Skip to content

Email triage for accounting firms

A firm’s inbox receives everything loose: client invoices to file, questions, letters from the tax authorities, platform notifications. Every misfiled document costs twice — at filing time, then at search time during the annual accounts.

In your day-to-day

A typical scenario

This typical firm’s main inbox absorbs 200 messages a day in busy season: loose invoices, client questions, platform notifications, letters from the tax authorities — all in the same stream.

  1. 01

    Through the morning, some sixty accounting documents are recognised, attached to the right client file and financial year in the document system — duplicates set aside on the way.

  2. 02

    At 11 AM, a tax reassessment proposal turns up inside an unremarkable attachment: the partner and the staff member concerned are alerted within minutes, and the statutory response window starts being counted.

  3. 03

    At 4 PM, clients missing documents for the month’s VAT return receive a reminder listing precisely what is expected — not a generic “please send us your documents”.

What changes

Filing mornings disappear, and no letter from the authorities ages in an inbox anymore: anything that starts a legal clock is seen the same day.

Order of magnitude

Working assumptions

  • 200 messages a day in busy season
  • an average of one minute’s reading and filing per message
  • 5 working days a week

In the region of 15 hours of sorting per week removed from staff mornings — close to two production days recovered weekly, at the firm’s charge-out rate. An order of magnitude you can measure on your own flow during the observation phase.

Indicative estimate built on average sector assumptions — it gets recalibrated on your actual volumes during scoping.

What eats your days

How it works

  1. 1

    Measure before sorting

    We start by measuring the real flow: who writes, about what, what actually needs action. Sorting rules are built on that data — not on impressions.

  2. 2

    Sort, extract, route

    The agent classifies each message, extracts the useful data (references, amounts, deadlines) into your tools, and routes to the right person. Urgent items are flagged within minutes.

  3. 3

    Answer the recurring

    Acknowledgements, document requests, standard questions: the agent replies on its own within the approved fence. The rest lands pre-qualified in your team’s queue.

Typical results

95%

of messages classified and routed with no human involved

4 min

between an urgent message arriving and being flagged

0

client requests lost in the pile

Orders of magnitude observed in production; your diagnostic sets your own baseline and targets.

Frequently asked questions

Can the agent file directly into our document system and practice software?+

That is the goal: it identifies the client, the nature of the document and the period, then files into your existing structure. Anything it cannot attach with certainty goes to a validation queue instead of being filed at random.

Each staff member manages their own inbox — how does this fit?+

The agent can cover the firm’s general inbox and, if you wish, assist individual inboxes with the same rules. Sorting becomes consistent — the definition of “urgent” stops depending on who happens to be reading.

Is this the problem eating your team’s time?

Tell us how you work today — 30-minute call, then a free written diagnostic of what this agent would change for you, with numbers.

Get my free diagnostic

Free resource

Get the self-assessment grid for your sector

Sales, admin, support, operations: the 20 tasks AI agents already handle in SMEs — with, for each one, the tell-tale sign that your team is concerned.