By leveraging data and giving team members more control over their work hours, Coles achieved a massive reduction in turnover. One of its HR leaders shares their approach.

As one of Australia’s largest employers, Coles has a workforce of over 120,000 team members nationwide. With such a vast operation, keeping employees engaged and motivated isn’t just a priority – it’s a business imperative. 

In FY22, we faced a major challenge: turnover surged by 50 per cent year-on-year, with 55,000 team members leaving the business. At the same time, we onboarded 45,000 new hires in our supermarkets – an effort that came with significant costs in recruitment, onboarding and training.

Faced with this growing retention challenge, we knew we needed to rethink our approach. We had limited understanding of why this high turnover was happening. We needed to work out what was going on, so we leveraged data to uncover the root issues and turn things around. 

Our data-led discovery process found that Coles had become overly reliant on recruitment. Team members were being hired and then not receiving the number of hours they wanted to work, which we discovered was contributing to high levels of attrition within the business. 

This was a turning point for us. 

We realised we needed to break our reliance on constant recruitment and focus on long-term retention instead.

To meet this challenge, we developed a Meaningful Work Strategy, which won us the Best Recruitment, Attraction and Retention Strategy Award at the AHRI Awards last year. This strategy centred on giving team members more say in how often they worked. As a result, we’ve managed to reduce turnover by 7.2 percentage points – from 43.5 per cent in FY23 to 36.3 per cent in FY24. This equates to 11,500 fewer team members leaving the business. 

We’ve also: 

• Increased the number of team members receiving the hours they want – which we call ‘desired hours’ – from 59 per cent (FY23) to 75.5 per cent (FY24) – equating to 20,000 people. 
• Enabled 3,300 team members on student visas to receive more hours during peak trade, aligning with education breaks.
• Hired 14,000 fewer team members in FY24 vs FY23 (a significant cost avoidance) because of turnover reduction and increase in desired hours.
• Increased team member engagement and positive attitudes towards flexible work. 

Leveraging data analytics tools for better insights

In FY23, a data-intensive review uncovered around 40 possible reasons for high turnover rates. We narrowed this down to two main themes: shift length and hours worked. Ultimately, people weren’t getting the shifts or the hours they wanted after they were hired. This meant people were 1.7 times more likely to leave the company. 

To address this challenge, we developed two key tools. The first, the Headcount Forecasting Tool (HCFT), determines the optimal number of team members needed in each store for a given period, along with the specific shifts required to meet demand. This allows us to make less reactive hiring decisions by encouraging store managers to slow down and look at the data, instead of simply hiring more staff. 

Working in partnership with our external partners, we rapidly built the first version of the HCFT and then trialled it with around 100 stores. Using feedback from this trial period, we refined the tool and launched it nationally in October 2023.

“Being recognised by AHRI as an industry leader is incredibly gratifying. After the immense effort that went into this project, receiving this award feels like the final ribbon tying it all together.” 

A tailored approach to working hours

After establishing the number of staff we needed in stores, we then needed to understand more about the team members we currently had – and what they wanted. We built the Desired Hours initiative to enable staff members to indicate how many hours and shifts they would ideally like. 

Team members can now easily update their desired hours through the ‘mycoles’ system on a regular basis. 

We know people might have study or life commitments they’re working around, so their availability and desired hours fluctuate, and this allows them to indicate that. 

This streamlined process replaces manual, ad-hoc systems that meant store managers were often making rostering decisions based on gut instinct rather than data. We’ve found it also encourages more one-on-one conversations between store managers and team members, to better understand individual circumstances. 

Four lessons we learned along the way

1. Data is integral 

Data has been key to this project from start to finish. In the early stages, it allowed us to run statistical models to uncover the root causes of the issue. Without that data, I don’t think we’d have been able to showcase the scale of the problem to the business. 

To make it stick, we developed metrics around how many team members are receiving their ‘desired hours’, which is included as part of our people reports to the CEO and board. This means we have to hold ourselves to account to make sure that we are reporting on how we’re tracking our turnover figures.

2. Align HR’s work to business challenges

As this initiative was designed with one of our CEO’s key business concerns in mind – rising turnover rates – we’ve had strong business sponsorship from the beginning, which has been very helpful.

To me, this is a great example of HR demonstrating their true value to a business. Rather than designing initiatives within an HR silo to serve our own agenda, we thought about what the business needed as a whole and designed a solution to match this.

3. Explain the ‘why’ to change behaviour

A key challenge was shifting people’s mindsets away from recruitment as an instinctive reaction in busy periods. For the initial roll-out period of about 12 to 18 months, we really had to hold the line and say, “This might be challenging now but – trust us – it’s here for a reason, and that’s to make sure that we’re recruiting and building an engaged team.”

Once our operational leaders understood the ‘why’ and started having conversations with their team members, they were excited about it. When they started seeing the data and started rostering differently, then we saw the fruits of our labor, and saw turnover start to improve.

4. Business and engagement metrics 

By giving employees more autonomy over their work hours, we have seen great results not only via stronger business outcomes, but also from an engagement perspective.  

Our engagement surveys show that people who are able to work their desired hours are 3 per cent more engaged than the rest of the workforce, meaning team members are happier at work and more likely to stay in their jobs for longer. 

Being recognised by AHRI as an industry leader is incredibly gratifying. After the immense effort that went into this project, receiving this award feels like the final ribbon tying it all together and is a testament to the dedication, innovation and hard work of everyone involved. 

This recognition not only validates our approach, but also reinforces the impact of our efforts in shaping best practices within the industry.

Liam Mahon is the Head of People & Culture, Supermarket Operations at Coles.

A version of this article was originally published in the April 2025 edition of HRM Magazine.

Enhance your people analytics expertise with AHRI’s People Analytics & Insights – Advanced short course, designed to elevate your workforce data management and analytical skills in a business context.