wójcik builds a custom real-time dashboard for each consolidation client — giving you a single view across all accounts for the first time. For Harpreet Gill, that dashboard was the moment everything clicked: "I could see the overlap on one screen. It was undeniable."

It's a proprietary metric we developed to quantify how well-coordinated a multi-account portfolio actually is. It measures holding overlap, fee efficiency, tax-structure alignment, and asset allocation consistency across all accounts. A score of 100 means every account works in concert. Gill Demolition started at 31/100 and reached 87/100 within four months. Most new clients score between 25 and 45 — because nobody has ever looked at their accounts as a single system before.

It's built precisely for that scenario. Each stakeholder's risk preference is modeled as a constraint. The AI generates Pareto-optimal allocations — portfolio structures where no one family member can be made better off without making another worse off — across 50,000 simulated market scenarios. We facilitate the conversation; the math depersonalizes the disagreement. For the Fraser Valley Brewing Collective, four families reached unanimous agreement in six weeks after fourteen months of deadlock.

Facility construction timing, succession buyout structuring, education funding, joint real estate investments, pooled capital vehicles for co-operatives, or any decision where capital is shared and opinions diverge. The model is decision-agnostic — it works anywhere multiple stakeholders need to commit shared capital under uncertainty. We've used it for everything from $1.6M brewery co-op investments to $8M succession buyout structures.

Absolutely not. Priya Dhaliwal facilitates in-person sessions where families review projected outcomes on screen, adjusting preferences in real time. No equations involved. Priya is fluent in Punjabi, Hindi, and English, and has built her career on making complex AI outputs accessible to families who care about results, not methodology. If you leave a session confused, we've failed — and we want to know about it.

Usually 2–3 facilitated sessions over 4–6 weeks. Session one: we listen. Each stakeholder articulates their goals, concerns, and non-negotiables privately and together. Session two: we present initial model outputs and let the family react, adjust, and challenge. Session three: refined scenarios incorporating feedback, leading to a consensus allocation. Some families need additional sessions — the Fraser Valley Brewing Collective required three — but the structure ensures progress at every step.

Yes. Our NLP due diligence tool parses the offering memorandum, compares fee structures and waterfall provisions against a database of 2,400+ Canadian alternative offerings, and flags unusual terms, conflicts of interest, or structural risks. The AI doesn't make the decision — it ensures you understand what you're buying before you commit capital. Our research on AI in investing explains the difference between genuine analytical tools and marketing veneer.

Fee provisions buried deep in the document — management fees, performance fees, and acquisition fees that compound in ways the marketing materials don't disclose. Exit penalty clauses that restrict liquidity for years beyond what was verbally represented. Misaligned waterfall structures where the manager profits before investors reach their preferred return. Liquidity restrictions that contradict how the product was marketed. Undisclosed conflicts between the issuer and distributor. And provisions that deviate significantly from market norms across our 2,400+ offering database.

Flat fee scoped during an initial call. Typically $1,500–$4,000 depending on document complexity and the number of comparative offerings in the analysis. No asset minimum — this is a project-based engagement. One client avoided $86,000 in exit penalties from a single review. The ROI on this service is often the most immediately quantifiable of anything we offer.

Absolutely. This is one of our most popular standalone services. Many clients come to us initially for a due diligence review on a specific offering and later engage us for broader portfolio management after seeing the depth of analysis. But there's no obligation — if all you need is a single review, that's a complete engagement on its own.

Not necessarily — but it's almost certainly not optimal. A fixed cycle ignores condition variance, fuel efficiency trends, resale value windows, and interest rate environments. Two identical trucks purchased the same year may reach their optimal disposal point months or even years apart depending on route assignment, maintenance history, and accident exposure. Our model recommends unit-by-unit timing rather than fleet-wide cycles. For Sidhu & Sons Transport, the difference was $640,000 in CapEx savings over 18 months.

Maintenance records (we digitize paper shop invoices if needed), telematics data showing route-specific wear patterns, auction and resale databases (e.g., Ritchie Bros. historical results), interest rate forecasts from our market monitoring infrastructure, diesel and fuel price trend models, and Canadian tax schedules including CCA classes and immediate expensing rules. Our market intelligence on the freight and logistics sector provides additional context on EV transition timing and fleet replacement cost trends.

Sidhu & Sons Transport reduced fleet CapEx by $640,000 (23%) in the first 18 months while maintaining equivalent fleet age at 5.1 years. Resale proceeds increased 14% because units were sold closer to optimal value windows rather than on an arbitrary calendar cycle. The model has since been expanded to cover trailer replacements. For smaller operators with 10–25 units, savings typically range from $40,000 to $180,000 annually.

Yes. The same modeling framework applies to any depreciating business asset with a resale market — excavators, cranes, commercial kitchen equipment, medical devices, or commercial property. The data inputs change, but the optimization logic is the same: when does the cost of continuing to operate an asset exceed the net benefit of replacing it, given current market conditions, tax provisions, and financing costs?