All Desks Have Retyping Inefficiencies - Trade Booking is an Expensive One.

This is a front-office automation problem: the manual re-entry that follows any negotiation in chat. Agreed trades need to be re-typed into other systems by hand: pricing blotters, booking systems, lending desk's spreadsheets. The systems may change, but the problem doesn't. It exists on every desk that negotiates in chat, all day, every day.

While inefficient capture of pre-trade information from chat platforms can lead to missed trading opportunities and poor pricing, delays and inaccuracies in booking trades have an even greater impact leading to increased market risk, out-of-date trading limits, and resultant sub-optimal use of capital. Incomplete information means post-trade compliance queries can take days to trace back to the original chat message and missed reconciliation and reporting deadlines mean big fines.

Chat platforms have been opening up for a reason

The direction of travel over the last few years has been the same across every major chat platform: more of what happens inside a chat platform conversation can now be accessed in real time by other systems (via API). But access alone isn't enough. The chat messages needs to be accurately interpretted, classified and structured. Deterministic approaches like regex may work adequately for some highly structured source message data while general-purpose LLMs are increasingly useful for summarising semi-structured chat messages. However, neither approach copes well when applied to real time processing of the shorthand desks use to confirm a trade, a vocabulary which varies widely across desks, counterparties and by asset classes. Why Trade Capture is Still a People Problem in 2026 explains why this has been such a barrier to automation and efficiency of trade booking for so long.

What changes: the conversation becomes the data

ipushpull is a data capture and workflow automation platform combining seamless integrations, powerful message interpretation and mapping, human-in-the-loop approval and auto-learning. Whatever the source of the conversation - a chat message, an email, a spreadsheet- its content is captured, interpretted, classified and turned into structured data. The platform usitlises the latest AI combined with more traditional approaches to optimise accuracy, performance and cost. The desk oversees all the processes in a configurable, interactive dashboards and can approve, enrich or correct information before anything moves downstream. A feedback loop means the platform is always learning that all-important vocabulary to automatically maintain and improve performance and accuracy. 

Trade booking: the biggest impact, and the most expensive

Trade capture and booking is where inefficiencies of manual capture, reformat and re-entry potentially costs the most and automation can have the biggest impact. Koch Energy Services, one of North America's largest physical gas traders, replaced manual booking with exactly the ipushpull approach. Accuracy of trade capture from chat is in the high 90%s, across more than 50,000 trades processed. This means re-keying and addition of futher information has been almost completely replaced by a quick approval step, so the time spent by the desk on trade booking is down by over 85%. Importantly none of this has has required the desk to change how they interact on chat. Traders keep chatting with brokers exactly as before; the retyping is what disappears.

As Koch's Tim Flynn put it: "This is about freeing traders to focus on markets, not mechanics. Less typing, fewer errors, faster outcomes."

Different Desks, Different Workflows - Same Problem 

Trade booking gets the attention because the cost of getting it wrong can be the the highest but it's by no means the only place a desk still needs to copy information from one system and format to another manually. For example:

• Investment manager's FICC desk - Traders copy portfolio manager instructions into chat, negotiate across several dealers, then key the agreed prices back into internal systems by hand. Retyping ineffiencies mean delays in PMs and business knowing their position and risk.

• Securities Lending desk - The desk fields availability and rate updates from agent lenders over email then needs to re-key all those updates into a centralised blotter by hand. 

• Broking desk - The desk sends out multiple RFQs and receives responses. Collating into a single order book is time-consuming and slow meaning trade opportunites can be missed.   

Automating capture of messages and re-entry means the desks become faster and the business becomes more efficient carries less risk. With ipushpull the technology already exists to address these problems and inefficiencies.

Get in touch and we'll show you how you could be tackling these problems now.

Related reading: