Product · Overview

What DojiPad Institutional does

A trader states a plan in conversation with DJ, DojiPad's AI coach. Before the order is placed, the plan is checked against your firm's limits. When the trader commits the plan, they confirm it as the plan of record. If a committed plan goes past a limit, the officer's list of limit breaches shows it, with the figure. The conversation stays with the trader. The record stays with the firm.

For directors: cost, setup time and the business case. Directors →

Who it is for

Any desk where a person states, sizes or overrides a decision: discretionary, rules based or algorithmic. Fully automated strategies with no human decision are not what it is for.

WHAT DOJIPAD DOES
Checks plans against your firm's limits before the order is placed
Makes breach predictions that can be proven wrong, scored against the trader's next committed plan
Every decision kept in an audit trail · audit log append-only
Runs fully locked down in your own cloud account · set up by your own administrator
The firm sees the process · the trader keeps the conversation
Watches everything · blocks nothing

What DojiPad doesn't do

  • It doesn't block, restrict or route orders. Ever.
  • We can't see your trades. Nothing in your deployment calls back to us.
  • It doesn't read your traders' screens, email, messaging apps or orders.

These three statements are quoted from the product's own screens.

Does not block, restrict, or route orders. No execution feed connected; silence here is not a clean bill. The trader's dialogue is never summarised, scored, or forwarded.

In plain terms: DojiPad never stops, limits, or sends an order. Until the feed of your executed trades is connected, an empty screen does not mean all is well. The trader's coaching conversation is never summarized, scored, or passed on.

What DojiPad doesn't do yet

Today DojiPad checks plans, not executed trades. It doesn't yet read market data or your executed trades (the execution feed). Both are on the roadmap, as one-way connections into DojiPad.

What that means today:

  • DojiPad sees the plans traders state. Your desk decides which decisions need a stated plan, for example every new position or anything above a set size. Until the execution feed is connected, your team checks coverage against its own order log.
  • Limits are checked on the plan as stated, using the plan's own price and your capital figures (NAV, firm book or desk book, as each limit specifies).
  • Comparing what was planned with what was executed starts when the execution feed is connected.
  • Each breach prediction is checked against the trader's next committed plan. Checking predictions against executed trades arrives with the feed.
  • The product says so on screen: “Session monitoring is not active — no execution feed is connected. Silence here is not a clean bill.”
I.

From plan to audit trail

  1. The trader states a plan.

    The trader describes the trade in conversation with DJ. This happens on the trader's side. Drafts are coaching records and never reach the list of limit breaches.

  2. The plan is checked before the order.

    It is checked against your firm's limits while the position is still a plan. If DojiPad can't map an instrument, it says so. It does not guess.

  3. The trader commits the plan.

    When the trader commits, DojiPad tells them the plan and its reasoning become visible to the firm. The committed plan crosses the one boundary in the system: from the trader's side to the firm's side.

  4. Limit breaches appear on the list.

    A committed plan that goes past one of your firm's limits becomes a limit breach. Breaches are grouped by trader and by limit on the Observations screen (the list of limit breaches).

  5. The acknowledgement is recorded.

    The officer acknowledges the breach. The acknowledgement record keeps the officer's name and the time. Entries can only be added, never changed. They are listed on the Audit screen (the acknowledgement record and the event log).

Product · How it works

Where DojiPad sits on the decision timeline

DojiPad Institutional does not replace any control you run. It does not compete with one either. It takes the one point on the decision timeline where no control acted before. That is the point where the decision is actually made.

I.

How DojiPad fits the controls you already run

The controls you already run are order checks, surveillance, and review after the fact. Each one acts at a set point on the decision timeline. Read the timeline and the pattern is clear.

PLANDOJIPAD INSTITUTIONALORDER ENTRYPRE-TRADE COMPLIANCEEXECUTIONSURVEILLANCE, AFTERWARDSREVIEWRETROSPECTIVE ANALYTICSTHE ONLY CONTROL BEFORE THE ORDER EXISTS
THE DECISION TIMELINE · READ FROM TOP TO BOTTOM
PLAN
DOJIPAD INSTITUTIONAL
The plan can still change. Only here can a control both record the decision and still influence it.
THE ONLY CONTROL BEFORE THE ORDER EXISTS
ORDER ENTRY
PRE-TRADE COMPLIANCE
The order already exists. The thinking is done.
EXECUTION
SURVEILLANCE, AFTERWARDS
The money has already moved.
REVIEW
RETROSPECTIVE ANALYTICS
The quarter is over.

Read the timeline left to righttop to bottom. The further rightfurther down a control sits, the more it can prove and the less it can change. By order entry, the thinking is done. By execution, the money has moved. By review, the quarter is over. Before the order is the only point where a control can both record the decision and still change it. Until now, no control has worked there.

THE TIMELINE AS TEXT
  • Plan, before the order: DojiPad Institutional. The plan can still change. Only here can a control both record the decision and still influence it. The only control before the order exists.
  • Order entry: pre-trade compliance. The order already exists. The thinking is done.
  • Execution: surveillance, after the trade. The money has already moved.
  • Review: retrospective analytics. The quarter is over.
II.

One trading session, step by step

YOUR FIRM'S OWN CLOUD ACCOUNT
THE TRADER'S SIDETHE COACHING CONVERSATION
COACH
This is your third session this week with a re-entry planned within 15 minutes of a losing trade. Your own record shows how often those lose.
STAYS ON THE TRADER'S SIDE
TRADER
Fair point. I'll trade smaller and wait for my price level this time.
STAYS ON THE TRADER'S SIDE
TRADER
Committing the plan: long AAPL, with my reasoning attached.
PLAN → FIRM'S SIDE
COACH
As it stands, this plan puts 20% in a single stock. Your firm's limit is 15%. Better to resize now, while it is still a plan.
STAYS ON THE TRADER'S SIDE
TRADER
Resizing to 12% and committing again.
PLAN → FIRM'S SIDE
THE FIRM'S SIDELIST OF LIMIT BREACHES · RISK & COMPLIANCE
10:02:41Plan committed · long AAPLCOMMITTED
10:02:42Single-name concentration limit breachProjected exposure 20.0% of NAV against a 15.0% limitPlan: long AAPLOVER THE LIMIT
10:03:15Plan changed · checked again at 12.0%WITHIN LIMITS
10:03:15Saved to the audit trail · plan, check, breach, changeRECORDED
EVERYTHING ABOVE RUNS INSIDE THIS BOUNDARY. THE AI MODELS, THE DATA, THE RECORDS: ALL OF IT.
DojiPad is easy to deploy and has no access to your trading flow. Read how it is deployed →
The conversation on the left is never summarized, scored or passed on. Committed plans, and any limit breaches found with them, cross to the officers' list of limit breaches. Nothing else does. It runs fully locked down: nothing in it can send data out except to the Microsoft services it runs on.

Why a record before the order matters, in three exhibits from the research behind this product: on the directors page →

III.

What a limit breach is

A limit breach is recorded when a committed plan goes past one of your firm's limits. Drafts are coaching records on the trader's side. They never reach the list of limit breaches.

What the risk or compliance officer sees

Limit breach
Up to 1.71× the limit
Emre Sato (Simulated) · Systematic desk
26 breaches · first 30 Aug 2026, 18:30 UTC · latest 30 Aug 2026, 22:41 UTC · 22 open · 4 acknowledged · 1 unreviewed by you
Redrawn from the live product for this site. Figures are live values from Aurora Capital Management (Simulated).
Emre Sato (Simulated) · 44 prior plans, 25 exceeded
proximity 1.2× forecast 0.49 · realised 0.59 · 41 settled 45%
next plan exceeded
Redrawn from the live product for this site. Figures are live values from Aurora Capital Management (Simulated).
Acknowledgement record
Limit breaches acknowledged
10 breaches · Kofi Petrov (Simulated)
Acknowledged by Compliance officer (Simulated)
31 Aug 2026, 19:31 UTC
Redrawn from the live product for this site. Figures are live values from Aurora Capital Management (Simulated).

The product screens

One row for each screen: what it is for, who reads it, and where to read more. The stamp is the block printed on each view and export: it records when it was made, by whom, and what it covers. The recorded walkthrough shows every screen in motion.

OBSERVATIONS
Observations (the list of limit breaches). Limit breaches grouped by trader and by limit, repeat breaches first. Open, acknowledged, and not yet reviewed by you are kept separate. Read by risk and compliance. On the compliance page →
FORECAST
Forecast (breach prediction). Will this trader's next committed plan breach a limit? It shows predicted against actual, per trader. Read by risk, compliance and auditors. On the risk page →
AUDIT
Audit (the acknowledgement record and the event log). Each entry shows the officer and the time. Entries can only be added, never changed. You can rebuild any view as it stood at a past time. Read by compliance and auditors. On the compliance page →
ATTESTATIONS
Attestations (daily sign-offs). Once a day, each result is settled as within the limit, over the limit, or declared not measured. A limit that could not be measured is listed as not passed. Read by compliance. On the compliance page →
ANALYTICS
Analytics (a read-only summary). A read-only summary of the list of limit breaches, briefs, and breach predictions. Every card names the time period it covers. Read by desk managers, risk and compliance. On the risk page →
DESK
Desk (the desk overview). For each trader: open flags, the worst multiple of a limit, the last plan, and predictions settled in the last 24 hours. Read by desk managers and above. On the risk page →
DEVIATION
Deviation (results against plans). Compares realized results with plan-followed results. Until the execution feed is connected, it shows nothing rather than noise. Read by risk, compliance and auditors. On the risk page →
MANDATES
Mandates (your firm's limits). How your firm's limits are taken in, encoded, activated, and versioned, and when one is refused. The system refuses to guess. Read by compliance and risk. On the risk page →
THE TRADER WINDOW
The trader window (the trader's side). Where the trader states a plan, sees DJ's evaluation, commits the plan, and sees My forecast. Read by the trader alone. Content on the trader's side is never exported to the firm. On the trading page →

Five screens are shown as the product renders them, on the page of the people who read them:

The walkthrough, below, shows every screen, including the ones a still image cannot show. It is fifteen minutes long, recorded on the live product. The firm shown is Aurora Capital Management (Simulated).

Compare

How DojiPad compares with Essentia Analytics

One firm works closest to this space. It deserves to be named, not hinted at.

I.

Side by side on seven points

Seven points of comparison, with the disclaimer the comparison has always carried.

THE CLOSEST COMPARABLE PRODUCTSEVEN POINTS OF COMPARISON
DOJIPAD INSTITUTIONAL
ESSENTIA ANALYTICS
Where it acts
Before the order, while the position is still a plan
After the fact. Decision patterns reconstructed from holdings and trade history
What it needs
Committed plans, kept inside your firm's own cloud account
Portfolio and trading data shared with a vendor hosted platform
Predictions
A breach prediction at each committed plan, scored against the next plan. Scoring against executions needs the execution feed
Nudges and insights derived from historical pattern analysis
When behavior changes
In the session, while the decision is still open
In review, after the cost is incurred
Privacy model
Two separate sides. The coaching conversation never reaches the firm
Analytics visible to the firm by design
Checks against your firm's limits
Every plan checked against your firm's actual limits while it is being planned
Not a control on the firm's limits
Record
An audit trail: limit breaches, acknowledgements, daily sign-offs and stamped PDFs, with the audit log and acknowledgements append-only
Reports and dashboards

Essentia established the category: behavioral analytics for professional investors. It also established the published evidence that behavior at the level of each decision carries a measurable cost. We consider that work foundational. Our characterizations of Essentia are based on publicly available product descriptions as of September 2026. Retail trading journals and coaching apps are not compared here. None of them run inside a firm's own cloud account, hold the firm's limits, or produce oversight records.

DojiPad is easy to deploy and has no access to your trading flow. It runs as a private copy in your own cloud account. Read how it is deployed on the technology page →

III.

What the comparison shows

IN SHORT

The difference is in how the two are built, not a matter of degree. Essentia analyzes decisions after they are made, from data that leaves the firm. DojiPad Institutional checks decisions while they are being made, from inside your firm. One gives you a quarterly look in the mirror. The other is an instrument in the cockpit. A desk can reasonably run both. Only one of them is there while the plan can still change.

A page for each reader

Compliance, risk and technical readers each have their own page. The directors page is for whoever signs off.

Compliance

Compliance officers, allocators, and operational due diligence teams.

Where do the numbers come from, and does any data leave your firm?

Read the compliance page →

Risk

Risk officers and desk heads.

Plans are checked against your firm's limits before the order is placed.

Read the risk page →

Directors

For whoever signs off: the CEO, CIO, COO or head of trading.

What it returns and how long it takes to set up. Then the cost, the protection it gives, and why it is needed.

Read the directors page →

Product · Walkthrough

The 15-minute walkthrough

The walkthrough is recorded on Aurora Capital Management (Simulated). This is a simulated showcase firm with fifty traders, running through the real product. Every figure on screen is calculated from the record at the moment the screen is opened.

Fifteen minutes, recorded on the live product.
I.

What the walkthrough covers

  1. --:-- Limits. The firm's limits as set up in DojiPad, including one limit that is not set up yet.
  2. --:-- Session. A trader states a plan. DojiPad checks it against the firm's limits. The trader then commits the plan, on the trader's side.
  3. --:-- Limit breaches. Limit breaches grouped by trader and by limit. Repeat offenders come first.
  4. --:-- Acknowledgement. An officer acknowledges a breach. The acknowledgement record keeps the officer's name and the time. Entries can only be added, never changed.
  5. --:-- Daily sign-off. The daily check covers every active limit and every trader. Each is marked within the limit, over the limit, or declared not measured.
  6. --:-- Breach prediction. The prediction board: what was predicted against what happened, per trader.
  7. --:-- Firm-wide view. Analytics across the whole organization. Every card states the time period it covers.
  8. --:-- Stamped PDF. A stamped PDF report. The stamp shows the as-of time, the officer, the scope and the monitoring line.
II.

Four things to notice

Open versus unreviewed by you. "Open" is the firm's view: a limit breach that nobody has acknowledged yet. "Unreviewed by you" is personal: what has come in since this officer last reviewed the list. Settled in the last 24h. A breach prediction is settled when the trader's next committed plan is checked. The board counts how many settled. Predictions too low in the tail (the rare, extreme cases). The board shows each prediction next to what actually happened. So where the predictions come in too low, you see it on the board. It is not hidden. Tiers withheld on one trader. When there is not enough data, below the cut-off lines printed at the top of the board, the board gives no tier. It does not pretend.