or How to end the Labours of Sisyphus

Note: This article is discussing elements of the European Working Time Directive, but nothing in the article shall be considered legal advice. Before making any decisions possible affecting regulations, do get your own legal advice first

Sisyphos.jpgI spent some of the more frustrating hours of my life as an HR systems consultant implementing the European Working Time Directive in SAP Time Management in some shape or from. It was frustrating so often not because the system couldn’t do it. No, the frustrating element comes from organisations, who want to implement the regulation through a system only. And you simply can’t.

Time administrators often experience the same frustration as the clean up hundreds of

warning and error messages every day only to find the same number again the very next morning. So, they could be excused to feel like ancient King Sisyphus, who was tasked by the gods (they didn’t have the European Commission yet back in those days) to roll a heavy boulder up a hill only to see it escaping and rolling back rolled back down every time he was about to reach the top.

But let’s take one step back and look at the most significant elements of the regulation.

1) Maximum daily working time is 10 hours: the possibly most important rule

Easy: cutting off working time after 10 hours even comes with the out-of-the-box configuration in the time schema TM00 and who doesn’t know constant TGMAX in table T511K? It usually also comes with a warning for the time administrator to see.

Hold on a minute. Of course, there are exceptions. On individual cases like emergencies (you don’t want your surgeon to leave you lying there cut open, because the operation took longer than planned and he reached his 10 hours for the day, right?). That’s usually dealt with using a custom subtype for infotype 2007 and a custom rule. And then there are planned exceptions, e.g., when it does make sense and is safe to compress the monthly working time into fewer days. Then you’ll manage these exception using work schedule parameters or groupings. Fine.

So, it”s quite simple: If someone clocks more than 10 hours without special permission, they don’t get paid for the extra time. Or is it? Sounds to me like there is an assumption that working time you don’t get paid for is actually not making you tired. It counts as regeneration time. Well, if you believe that, could you please come to London and work for us? We promise not to pay you, so it will be very relaxing 😉

And this is where the whole idea of “implementing the European Working Time Directive” in SAP HR Time Management, Employee Central Timesheet or any other system falls down. The system can only make sure your employees don’t get paid for more than 10 hours, but it’s got no control over actual working time (unless it’s linked to access control). I get the idea that it does discourage employees from working too long, if you tell them, they won’t be paid for it and payment is also a strong incentive to obtain approvals for exceptions, so you know about them. But it can’t be reduced to a Time&Attendance problem. It’s a People Management problem and therefore line managers need to be involved. So, the very least you need to add is that line managers get the right analytics tools to recognise excesses in working time AND someone to make sure they actually act. Whether or not employees are actually paid for the excess time becomes a side issue (though not paying them sets the right incentives and, let’s be honest, is convenient for the employer).

Another question you should ask: are we making sure employees not capturing clock times are also compliant? This group of people is often conveniently excluded from the strict regime, as there is no easy proof for them working excess hours. Well, that doesn’t make it more legal. Clearly: this is a supervisor’s task to monitor.

2) max 48h working time per week over long term average

in a 6 day week, employees can clock 60 hours without going above 10h per day. The regulator believes this is too much (probably right for most people and jobs imo), but acknowledges you might have to work a bit more during peak periods. That’s why the weekly limit of 48 hours (actually the limit is 8h per day based on a 6 day work week – and using holidays or sickness to get the average down doesn’t count) can be exceeded, but should be compensated for with shorter days during a 6 months period.

There are some customers, who want their time evaluation to throw error messages at their time admins or even cut pay. Most customers, however, don’t do anything about this rule or have a report to monitor the numbers.

The last way is by far the best imo, assuming this report is provided to and acted upon by line managers. These warning messages fired at time admins are usually ignored as a time consuming nuisance and cutting pay – well: dto.

3) Taking the right breaks

This is arguably turned into the most complicated bit in the system. The rules are (almost) clear:

– if you work more than 6h on a day you need to take 30min break

– If you work more than 9h it’s 45min all together

– you are not meant to take the break in the first or last hour or in pieces smaller than 15 minutes. Plus a few more details

Sleeping_employee.jpgThe standard time evaluation schema is set up to deal with the 30 and 45 minutes break. An even easier way is possible through work schedule rule configuration. The schema solution doesn’t really care, when the break is placed. In fact, technically, it does often place it at the end of the work day. This is against the rule, but the argument goes that it’s only about cutting 30 or 45 minutes off no matter when the employee really took her break. As overtime often goes into time, where night shift is paid, this doesn’t always work and you need to mess around a bit with the schema at that point. A discussion is: if someone clocks 6 hours and 1 minute:

– should we deduct just one minute, because than we don’t exceed the 6 hours any more.

– or 30 minutes, because that’s the break the employee should have taken already by that time, had he planned to stay for more than 6 hours

– or 15 minutes, because the first point applies, but breaks need to be taken in 15 minutes chunks?

This example illustrates how futile the exercise is: If an employee hasn’t done a break yet, when reaching 6 hours, but realises she needs a few more minutes, surely she wouldn’t have a break first before working another 5 minutes?

Well, he same as above applies: cutting breaks out of paid working time doesn’t guarantee a break, but if employees are responsible for taking their breaks and line managers do some monitoring, this may just be fine.

Where you really see a lot of futile effort is, when employees do actually clock out and in for their breaks. The standard rules (to some extend) and custom rules (often quite complex) are designed to consider breaks clocked by the employees and then adding break time as and when required. E.g.: an employee works from 8am to 18am (10 hours), clocks out 13:00 and back in 13:30 for lunch break. The schema now calculates that there are still 15 minutes of break missing. And pretends there was another break, say, 13:30-13:45. Now, tell me if I’m wrong here: I can understand that breaks are not clocked and standard breaks therefore assumed by the system unless told otherwise. However, if employees do clock breaks, then adding to the recorded break time looks suspiciously like cheating to me. And all that at the cost of considerable, unnecessary complexity added to your system. Yes, it’s not rocket science, but as it usually multiplies with a lot of other complexities, weeding these rules out would often be a real gain in simplicity (yes, consciously using the buzz word here).

4) and on and on and on

There are quite a few more rules in the regulation concerning, amongst other things:

  • a minimum of 11 hours off in each 24h period (often neglected, but an important safety issue)
  • a minimum of 4 weeks annual leave
  • special rules for employees under 18 years old

You may want to argue about the rules themselves. They do make sense, but as they are pretty indiscriminate often may take it too far and feel a bit patronising. But that”s not, what I want to discuss here.

So, can the system do this for me?

There’s one big problem in time management in many organisations and the working time directive is symptomatic for this problem:

The management of what’s often called the organisation’s most important resource is delegated to a machine, when it should be line managers’ job. In many cases, it would be possible to disentangle from several automated rules and use the energy freed up to provide line managers the right information to manage their team members.

And that’s where SuccessFactors Employee Central Time Sheet comes in, in case you have wondered about the headline. Yes: if you have extremely complex rules (be it to calculate fake breaks to comply with the working time regulation or to do more sensible things to actually support your core business), that require highly complex rules or programming of custom operations in transaction PE04, then you probably won’t be able to use Employee Central Time Sheet. In that case you need to turn to 3rd party products, when you move to the cloud.

However, if you manage to streamline your calculation (and a lot of customers I have seen could do that) to have the time and attendance system doing its own job and line managers doing theirs – supported with the right reporting, then you’ll often realise that EC Time Sheet is the better solution for you. SuccessFactors Employee Central is perfectly designed to empower employees and line managers to do their job, so you’ll be more likely to engage line managers with an easy to use system.

If you go to the very basics, Employee Central Time Sheet in release 15_08 may already be all you need. In that case: congratulations for a great job in simplifying your HR process! However, it’s quite possible you still want to use actual times rather than just durations in hours. Otherwise you won’t be able to record breaks at all. In that case you’ll have to look further down the roadmap (possibly not very far 🙂 ) and pick the right time to switch depending on the speed of evolution in EC Timesheet and your exact requirements.


Employee Central Time Sheet Screen from 15_05 release

And while you are waiting, some simplification may already be possible to prepare for the great day. Most notably: fix Sisyphus’ boulder on the top of that hill. Stop generating these hundreds of warning and error messages for time administrators, when, at the end of the day, they don’t really have the power to do anything about them. Why have them spend an hour per day ticking off and approving messages, if there is no real decision to be made. You’ll be better off monitoring it with a report and you can use the freed up time of your time admins to plan for the Employee Central migration – or for the 3rd party integration.

Comply with a single tick box

Well, it’s not all gloom and doom. There is actually one very simple way in SAP HR to comply with the working time directive. And that’s having employees opting out of it and recoding that decision in infotype 0016:


Opt-out field for UK in infoytpe 0016 in SAP HCM

To report this post you need to login first.

1 Comment

You must be Logged on to comment or reply to a post.

  1. Luke Marson

    Great blog Sven and excellent points about managing complex laws in and out of the system. I love how the EC Time Sheet is coming along and how it fits in the time and attendance landscape. More to come on that from me soon… 🙂


Leave a Reply