The FCA’s 2026 policy statement on operational incident and third-party reporting has added to the load on smaller firms’ compliance setups[1]. When a small wealth manager looks at its anti-money laundering (AML) transaction monitoring, it usually sees two bad options: an enterprise software package costing tens of thousands of pounds a year, or manual spreadsheet checks that rely on someone remembering.

There’s a third option. Most advice firms with under two hundred staff don’t need a custom AI platform for this. They need a quiet, connected workflow that flags unusual transactions for a person to review.

Why enterprise software is rarely the right fit

Enterprise AML platforms are built for much larger institutions, so a small firm pays for scale and features it won’t use. They often carry significant licensing fees as well.

If your firm processes a few hundred transactions a week, volume isn’t your main risk. Inconsistency is. Manual checks slip when people are busy or off. An enterprise platform fixes that by giving you a huge suite of tools and asking you to rebuild your process around them.

You can get the consistency your compliance function needs more simply, by building the monitoring into the software you already use.

Building the monitoring with tools you already have

You automate transaction monitoring by connecting the systems you already pay for. I treat this as an integration job rather than a custom build: a platform such as n8n or Make links your portfolio management system to your compliance log and your internal messaging channel.

When a transaction happens, the integration pulls the data and checks it against rules you’ve written. If a client suddenly moves a large sum from an unexpected jurisdiction, the system flags it. It doesn’t decide whether the transaction is suspicious. It sends the flag to a compliance officer so nothing gets lost in a busy inbox, and that person reviews it and applies regulated judgement before anything happens.

How to start automating your AML checks

Start by mapping where the data lives, then write the rules, then build the alerts. In that order.

  1. Map your data flow. You can’t automate a process until you know where its data lives. Identify which system holds your transaction records and which holds your client risk profiles.
  2. Write fixed rules before adding any AI. Don’t ask a language model to guess whether a transaction looks suspicious. Write rules for amounts, frequencies and locations that a person can read and an auditor can check.
  3. Build the alert route. Set up an integration that posts each alert to a dedicated compliance channel in Teams or Slack, with a direct link to the transaction record so the reviewer can act straight away.

Where these projects go wrong

The tools for building these alerts are easy to get hold of. Projects stall at deployment, and usually because the build is bigger than the problem. A firm commissions a custom system when a simple webhook between existing services would have done the job in a fraction of the time.

Keeping the rules written down and the connections simple gives you a system that’s easy to maintain and easy to audit. For the wider picture on governing this kind of automation, see the UK AI governance guide.

Automating AML monitoring is about consistency. Human oversight stays exactly where it was: the system makes sure the right transactions reach the right person, and that person makes the call.

This article is for informational purposes only and does not constitute regulated financial advice or a compliance opinion. Consult a qualified compliance professional for advice specific to your firm.

Sources

[1] FCA, ‘PS26-2: Operational incident and third-party reporting’, Financial Conduct Authority, 23 May 2026. Available at: https://www.fca.org.uk/publications/policy-statements/ps26-2-operational-incident-third-party-reporting