| Tool assessed | www.bicon.co · about.bicon.co |
| Assessment date | 11 June 2026 |
| Method | Live browser analysis + JavaScript prompt extraction |
| Assessed by | Claude (Anthropic) |
| Scope | Features, competitive benchmarking, AI capabilities |
bicon.co is a free, account-free, browser-based ETF portfolio construction and analysis lab built specifically for European/Swiss investors. It occupies a unique niche: the only tool in the market that combines rule-based portfolio construction (from investor profile → asset allocation → named UCITS ETF implementation) with deep quantitative analytics (Monte Carlo, CVaR, stress tests, look-through decomposition) in a single, fully transparent, offline-capable interface.
Its closest analogues — Portfolio Visualizer, justETF, and Morningstar — each cover one dimension well but none combine all three layers (construction + ETF selection + analytics) for the UCITS/European context.
The tool also includes a sophisticated AI bridging layer: a structured prompt export system that connects the deterministic Lab engine to any external LLM, enabling a round-trip workflow where an AI's portfolio recommendation can be imported back for full quantitative benchmarking.
| Dimension | Detail |
|---|---|
| Stack | 100% frontend (no server calls, offline-capable) |
| Data refresh | Nightly for ETF listings/TERs; CMAs are static |
| ETF data source | justETF (look-through profiles, ISIN metadata) |
| Storage | Browser localStorage (no account required) |
| Languages | English / German |
| Export | PDF (one-page), PDF (detailed), PPTX, Excel, JSON file |
| Positioning | "A BICon showcase" — a technology demonstration |
This is the core engine. It takes a structured investor profile and outputs a complete, named ETF portfolio with full analytics.
| Parameter | Options |
|---|---|
| Base Currency | CHF · USD · EUR · GBP |
| Investment Horizon | Any integer (years) |
| Risk Appetite | Low · Moderate · High · Very High |
| Target Equity Allocation | Slider 0–100% |
| Number of ETFs | Auto or manual (tested up to 8) |
| Preferred Exchange | SIX Zürich · XETRA · LSE · None (all European) |
| Thematic Tilt | None · Technology · Healthcare · Sustainability · Cybersecurity |
| Currency Hedging | Full FX hedge toggle · Bonds-only hedge toggle |
| Synthetic ETFs | Toggle — swap-based replication eliminating 15% US dividend WHT (~20–30 bps/yr) |
| Look-Through Analysis | Toggle — routes charts/metrics via actual ETF holdings |
| Satellites | Commodities (Gold) · Listed Real Estate (REITs) · Crypto |
Every allocation row resolves to a named UCITS ETF with: Asset Class · Weight (editable) · ETF Name (dropdown with alternatives) · ISIN · Ticker(Exchange) · TER · Domicile · Replication method · Distributing/Accumulating · Currency · Qualitative comment. Users can override any ETF from a dropdown of alternatives and reset to defaults. A snapshot freshness indicator shows the last nightly refresh time.
Expected Return (p.a.), Volatility (stdev), Sharpe Ratio (with per-currency risk-free rate), Max Drawdown (heuristic), Beta vs. ACWI, Alpha vs. ACWI, Tracking Error, Outperformance vs. ACWI.
Each scenario shows portfolio loss with a "Top Drivers" breakdown.
User sets a target loss threshold; the engine solves for the scenario multiplier needed to hit it for each historical scenario, plus an equity-only uniform shock. A unique feature not found in comparable tools.
Side-by-side comparison of two independently configured portfolios with full structural and quantitative diffing. Portfolio A is linked to the Build tab; Portfolio B is configured independently. Both can be loaded from the Explain tab or from saved files.
"Bring your own ETFs" — the reverse workflow. Users enter an existing real portfolio and receive the same analytics suite as Build.
Analysis output is identical to the Build tab. "Send to Compare" routes the portfolio to Slot A or B on the Compare tab.
Full transparency documentation covering every assumption, formula, and data source.
Live-editable parameters (stored locally, override app-wide defaults):
An Excel snapshot download provides a full model with live Excel formulas for independent verification — including allocation, CMAs, correlations, covariance matrix, and all 8 risk metrics.
| Feature | Significance |
|---|---|
| Synthetic ETF toggle with WHT quantification | Explicitly models the ~20–30 bps/yr benefit of swap-based replication vs. 15% US dividend withholding tax — no other public tool does this |
| Reverse Stress Test | Inverts the stress test: "how much worse does 2008 need to be to break your plan?" Novel framing for downside anchoring |
| CVaR(95) and CVaR(99) reporting | CFA/Solvency II-standard tail-loss metrics in a free consumer tool |
| Student-t distribution option | Explicitly enables fat-tail Monte Carlo, vs. the Gaussian default in most tools |
| Crisis vs. Normal correlation regimes | Stress-correlation Monte Carlo is typically an institutional feature |
| Home Bias Analysis with cap-weight multiplier | Explicit assessment of Switzerland/home-currency overweight — unique to the Swiss market context |
| Look-through decomposition across custom ETF mixes | Goes beyond the ETF itself to surface concentration in underlying holdings across the entire portfolio |
| AI Prompt export (Basic & Strict modes) | Bridges rule-based output to conversational LLM analysis — users can send the portfolio to ChatGPT/Claude for narrative interpretation |
| Full transparency + Excel download | Downloadable model with live formulas for audit — virtually unheard of in this category |
| PPTX export | Professional-grade client presentation export from a free web tool |
| No account required | Zero friction; no data collection |
| Dimension | bicon.co | Portfolio Visualizer | justETF | Morningstar Investor | iShares Core Builder | ETFdb.com |
|---|---|---|---|---|---|---|
| Primary focus | Construction + analysis | Backtesting + optimization | Tracking + screener | Research + monitoring | Model portfolios | Screener + data |
| ETF universe | UCITS (European) | US (ETFs/funds/stocks) | UCITS (European) | Global | iShares only | US |
| Construction engine | ✓ Full (profile→alloc→ETF) | ~ Asset-class only | ~ Savings plan backtest | ✗ None | ~ Model portfolios | ~ Model portfolios only |
| Named ETF recommendations | ✓ With ISIN/Ticker | ✗ Asset class only | ✗ User must pick | ✗ N/A | ~ iShares only | ~ Generic models |
| Exchange support | SIX · XETRA · LSE | US exchanges | European | Global | US | US |
| Base currencies | CHF · EUR · USD · GBP | USD | EUR (paid) | USD | USD | USD |
| Historical backtesting | ✗ | ✓ Deep (since 1972) | ✓ (since 2005) | ✓ | ✗ | ~ Limited |
| Monte Carlo simulation | ✓ 2k paths, fat-tail, crisis regime | ✓ | ✗ | ✗ | ✗ | ✗ |
| CVaR / tail-loss metrics | ✓ CVaR(95) + CVaR(99) | ~ Percentile only | ✗ | ✗ | ✗ | ✗ |
| Stress testing | ✓ 3 scenarios + reverse stress | ✓ | ~ Drawdown chart | ✗ | ✗ | ✗ |
| Look-through decomposition | ✓ Geo + sector + Top 10 | ✗ | ~ (paid) | ✓ X-Ray | ~ Limited | ~ Holdings only |
| Home bias analysis | ✓ Explicit × cap-weight | ✗ | ✗ | ✗ | ✗ | ✗ |
| Currency hedging modelling | ✓ Granular (full/bonds-only) | ✗ | ✗ | ✗ | ✗ | ✗ |
| WHT / synthetic ETF logic | ✓ ~20–30 bps quantified | ✗ | ✗ | ✗ | ✗ | ✗ |
| Side-by-side comparison | ✓ Full structural delta + analytics | ✓ (up to 3 portfolios) | ~ (paid) | ~ | ✗ | ✓ (ETF-level) |
| "Explain my portfolio" | ✓ | ✓ | ✓ (tracking) | ✓ | ✗ | ~ |
| Portfolio coherence checks | ✓ | ✗ | ✗ | ✗ | ✗ | ✗ |
| Fee estimator (horizon-adjusted) | ✓ Bucket-level TER drag | ~ Expense ratio only | ✓ | ✓ | ✓ | ✓ |
| Factor / style analysis | ✗ | ✓ Fama-French, momentum | ✗ | ✓ Style box | ✗ | ✗ |
| Efficient frontier / optimizer | ✗ | ✓ | ✗ | ✗ | ✗ | ✗ |
| Export (PDF/PPTX/Excel) | ✓ All three | ~ Charts/CSV | ~ Excel (paid) | ✗ | ~ CSV | |
| Full methodology transparency | ✓ + Excel download | ~ Partial | ✗ | ✗ | ✗ | ✗ |
| Account required / cost | ✓ Free, no login | ~ Free tier limited | ~ Paid for deep features | $249/year | ✓ Free | ~ Partial paywall |
| Mobile app | ✗ | ✗ | ✓ iOS/Android | ✓ | ✗ | ✗ |
| Broker integration / live sync | ✗ | ✗ | ✓ (partner brokers) | ~ | ✗ | ✗ |
| Transaction tracking | ✗ | ✗ | ✓ | ✓ | ✗ | ✗ |
✓ Yes/Full ~ Partial/Limited/Paid ✗ Not available
Portfolio Visualizer is the gold standard for US-based backtesting and quantitative analysis. It offers historical simulation back to 1972, Fama-French factor regressions, and mean-variance optimization — capabilities bicon.co does not have. However, it is entirely US-centric (no UCITS ETFs, no European exchanges), has no portfolio construction engine, and is notoriously complex for non-specialists. bicon.co outperforms it on construction guidance, UCITS coverage, currency hedging logic, WHT modeling, and transparency. The two tools are largely complementary rather than competing for the same user.
justETF shares the same European/UCITS universe and provides the underlying look-through data that bicon.co consumes. justETF excels at portfolio tracking, historical backtesting, ETF discovery/screening, mobile, and broker integration. However, it has no portfolio construction engine, no forward-looking Monte Carlo, no stress testing, no CVaR, no currency hedging modeling, and deep analytics sit behind a paid subscription. bicon.co is effectively a more analytical, forward-looking complement to justETF's tracking and screening strengths.
Morningstar's X-Ray and Portfolio Manager are best-in-class for fund research, style box analysis, and mutual fund composition. However it has no construction workflow, no Monte Carlo, no stress testing, no UCITS-specific features, and costs $249/year. For European ETF portfolio construction, it is largely irrelevant.
iShares Core Builder is entirely restricted to the iShares fund family and US markets, with very limited analytics. It is a marketing and sales tool rather than an independent analytical platform.
Primarily US-focused screener and data platforms. They offer portfolio analyzers for expense ratios and holdings exposure, but no construction engine, no simulation, and no UCITS coverage.
| User Type | Fit | Notes |
|---|---|---|
| Swiss/European retail investor (self-directed) | ★★★★★ | Ideal primary tool |
| European wealth manager / IFA | ★★★★★ | Strong for client presentation (PPTX/PDF); limited by no live data or broker integration |
| Finance student / educator | ★★★★★ | Methodology transparency and Excel download make it outstanding for learning |
| US-based investor | ★★★★★ | Wrong ETF universe; use Portfolio Visualizer instead |
| Institutional portfolio manager | ★★★★★ | Too simplified; missing factor analysis, optimization, live data |
| Journalist / analyst researching ETF portfolios | ★★★★★ | Quick structured scenarios, excellent PDF/PPTX exports |
bicon.co is a remarkably comprehensive free tool for its target audience. No other publicly available product combines rule-based UCITS ETF portfolio construction, deep quantitative simulation (with fat-tail distributions and crisis correlation regimes), look-through decomposition, currency hedging modeling, WHT optimization, and professional-grade export in a single zero-friction interface.
Its main limitations — no historical backtesting, no live data, no optimization engine — are acknowledged honestly in the Methodology tab and are natural trade-offs of the frontend-only, transparency-first design philosophy. For backtesting, Portfolio Visualizer is the complement; for real-time tracking and broker sync, justETF fills the gap.
For a European investor building or reviewing an ETF portfolio from scratch, bicon.co currently has no single direct competitor.
bicon.co's positioning is explicitly "Deterministic finance meets AI intelligence" (about.bicon.co tagline). The AI layer sits entirely outside the tool's calculation engine — the Lab itself is rule-based and deterministic — but it provides a structured bridge to any external LLM. This is an architecturally deliberate choice: the Lab generates facts, the AI generates an independent opinion, and the Compare tab puts both side-by-side.
| Feature | Location | What it does |
|---|---|---|
| Show & Copy AI Prompt | Build tab — left panel | Generates a structured CFA-level prompt pre-filled with the user's constraints; copies to clipboard for pasting into any LLM |
| Import via copy / paste | Explain tab | Accepts the AI's response (ISIN + weight lines) back into the Lab for full analysis |
| Professional dialogue | Top-right, persistent | mailto:info@bicon.li — human-level contact with the BICon team |
The modal shows both modes simultaneously with tab switching so users can compare before choosing.
| Dimension | Basic & Fast | Pro & Strict |
|---|---|---|
| Prompt length | ~6,700 chars | ~8,750 chars |
| AI persona | CFA-level portfolio strategist | CFA-level portfolio strategist |
| Execution style | "Focus on speed and clarity. Pragmatic, heuristic approach." | "MANDATORY reasoning discipline. Follow steps in order. Justify every allocation decision." |
| Construction methodology | Sound portfolio design principles. Heuristic diversification. | Mean-variance / efficient frontier: base allocations on return, volatility, and correlations. |
| Internal validation | Not required | MANDATORY self-check before final answer |
| Output depth | Table 1 (allocation) + Table 2 (ETF implementation) + import block | All of Basic + currency overview + top-10 look-through + rebalancing concept + cost estimate + efficient frontier rationale |
| ETF validation | Critical requirement, mandatory | Critical requirement, mandatory (same) |
The prompt is generated fresh from the user's current Lab settings. Every constraint configured in the Build tab is injected:
- Base currency: CHF ← from currency dropdown - Risk appetite: High ← from risk selector - Investment horizon: 10 years ← from horizon input - Equity allocation between 50% and 70% ← derived from risk (High → ±10% band) - Target range of 8 positions in total ← from number of ETFs - Prefer ETFs tradable on SIX Swiss Exchange ← from preferred exchange - Satellites: none requested ← from satellite toggles - No broad currency hedging on equity ← from hedging toggle state - Use physical replication only ← from synthetic ETF toggle
Changing any parameter immediately generates a different, tailored prompt — the LLM receives a precisely scoped brief rather than a generic question.
Both prompts share the same skeleton; Pro mode adds mandatory validation layers:
| Section | Content | Applies to |
|---|---|---|
| 1 — Role | "You act as an independent CFA-level portfolio strategist." Single, clean persona instruction. | Both |
| 2 — Objective | User-specific constraints injected (currency, risk, horizon, equity range). Concrete, machine-readable brief. | Both |
| 3 — Execution mode | Basic: speed and pragmatism. Pro: structured reasoning discipline with explicit justifications. | Both |
| 4 — Construction methodology | Basic: heuristic diversification. Pro: mean-variance framing, approximate efficient frontier. | Both |
| 5 — Internal validation | 4 mandatory self-checks: table consistency, no redundant exposures, min position sizes, simplification test. | Pro only |
| 6 — Eligible asset classes | Derived from satellite toggle settings. | Both |
| 7 — Requirements & constraints | 11–13 numbered items covering exchange, hedging, replication, home bias, diversification. | Both |
| 8 — ETF validation (MANDATORY) | Field-by-field verification: ISIN↔name, Ticker↔exchange, asset class, replication, UCITS status. Cites SIX, justETF, Deutsche Boerse as sources. | Both |
| 9 — Final consistency checks | All ISINs unique; all ETFs active and tradable; exchange preference respected. | Both |
| 10 — Output format | Precisely specified columnar output mirroring the Lab's own ETF Implementation table. | Both |
Anti-hallucination guardrail (Section 8):
"Never rely purely on model memory for ISINs, tickers, exchange listings, ETF names, share classes. If an identifier cannot be verified with high confidence: explicitly state uncertainty, do not guess, and propose a verified alternative."
The full workflow — as documented on about.bicon.co and confirmed via the import dialog:
1. BUILD → Lab generates a rule-based reference portfolio 2. PROMPT → One click generates a CFA-level brief (Basic or Pro) 3. COPY → User pastes prompt into ChatGPT, Claude, Gemini, or any LLM 4. AI → LLM returns an independent portfolio + import block 5. PASTE → User pastes import block into Explain tab 6. IMPORT → Lab parses ISIN/weight pairs, maps to catalog, runs full analysis 7. COMPARE → Compare tab shows Lab portfolio (A) vs AI portfolio (B) across all metrics
IE00B4L5Y983 / 35 IE00B5BMR087 / 25 IE00B3F81409 / 40
| Parsing rule | Detail |
|---|---|
| Separators accepted | /, tab, ;, or , |
| Header rows | Auto-detected and skipped (e.g. "ISIN\tWeight") |
| Decimal separator | Comma or dot both accepted |
| Comments | Lines starting with # skipped |
| Parser design | Tolerant/forgiving — accepts raw LLM output with minimal cleanup |
This round-trip is the product's key workflow innovation. It turns the Lab into a benchmarking and verification layer for LLM-generated portfolio advice, rather than treating AI as a replacement for structured analysis.
The "Professional dialogue" button (persistent, top-right) is a mailto:info@bicon.li link. From about.bicon.co:
"The Investment Decision Lab is a non-commercial technology showcase. If you want to discuss the methodology, explore the analytical capabilities, or have a professional conversation about portfolio construction, the BICon team is open to dialogue."
This is not an AI chatbot — it is a direct human-to-human channel for methodology discussions, capability exploration, and business development conversations. It serves as the contact mechanism for what is fundamentally a technology showcase, not a consumer product.
No comparable free ETF portfolio tool offers this workflow:
The bicon.co approach is architecturally distinct: rather than replacing the structured tool with AI, it uses the structured tool to frame, brief, and benchmark the AI. This is a more defensible and transparent use of LLMs in a regulated financial context — the deterministic tool provides ground truth for comparison, and the AI's role is clearly labelled as an independent second opinion.
The prompt engineering in bicon.co is genuinely sophisticated for a free tool. The dynamic constraint injection, mandatory ETF validation guardrails, two-mode design, and structured import round-trip represent a well-thought-out human-AI workflow.
The key insight — that a deterministic tool and a probabilistic LLM serve complementary roles and should be compared rather than conflated — is both intellectually honest and practically useful.
The main limitation is the absence of an in-app AI interface, which creates friction and leaves output quality uncontrolled. But given the tool's non-commercial, frontend-only, no-API architecture, this is a deliberate trade-off rather than an oversight.
Sources: Live browser analysis of www.bicon.co and about.bicon.co (11 June 2026) · Prompt text extracted via JavaScript · Import dialog inspected live · Competitor feature research via web search