8 October 2014 707 words, 3 min. read

Quality improvement through reengineering of business processes

By Pierre-Nicolas Schwab PhD in marketing, director of IntoTheMinds
We are currently conducting a big assignment for a BEL20 company to improve financial provisions processes and increase operations quality. This assignment encompasses the description of AS-IS processes across several divisions, reengineering them into something “lean” and improving the quality […]

We are currently conducting a big assignment for a BEL20 company to improve financial provisions processes and increase operations quality.

This assignment encompasses the description of AS-IS processes across several divisions, reengineering them into something “lean” and improving the quality of the output. A big challenge given the complexity of those financial processes and the variety of people involved. Most assignments of this type end up on recommendations that are never implemented. But what’s so difficult about it ?

One big system to collect the data

When talking about collecting, organizing and retrieving data, you’ll undoubtedly have to deal with databases, ERP’s and the like. For financial data, Oracle and SAP are the two famous systems you’re likely to be faced with. For operational data, each industry has its own.

Central databases are key for the provisions processes : you need one financial database to have all your financial transactions, and one operational database to perform a comparison and calculate your provisions.

This being said, I know many assignments that ended up in nothing tangible because the central database was seen as the solution for everything. It’s not. A database remains a archiving system that needs to receive data from somewhere. This data needs to be input by someone. And the heart of the problem is there.

Don’t underestimate processes

Whatever the IT system, it needs to be fed with something. Everyone knows the saying “Garbage in, garbage out”. I like this saying because in 100% of the process reengineering exercises we’ve done in the past, something was wrong with the data entered in the system.

What you must do to ensure that your central system works well (I’m sorry to write such obvious things but sometimes it’s really necessary) is that the data entered does match the reality of the operations.

3-step method to detect the bugs in your process

What we usually do to fix errors is the following.

  • First of all we examine whether the output can be obtained by a different way. Then we try to apply it.
  • Second we run the normal process and the alternative process in parallel.
  • Third, we examine the discrepancies and look for the rootcause

The rootcauses of the Delta’s are the errors in the process that you must fix. Most of them will be of operational nature and will be caused by Humans.

What to do to avoid errors

The biggest challenge, once you have discovered the weaknesses of your process, is to find solution. And we know no better solution than what Japanese call Poka Yoke.

You may have never heard of this word. It doesn’t matter. The concept behind it is the following (Wikipedia definition) :

A poka-yoke is any mechanism in a lean manufacturing process that helps an equipment operator avoid (yokeru) mistakes (poka). Its purpose is to eliminate product defects by preventing, correcting, or drawing attention to human errors as they occur

Poka Yoke is the solution

In other words you must design some parts of your process in such a way that Humans can’t misuse it. Although Poka Yoke has been used in industrial settings, you can use the concept in any context. For example : if you are using an Excel file to input data, you may want to protect certain fields to get the data in the expected format, use drop-down lists to restrict input to certain pre-defined choices, forbid blank fields.

Once again this may seem obvious but AS IS processes are most of the time very far from being used as they had been designed originally.

Conclusion

Central database and ERP will only work well if they contain data of good quality. The processes enabling data input must be bullet proof and they usually aren’t.

Using an alternative process to reach the same end results will enable you to see discrepancies. Those discrepancies will lead you to the rootcauses of these Delta’s.

The Poka Yoke principle will help you imagine new secured ways to input data and will help you guarantee the quality of your results. If you want to hear about a few success stories, feel free to contact us.

Image : copyright tec_estromberg via Flickr



Posted in Strategy.

Post your opinion

Your email address will not be published. Required fields are marked *