Agentic information chasing
Additional detail, process and artefacts available at interview. Part of this work is heading into closed beta as document chasing in Xero Partner Hub, announced at Xerocon London 2026.
TL;DR
Led the concept, design and prototyping of an agentic experience that automatically identifies missing client documents and information, then chases clients on an accountant’s or bookkeeper's behalf. I owned the agent, its jobs, and the client communication experience end to end, working alongside another designer, senior PMs and engineering leads. The work was validated with customers, with 5 of 7 accountants and bookkeepers choosing it as their preferred feature in concept testing, and a version of it is heading into closed beta in Xero Partner Hub.
JAX answer agent concept - an AI agent for continuously chasing clients for missing information and documents needed to do billable work
The problem
Around 40-50% of ABs regularly get stuck waiting on a document or piece of information from a client, which traps them in a manual cycle of chasing. The existing thinking treated this as ad hoc requests, like chasing three missing invoices at month-end. That doesn't scale to a bookkeeper with a hundred clients. I reframed it around agent jobs: standing goals like bank reconciliation, each with its own definition of "ready". Once a job is set up, the agent monitors client data against that definition and starts the conversation itself. Smart batching merges gaps across jobs into one comms plan, so clients get one request, not several.
Designing for trust
In concept testing, 5 of 7 ABs picked document chasing as their favourite feature. It solved a pain their existing tools hadn't. But it touches the client relationship they own, and they weren't ready to hand it over untouched. Three things became non-negotiable:
Review before send. ABs wanted to see everything before it went out, in one central place. Fire-and-forget was a no.
Edit and personalise. A generic AI draft wasn't good enough. ABs wanted formal for corporate clients, warmer for others, and room for personal touches inline.
Control over what's requested. One participant didn't want a generic dollar limit sending them hundreds of documents to find the one that mattered.
I designed a graduated trust model instead of an on/off switch:
Every job starts in approval mode. The AB reviews and approves each message, with inline editing and per-client tone controls.
Messages go out under the individual AB's name, not the practice, to keep the personal relationship clients expect.
After a set number of manual approvals, the AB can unlock autonomous mode for that job.
Redundancy checks flag when a document has likely already arrived, so the agent never asks a client for the same thing twice.
Set up goals like month-end bookkeeping for the agent to continuously monitor and resolve
Initially set up with review and approval by the user to build trust before unlocking agent autonomy
What I cut
My first direction was a one-size-fits-all approach: a single threshold cutoff for all requests. It tested well as a simplicity story, then fell apart against real clients. A construction client needs every receipt for compliance, while for another, only items above a certain value matter. I pivoted to a per-client configuration, trading simplicity for something that matches how practices think about risk.
Result
In a research panel (across the US, UK and AU) customers chose document chasing over everything else when tested against other agentic and AI feature concepts. The concept met with positivity and a healthy dose of criticism.
“Everybody wants [this] now… This would be transformative”“[My client would go] Hey, your bot just keeps hitting me up…”
A version of this document chasing capability is heading into closed beta in Xero Partner Hub, where JAX automatically identifies transactions missing a receipt. The announced beta is targeting around 57,000 active partners. The work also pulled me into Xero's AI and agentic strategy group to represent the accounting and bookkeeping lens.
Learnings and open question
Reframe before you refine. The biggest shift in this project wasn't a better UI for chasing documents, it was reframing the problem from single requests to standing, monitored jobs. That systems-level reframe is what made the later design decisions (batching, autonomy) possible.
Trust is a designed mechanism, not a setting. Rather than a binary on/off for automation, giving practitioners a visible, earned path from approval to autonomy turned a source of anxiety into something they opted into.
Design work can outgrow its original brief. Bringing the accounting and bookkeeping lens to a company-wide AI strategy conversation wasn't part of the original scope, but it became one of the more consequential parts of the work.
The client experience is just as, if not more, important. We tested the AB experience, not the small business owners receiving the requests. ABs were clear it has to be just as easy for their clients, otherwise it all falls apart, so that's the most important next piece of research.