Key to his success is the focus of Team Sky on achieving the much-vaunted ‘marginal gains’ and the use of data to measure and manage performance under Team Principal Sir Dave Brailsford.

If you want a print industry equivalent of Brailsford then Anthony Thirlby is a contender. His data-driven approach at his previous business ESP garnered worldwide praise for its ability to improve production performance. He’s now chief operating officer at Pureprint, where he is charged with implementing a similar approach at a larger and more complicated operation. 

Pureprint’s raft of Heidelberg offset presses, HP Indigo digital kit and Inca Onset large-format devices at recent acquisition Inprint, alongside the Tharstern Primo MIS provide a formidable production arsenal just as impressive as Team Sky’s Pinarello bikes, Shimano components, Rapha kit and Training Peaks software.

However, for both operations it’s not all about the equipment; it’s the approach and attitude that are the most important aspects to successfully managing performance and driving ongoing improvements. Unfortunately, that can be the hardest thing to address, as it requires taking a hard look at your operations and assumptions and a lot of change.

“In my experience a lot of people don’t want to know where they are,” says Thirlby. “Most people know they need to do something but they don’t know where to start. You’ve got to be honest, then you can free up capacity or increase margins. Get beyond your ego.”

His belief that it is a cultural issue that is the biggest barrier to the adoption of data-driven performance management is echoed by Tharstern’s sales director Lee Ward: “I don’t think it’s the technology that’s the barrier anymore it’s the mind-set. I get the feeling a lot of firms are considering it, but not taking the leap of faith and doing it. I get the feeling people are worried about what it will uncover, or they are oblivious to the fact that it is achievable. 

“If you’ve got the data you can act on it. If you’re scared about what the data will show, how bad can it be?”

More data is always better

Dorchester-based books and journals specialist Henry Ling is an example of a firm that has long embraced data collection as a way to monitor and manage the business. It makes extensive use of shopfloor data collection (SFDC) feeding into its Iteba Solprint MIS. In the press hall it uses Heidelberg’s Analyze Point to provide detailed figures based on the machines’ performance rather than just operator entered information. For managing director Helen Kennett some data is good, more accurate and impartial data is better: “We use it for performance management. It means we can work with facts, not opinions. You think that you know from gut feeling what needs to be done, but the reality is often different.”

Kennett’s comment highlights a problem: whether you choose to trust the data or intuition when they don’t agree. Her experience suggests that while some scepticism is healthy it’s also important to set aside established customs, folklore and assumptions and trust what the data shows you. That, however, can take time. 

Fundamentally all the measurements you will be making come down to time and money. 

“It all comes down to cost centre recovery,” says Thirlby. “Think of each machine as a profit and loss centre and look at what it recovers every day. 

“The first step is to do the analysis: what does it cost to produce 1,000 sheets, and how long does it take to generate £1,000? What does a machine need to produce to recover its costs each day? If you have all the numbers you know how much you can produce.”

All the above are a stake in the ground so that you know where you are today. Getting all that data together is potentially scary – you may find out that you’re not as good as you thought you were. But getting away from gut feelings and having a real handle on your position allows you to aim for where you want to be.

“If you have all those figures worked out then it frees up minds to look at the next developments,” says Thirlby. “You can get away from fire-fighting to strategy.”

How can printers do that and what can businesses then do to improve productivity and profitability?

The options for collecting and analysing the data range from extremely basic manual processes through to fully automated. At its most basic, collection can be paper-based timesheets, with, in increasing levels of sophistication, detail and SFDC, direct machine interfaces through to JDF-based systems. When it comes to analysing the results there are similar levels of sophistication running from a bit of basic arithmetic and simple spreadsheets through to tools built into MIS. Some equipment vendors also offer production analysis tools as part of their press controls and front-ends. These can either be used standalone or feed the data to the MIS. So far the most notable examples are Heidelberg’s Analyze Point and HP Indigo’s SmartStream Production Analyzer.

Choosing what approach to take depends on the size and sophistication of your operation, the equipment you’ve already got installed and how involved you want to get. 

As previously stated many believe it’s a lack of knowledge and/or fear that is holding up adoption. 

“There’s a low level of awareness of the technology and what it costs and the perceived need to have fully JDF-enabled kit to be able to do anything,” says Ward. He estimates that only 5% of Tharstern customers use the built-in tools for analysis even though he also estimates that some 20% have JDF-enabled equipment that could be relatively inexpensively connected.

“The cheapest option is for firms that have equipment that is already capable,” says Ward. “If your MIS licence includes JDF then it’s a matter of three days consultancy to complete the integration. If not, then the cost of enabling the software is £1,000. Most of the latest HP Indigos are ready. It’s more costly to integrate litho presses. For instance to integrate Heidelberg Prinect Pressroom manager it’s about three or four days of consultancy. But it’s not the big project that it was in the past. It’s much more out of the box.”

If equipment isn’t JDF-enabled then it’s possible to fit sensors that can feed back to the MIS. You can retrofit sensors from £3,000 to around £10,000 to equip a press. 

“Why wouldn’t you do that for a press with five years more life – that’s £41 per week,” he says. “If it identifies that you’re running 10% slower than estimated for a number of jobs and enables you to address the pricing or the running speed it’ll pay for itself very quickly.”

Heidelberg Prinect Preprint Product Executive John Murphy says that firms using its Analyze Point software can see payback in weeks from the ability to monitor the performance of machines and the transition from one job to the next. 

“In one instance a customer found it was taking twice as long to get the paper and plates onto the press as the printing time of each job,” he says. Once that was understood the firm focused attention on making sure paper and plates were always ready, including moving the CTP next to the press to ensure no delay getting plates. 

“Often it’s about making sure that the operators have faith in their equipment,” he says. “It will very quickly identify if someone has bought a press with Intellistart, but isn’t using all the job loading and pre-setting. Say they continue to work as if they have CP2000.”

Once you have identified the problems it comes down to training to make sure staff can get the best out of the equipment. 

Both those examples show that two capacity issues that you might address through investing in additional equipment can be much more cost-effectively addressed through changing behaviour and practices, which costs significantly less money.

Thirlby’s analysis of the future for any company that doesn’t embrace a data-driven approach is stark: “Companies that aren’t run that way – as businesses – will go,” he states.

It’s not always possible to be a multi-tour winner like Froome but you could escape being swept up in the broom wagon by applying performance management to your production. 

Performance analysis: top tips

Produce reports regularly – once a week or once a month

Focus on identifying non-productive time. If you’re below 10% you’re world-class

Use 26-hour time periods so that you can see the impact of the shift changeovers

Look at makereadies – how many pulls and what are the gaps between pulls. Is that last pull really necessary?

If there are more pulls being made than normal, why? Has calibration slipped? Is the chemistry going off?

Monitor waste per job as well as overall to get a picture of emerging trends. High waste might be another sign that the calibration or chemistry are drifting off. 

Benchmark against yourself rather than other companies – try and improve your own performance not relative to others.

If a press isn’t running at maximum speed there is often a good reason. Look at the intent. Has the operator tried running it up and then backed off or did they just go easy?