Automate efficient processes only, but not all efficient processes need automation

There are many instances of automation projects  where a particular process  is targeted  for  automation but  in the midway they  realize that the process needs  to be streamlined in the first place . This is not an easy position to be in and if the budgets are high it can cause serious damage.

“ The First Rule of any technology used in a business is that automation applied to an efficient operation will magnify the efficiency. The second is that automation applied to an inefficient operation will magnify the inefficiency”: Bill Gates

The story of Marketo published by Mr Blume (http://www.brightcarbon.com/blog/marketing-automation-a-failure/ )  explaining  why their  marketing automaton failed , sites lack of a clearly defined process as the top reason.

When we don’t know what’s been automated or haven’t clearly understood the need or the efficiency that we can derive from it , even with great tools  and developers around, its bound to fail. Identifying an efficient manual process is only the first step but the  question is whether  we really need to automate it ?

An efficient process need not be a candidate for automation.  Lets take an example of a   SAP Unicode migration project  . We know that it’s a whole lot of process from preparing the Unicode system to Fixing languages  , Exporting and importing them and building the target system .

Now  the SAP Unicode migration is a clearly defined activity evolved over a long period of time and it mostly efficient . By the look for it we can come to a conclusion that there are possibilities of automating parts of it , especially in the post and pre migration steps , but do we need to automate it ?

The  time spent in automating the Post/Pre  Unicode Migration  steps may not be worth the effort   versus manually   performing  it.

Nowadays most of the systems are Unicode and Unicode migration projects are nearly negligible so automating  and maintaining the automation may not be really worth the effort.

Efficiency of Processes alone is not a determining factor of what needs to be automated.  Below are some of the things which we can look for..

• Identify Repeated Tasks
  • Tasks which are mundane and manually done periodically  with the same data input are ideal candidates for this . Automating this can save people complaining about doing the same thing repeatedly. It frees your core resources from performing L1 tasks and helps them focus on complex and innovative activities.
    • You may be doing a system copy from the same Production to Quality system and the same set of post steps will be carried out . Now this is repeated task with almost the same inputs every time its done.
    • System Install Post steps and Validation step steps, many of which are pretty standard are  great candidates to be scripted.

• Look for efficient manual processes:
  • Look for processes which works well manually and doesn’t have any redundancy and lack of flow

• Brainstorm with the delivery team
  • The Best people to tell you what’s worth automating are people who work on the manual  task .  When you let their creativity flow you will be filled with thoughts and possibilities .

• Identify Tools and development skills needed to automate the process or an activity
  • Based on your budget you have highly sophisticated automation tools or  even tools like SAPGUI scripts,  Excel macros/VB Script and Shell Scripts.  To do the Proof of concept you may  have to start with what you have so VB Scripts and SAPGUI scripts would be a good place to start.
• Document , Document and Document
  • Document and manage the versions of the tools  that you are developing . Remember,you cannot maintain a tool that’s not adequately documented.

• Never Compromise Security
  • You don’t want a automation tool that has security vulnerabilities . The tool has to be tested and should be compliant with Security needs of an organization.
• Have a feedback  mechanism
  • Maintain a team room where your team can give feedback about the tool . SAP Automations may be version dependent so you want to test it on different versions and document the results to improve the tool and to handle exceptions.

• Maintenance is the key
  • Remember, your task of automation will not end when you release it to the team. In fact that’s  only the beginning. Maintenance  / Bug Fix and Enhancements will be needed if you want the automation to be successful in the long run.

• Remember there is no replacement to Human thinking
  • Though cognitive computing has really taken off its still not a match to the creativity  or the ability and intelligence of a Human mind. You still need people to manage to manage exceptions and cases where the tool fails to react or comprehend.

• Don’t get too attached  to the automation
  • An efficient  Automation today need to be efficient 2 years down the lane . You need to measure the efficiency in terms of  using the automation against a upgraded tool which the vendor may provide. So don’t use it because its developed. Use the tool only if its brings value to the overall Business process.

• Make sure your resources don’t get out of touch
  • The  tradeoff  in automation is that  your resources may  be out of touch or may loose the corresponding manual skillset . Manual Drills once in a while is imperative if you want  your Business to run  incase  the automation collapses.

At the end of it ,the key is the return of investment .  Plan and only embark on a Automation project if you feel the automation is going to bring in value . A great candidate is usually recognized in the brainstorming session . Work closely with the delivery team , you will know it when they are happy using the tool . When you see a smile on their face you know you have created value.