Mandate breach
Where the portfolio has crossed a limit the advisor set — surfaces first, every time.
Investor Insights
Turn a client's ledger into the one thing an advisor needs to say next.
Every mandate breach, behavior pattern, stress read, and concentration risk in a client's portfolio, ranked into one ordered feed — each card tied to the client's own trades, not a generic score.
Insight feedranked · brand scoped
Mandate breachcritical
Riskelevated
Behaviorwatch
Concentrationreview
Model signalweak
Opportunityopen
Strengthsteady
Learnqueued
01 / The read
Eight kinds of signal. One ranked read.
Every card in the feed belongs to one of eight categories, surfaced in a fixed priority order, and every card is built from the client's own numbers — not a generic score.
Risk
What a genuinely rough month could do to this specific book.
Behavior
A repeating pattern in how this client actually trades.
Concentration
A hidden single-bet or a cluster of positions that behave like one.
Model signal
A holding the model currently rates weak.
Opportunity
Room to improve that isn't urgent enough to be a warning.
Strength
An evidence-backed habit that's actually working, not just the problems.
Learn
A short, targeted lesson tied to whatever just fired.
Ranked, not random
Mandate breaches lead every time; everything else falls in behind them by severity, then by how confident the read is.
Tied to the trade, not a guess
A concentration card names the actual pair of positions. A behavior card points to the client's own fastest-sold winner or longest-held loser — never a generic example.
Two registers, one card
Every card carries a blunt, quantified line for the advisor and a gentler one for the client — the same evidence, two ways of saying it. Roughly seven in ten signals also carry a real cost estimate; the rest are detected and cited without one.
Not a bias score. The client's own trade, the client's own number, the client's own limit.
02 / Trust
The read is only as good as what it refuses to claim.
Every mandate limit is a number the advisor set. Every behavioral signal traces to a named academic study. And nothing fires until there is enough evidence behind it.
The mandate's seven checks
Single-stock concentration
A cap on how much of the book one name can carry.
Sector concentration
The same cap, applied to a sector instead of a single name.
Maximum drawdown
A trailing comfort limit on how far the portfolio has fallen from its peak.
30-day loss review
A trigger on realized losses over the trailing month, as a share of the portfolio.
Trade-frequency ceiling
Set to the client's own stated trading style — long-term, balanced, active, or trader — not one number applied to everyone.
Turnover ceiling
How much of the portfolio has been traded, not just how many trades were placed.
Adding to losers
Flags a pattern of buying more of a position that is already underwater.
Every one of these is an explicit number the advisor sets for that client. None of them move on their own, and none are inferred from a risk-tolerance label — what counts as a breach is never ambiguous.
Twenty-one signals, each with a citation
The behavior category covers 21 named, evidence-tied patterns — not personality labels, observable trading behavior: selling winners too early while holding losers too long, trading more after early success, averaging down into a losing position, panic-selling into a loss, doubling exposure right after a big win, chasing a stock because it just hit a new high, anchoring a decision to a position's own breakeven price, buying into newly-listed stocks in their first days of trading, letting a stale rebalancing plan drift, and under-diversifying into a handful of lottery-like single-name bets, among others. Every signal that fires carries a real academic citation, shown to the advisor as a "Research-backed" badge — not a footnote, a live attribution.
Shefrin & Statman, 1985
Odean, 1998
Kahneman & Tversky, 1979
Barber & Odean, 2000
Thaler, 1985
Genesove & Mayer, 2001
Thaler & Johnson, 1990
Grinblatt & Keloharju, 2001
Ritter, 1991
Barber & Odean, 2001
Malmendier & Nagel, 2011
Barberis & Huang, 2008
Hartzmark, 2015
Built to run out of confidence before it runs out of nerve
Below the floor, nothing fires
Each signal needs a minimum amount of trading history before it can appear at all — no claim gets made off one or two trades.
Between the floor and full confidence, the claim is dampened
A real but modest pattern reads as moderate confidence, not near-certain, until the evidence is genuinely ample.
Thin overall history downgrades everything
If a client's total trade history is very short, every signal on that ledger is held to its lowest confidence tier — no exceptions.
The system is built to say "not enough evidence yet" rather than guess — and that rule is enforced in more than one place, not left to a single check that could quietly be skipped.
How the stress and correlation reads are built
Stress testing blends a library of real, calibrated past market shocks replayed against the client's actual current holdings with a forward-looking simulation for value-at-risk-style figures — and a reliability check that downgrades confidence rather than presenting a stressed number as certain. Concentration reads use a pairwise statistical correlation across the client's own holdings, surfacing the single most concentrated pair of positions and any same-sector holdings that behave like one bet despite looking diversified.
The system is designed to run out of confidence before it runs out of nerve.
03 / In practice
The feed doesn't end in a chart. It ends in the next sentence.
A single client record, read start to finish, the way an advisor actually uses it.
In practice
A client record, start to finish.
An advisor opens one client's feed and the mandate breach leads — the single-stock limit is over, ranked above everything else. Behavior comes next: a pattern of adding to a loser, tied to a citation, not a guess. A glance at the day-by-day view shows exactly which trading day drove it, and drilling in surfaces the client's own trade — the actual ticker, the actual date, the actual cost. The advisor closes the read knowing exactly what to raise, and how to say it.
Mandate-first ranking, every time
Real trades, not a generic score
Citation and cost estimate, where one applies
Client recordstep by step
1 Open the client record ranked feed loads
2 Mandate breach leads concentration, critical
3 Flagged day, calendar view real trades shown
4 Citation badge, cost estimate research-backed
The same feed also powers a gated, client-facing view in a gentler register — the mandate breach becomes a plain observation, the behavior pattern becomes a coaching moment, never the advisor's blunt framing. One engine, two voices.
The feed doesn't end in a chart. It ends in the next sentence the advisor says.
Platform fit
Nothing here runs outside the control plane.
Investor Insights is one application on FinCore's shared governance foundation — not a bolted-on analytics tool with its own rules.
Every institutional workspace on FinCore has its own operating boundary, a deliberately small three-role model, and an audit trail that treats every important change as reviewable evidence. Investor Insights inherits all of it.
Nothing here runs outside the control plane. This is what that looks like for one application.
brand-scoped, per client
One feed per uploaded client ledger, licensed and entitled per user, isolated to the institution that owns it.
Mandate history is the audit trail
A mandate change creates a new version rather than overwriting the last one — what changed and when is never lost.
Identity masked by default
Customer names and account numbers are masked in the interface everywhere profiling surfaces them — display masking, not encryption, but the default either way.
Fully bilingual, not machine-translated
The feed is rendered and cached in both English and Turkish, with the same depth in each — real trade examples, coaching moments, and cost estimates, not a thinner second-language pass.
04 / FAQ
The questions an advisor desk asks before it trusts a feed.
Short, specific answers — the same understated register as the rest of this platform.
What happens when a client's trade history is too thin to say anything useful?
Below a minimum evidence floor, a signal simply doesn't fire — no claim is made off one or two trades. Between that floor and full confidence, the claim is shown at a dampened, not-quite-certain level. On a very short overall history, every signal on that ledger is held to its lowest confidence tier.
Do you need to expose more of our clients' personal data to build these insights?
No. Investor Insights works from the account and portfolio data an institution already holds, and masks customer names and identifiers by default everywhere the feed surfaces them. Advisors see behavior and risk signal, not exposed identity, unless a workflow specifically requires it.
Is the client-facing view the same as the advisor's?
Same engine, different register. A client sees a gentler, plain-language read of the same signals — a mandate breach becomes an observation, a behavior pattern becomes a coaching moment — while the advisor sees the blunt, quantified version.
Can I export this as a branded PDF?
Not from Investor Insights directly — that's what Broker Report is for, drawing on the same governed client data. The feed is built to be read live and acted on, not archived as a static document.
Is this available in Turkish as well as English?
Yes — the feed itself is fully rendered and cached in both languages, with the same depth in each. This document's interface is translated; its longer analysis is being brought to Turkish next.
Give every advisor the read they'd otherwise have to build by hand.
Request a walkthrough focused on your mandate model, your client population, and what a first week on Investor Insights would look like.