The process of putting the cart before the horse; or why to build a people process before you build technology in collections

Working for a technology company, it’s easy to think the first step to any operational process is the technology that underpins it. In my view, technology is there to support people and therefore you’re on a losing streak if you build technology ahead of knowing your process.

My approach is to build out a robust manual collections process on a whiteboard, or with sticky notes, and execute it on a small scale with email and excel to see if it can be effective, and then layering in technology.

Technology scales fast… and has ability to piss off many customers at once

We tried to launch a collections tool out of a well known ERP system where we’d framed a loose process up for the technology team and asked them to automate all reminders. For the first month, it seemed fine and then we started to get complaints. First was the customers who’d been chased before term, then there were the customers who had been sent reminders for bills they’d paid and the last straw was when the system sent a large number of customers someone else’s reminder. Apologising to 40,000 customers for one cycle is a big pill to swallow

Manual processes are easier to automate than build from scratch

In short, the tech team had not built a robust technology to support a collections process because we had yet to finalize a clear effective process. In those days, our DSO was near 60 days, and bad debt was 2%. I took the decision to kill the ERP tool and asked our collectors to work off excel and email to collect from roughly 80,000 customers.

I encouraged my team to build the process based on excel and outlook mail merges, not technology and expensive collections tools, whilst at the same time as driving operational performance. They delivered a 100% increase in leverage increasing portfolios from 500 accounts per collector to over 1,000.

Over the course of the next 18 months, my teams built a manual collections process to chase our debtors based upon a simple collections framework I developed and still use today when starting each new collections project. As our business grew over 100% in both volume and value, we created an extremely effective collections process delivering DSO below 6 days and bad debt below 0.5%, those are not metrics which happen without a great team running a truely robust process.

Continuous Improvement is driven by collectors not management

The biggest improvements in our process was delivered based on two things

  1. Collectors observations; and
  2. Data driven prioritization.

The key tool in our processes was giving every collector access to every scrap of data on their portfolio, and training them all in pareto analysis. Every collector in the organization was able to analyze their biggest issue, and drive their own performance fast. It’s easy learn from your mistakes when you have 40 individual testers working out the kinks in the process for themselves, rather than making one big change at a time globally and seeing what happens.

I also encouraged healthy competition and sharing between collectors to see who could drive the biggest improvements first for themselves, and more importantly, to share and teach their approach broadly across their team.

I’ll tackle pareto in another blog, but in summary, it’s more important for management to model an open approach to collectors running their portfolios, and sharing best practice, than to try and own each change or force automation through to teams with an unsettled process.

Technology cannot fix a bad process

I recently worked with a company with an aged debt problem. They were surprised that when they moved onto a new collections tool, the problems got worse. The leadership had not spoken to teams, or investigated the reasons why debt was ageing, but had been sold that a tool would fix their DSO and bad debt issue.

Once we cracked open the data, the problem was a bad customer setup process (an easy fix), and lax chase procedures early in the debt cycle without any attempts to restrict risk. The tool can’t tell you to fix your data acquisition issues, it just reproduces the crap you put in.

This is why it’s critically important to buildĀ  strong effective (and simple) process before bringing in the tech guys to automate and iterate.

Modern tools offer the best of both worlds

Whilst you can build super efficient and effective processes without technology, there are a number of tools on the market now which allow for both individual freedom of personal development and automation through configurable tools.

These are great in organisations with slightly more experienced collectors who are not learning their trade, but in newer teams where collectors are learning collections skills, its hard to overlay learning a specific tool as well, hence I prefer to start teams on a manual process (which is very simple) and quickly level up analysis skills. You need to either hire in experience or risk a bad customer experience.

In short, technology cannot make a process, know how your collections process should work, and use the technology to automate. The hard work upfront to understand your customers needs and your business approach through manual workflows, developed and worked by your collectors, will mean you can focus on efficiency with technology sooner and not spend all your time apologising to your customers for bugs and a bad experience.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s