Risk Management

We predict your clients PNL so you can move them to the optimal execution venue ahead of time.

Average Increase in Profit

By moving your accounts to externalised venues before they profit and to internal venues before losses you can increase your book revenue considerably.

Consistent Positive Results

Our models out-perform 100% externalise, 100% internalised, and the historical results of all of our clients. These results are achievable because of the consistency of machine learning decisions.

Average Decrease in Value at Risk

Not only do our models deliver increase in revenue but they also reduce the risk on the book by taking advantage of external venues at the right time. This really is the best of both worlds.

Every Account Analysed

By using machine learning to predict the PNL on every account we can ensure the book is positioned correctly every day. This means your risk team can focus where they can do the most good.

How it works.

Our machine learning models allow us to forecast client PNL every day to ensure your clients are allocated to the correct book.

Collect and Enrich Data

We pull your data from Metatrader using an anonymised manager login at the end of the trading day. We then enrich this data from account currency to USD and we calculate the features that will be fed into the models.

Predict Client PNL

We pass the enriched data into our set of machine learning models. Initially we classify the account into eight profiles. We then pass these predictions in to our revenue prediction models. We are looking to predict you clients forward 10 day PNL.

Recommend Book Changes

The 10 day PNL prediction is turned into a recomendation once we filter out the predictions that are too small to overcome the cost of hedging. These recommendations are now ready for your Trading and Operations teams to put into action.

Four easy steps.

From Demo to Go Live in as little as one week.

Demo Video Call

A short video call to walk your through the platform and to understand your pain-points and requirements.

Free Backtest

Once you provide access to your data we run a free, no-obligation back test. This usually takes us about one week to complete.

Results Presentation

We will breakdown the numbers for you on a brief call and provide you with a presentation to help with the internal business case.

Go Live

With a great backtest to set realistic expectations we have everything in place to go live with your access to the portal.

See the platform in action.

See a short example of how our algorithms classify accounts to internal and external execution venues.


We are more than happy to answer your questions on a demo call but here are a few of the frequently asked ones.

The backtest is the best way for you to understand how well our models can work on your flow. This takes the guesswork out of the upside. We predict an execution venue for every account, every day. We then compare the model performance to the actual performance and present a report to you.

We support Metatrader 4 (MT4) only at this time. We do have some Metatrader 5 (MT5) integration work underway but we do not have a release date as yet.

Our Risk Management recommendations are delivered daily. If you are looking for a more realtime monitor, check out our Position Keeper.

The models are looking for quantifiable relationships that are consistently predictable. There are some obvious ones we can discuss in detail during the demo. The interaction between margin and unrealsied, and between realised and unrealised PNL are quite common triggers.

We have several models. Some categorise the trader profile while others predict the clients profit or loss before it happens. The models consider realised and unrealised trading performance, transactional behaviour and the customer profile.

From demo to live can be as little as one week. This depends a little on how complicated your markups, group codes and commission structures are.


no obligations
    • Backtest our models
    • on your tradeflow
    • Backtest presentation
    • No obligations