How it works

No black boxes.
No surprises.

Here is exactly how Lintu builds your portfolio — from your first answer to a stress-tested result. Plain language. No jargon required.

1
Who you are
We start with you, not the market.

Most investment tools ask you one question: "How much risk can you handle?" We think that's too simple. The way you actually behave with money — under pressure, in a crash, when things are going well — matters just as much as your income or your age.

So we ask 33 questions across four areas: how long you can invest for, how you react when markets fall, how much you know about investing, and how you naturally make decisions. Takes about 8–10 minutes. No financial jargon.

At the end you get a score from 0 to 100, and a bird.

🕊️ Turtle Dove — capital preservation 🪶 Heron — steady and precise 🦉 Owl — balanced and analytical 🦅 Falcon — focused on growth 🦁 Eagle — high conviction

The bird isn't decoration. It's shorthand for a specific combination of time horizon, loss tolerance, and decision style — and it directly shapes every number in your portfolio.

2
Where the economy is
We check the economic weather before building anything.

A portfolio built for sunny skies should look different from one built when storm clouds are gathering. Before we set a single weight, Lintu reads eight live economic indicators — things like unemployment trends, bank lending, and the gap between short-term and long-term interest rates — and estimates how likely the economy is to be heading into a recession.

This isn't a guess. It's a machine-learning model trained on 65 years of US economic history. When the probability of a recession is high, your portfolio shifts a little more defensive — more bonds, fewer equities. When things look stable, it leans into growth.

The honest bit
This model cannot predict crashes. No model can. What it can do is notice when the conditions that tend to precede downturns are building — and adjust your positioning before things get bad, not after.
The exception
The 2020 COVID crash happened in two months. Too fast for any model to react. We show this explicitly in the backtest — no benefit is claimed there.
3
Your allocation
Your score and the economic signal combine into a target mix.

Your investor profile and the current economic outlook together produce a target split across four buckets: equities (shares), bonds, commodities (things like gold and oil), and — if your risk profile supports it — a small allocation to digital assets.

A Falcon with a stable economic backdrop might look like 75% equities, 20% bonds, 5% commodities. A Heron in a high-recession-risk environment might be 45% equities, 45% bonds, 10% commodities. The mix is specific to you and to right now.

There's no leverage. No short positions. Nothing exotic. The goal is a diversified portfolio you can actually hold through a bad year without panicking.

4
Choosing the ETFs
The model selects candidate ETFs. You keep full control.

Within each bucket, Lintu selects from a universe of 38 exchange-traded funds — pre-screened for size, cost (under 0.75% per year), and track record. ETFs are diversified baskets of securities, not individual stocks. One ETF might hold 500 different companies.

The weights across those ETFs are then calculated to keep your risk as low as possible for the return you're targeting. We use a method called CVaR minimisation — which in plain English means: we try to make sure the bad scenarios aren't catastrophic, not just that the average is acceptable.

EU investors
If you're based in Europe, US-listed ETFs like VOO or QQQ are not legally available to retail investors. Lintu automatically shows you the European-listed equivalents (called UCITS funds) with their full ISIN codes, so you can buy them on any European broker like Nordnet.
5
Learning from history
Your portfolio goes through five real crashes before you see it.

We apply your exact portfolio weights to five historical market crises and show you what would have happened — how far your portfolio would have fallen, how long it would have taken to recover, and how much the Lintu macro model would have helped (or not).

Dot-com crash
2000–2002. Tech stocks fell 49%. Bonds rallied. A diversified portfolio held up better than most people remember.
2008 crisis
The worst in a generation. Almost everything fell at once. Bonds were the only safe place. The average portfolio lost 35%.
COVID crash
2020. Stocks fell 34% in two months — then recovered fully in five. The fastest crash and fastest recovery in modern history.
Rate shock
2022. The US central bank raised interest rates faster than any time since the 1980s. Both stocks and bonds fell — the rare year when there was nowhere to hide.
Ukraine/Energy
2022. Russia's invasion sent energy prices soaring. Commodities surged 35%. Portfolios with a commodity allocation did significantly better.

This isn't meant to scare you. It's meant to make sure you understand what you're signing up for — before you put real money in.

6
Looking forward
1,000 possible futures. Here's the range.

Nobody knows what markets will do. Anyone who says otherwise is selling something. What we can do is run thousands of plausible futures for your portfolio — based on how markets have behaved over the past century — and show you the honest range of outcomes.

After 10 years, the best 5% of scenarios might show your portfolio tripling. The worst 5% might show it barely breaking even. The middle — the median outcome — is what you should base your expectations on.

One detail
We deliberately model more extreme events than a standard approach would — because markets really do have bad years more often than most models assume. The 2008 crash was supposed to be a once-in-a-century event. It happened twice in our lifetimes.
7
Fine-tuning
Not happy with something? You can adjust it.

Once your portfolio is built, we show you exactly which industries your money is in — technology, healthcare, financials, energy, and so on. If you work in banking and don't want more exposure to financials, or if you have a strong view on energy, you can use the sector sliders to tilt your portfolio in a different direction.

The system will tell you what would need to change in your ETF weights to get there, and whether it's actually achievable given your current allocation. No promises we can't keep.

One thing to be clear about

Lintu is an analytics tool, not a financial advisor. We are not licensed to manage your money, execute trades on your behalf, or tell you what to buy. Everything you see on Lintu is information to help you make better decisions — the decisions themselves are always yours.

We don't give personalised investment advice
We don't predict what markets will do
We don't hold your money or execute trades
We don't guarantee any specific outcome
Data sources

Where our data comes from

Every data point Lintu uses is sourced from publicly available, institutional-grade providers. No proprietary data, no black boxes. Source attribution is shown here — not on user-facing pages.

Macro economics
Federal Reserve (FRED)
500,000+ economic time series. GDP, unemployment, inflation (CPI/PCE), yield curve, credit spreads (HY/IG), and the full Model A feature set. Updated monthly.
fred.stlouisfed.org →
Futures & positioning
CFTC Commitment of Traders
Weekly futures positioning for S&P 500, 10Y Treasuries, Gold, and Crude Oil. Used in Model B to detect institutional sentiment shifts. Published every Friday.
cftc.gov →
European data
ECB / Eurostat
ECB interest rates, Eurozone GDP, unemployment, and inflation. Used in EU structural dependency model and European recession probability estimates.
data.ecb.europa.eu →
Energy & physical economy
EIA / Eurostat / FRED
Electricity consumption, oil production, natural gas flows, and freight data. Physical economy signals that lead GDP by 1–2 quarters and cannot be revised away.
eia.gov →
ETF universe
2,800+ UCITS ETFs
UCITS-compliant ETFs available to EU investors, scored on AUM, liquidity, expense ratio, and index quality. Universe updated monthly.
Digital assets
CoinGecko API
Bitcoin and Ethereum prices, market cap, and 24h volume. Used for the digital assets section. Free public API, updated continuously.
coingecko.com →
Capital flows
Bureau of Economic Analysis (BEA)
US industry-level GDP and capital flow data mapped to GICS sectors for overextension scoring. Supplemented by FRED sector series.
bea.gov →
Update frequency: Macro data updates monthly when new FRED releases arrive. Market prices update continuously. COT data updates weekly (Fridays). Models recalculate automatically when new data arrives. All data is cached locally.