4

Fire your Data Scientists – They Don’t Add Value

Data ScientistYears ago, I was on a project to improve production and product quality through data analysis. During the project, I heard one man say: 

“If I had my way, I’d fire the statisticians – all of them – they don’t add value”. 

Surely not? Why would you fire the very people who were employed to make sense of the vast volumes of manufacturing data and guide future production?  But he was right. The problem was at that time data management was so poor that data was simply not available for the statisticians to analyze.

So, perhaps this title should be re-written to be: 

Fire your Data Scientists – They Aren’t Able to Add Value.

Although this statement is a bit extreme, the same situation may still exist. Data scientists frequently share frustrations such as:

  • “I’m told our data is 60% accurate, which means I can’t trust any of it.”
  • “We achieved our goal of an answer within a week by working 24 hours a day.”
  • “Each quarter we manually prepare 300 slides to anticipate all questions the CFO may ask.”
  • “Fred manually audits 10% of the invoices.  When he is on holiday, we just don’t do the audit.”

This is why I think the original quote is so insightful.  Value from data is not automatically delivered by hiring a statistician, analyst or data scientist. Even with the latest data mining technology, one person cannot positively influence a business without the proper data to support them.

Most organizations are unfamiliar with the structure required to deliver value from their data. New storage technologies will be introduced and a variety of analytics tools will be tried and tested. This change is crucial for to success. In order for statisticians to add value to a company, they must have access to high quality data that is easily sourced and integrated. That data must be available through the latest analytics technology. This new ecosystem should provide insights that can play a role in future production. Staff will need to be trained, as this new data will be incorporated into daily decision making.

With a rich 20-year history, Informatica understands data ecosystems. Employees become wasted investments when they do not have access to the trusted data they need in order to deliver their true value.

Who wants to spend their time recreating data sets to find a nugget of value only to discover it can’t be implemented?

Build a analytical ecosystem with a balanced focus on all aspects of data management. This will mean that value delivery is limited only by the imagination of your employees. Rather than questioning the value of an analytics team, you will attract some of the best and the brightest. Then, you will finally be able to deliver on the promised value of your data.

FacebookTwitterLinkedInEmailPrintShare
This entry was posted in Big Data, Business Impact / Benefits, Data Integration, Data Integration Platform, Data Warehousing and tagged , , , . Bookmark the permalink.

4 Responses to Fire your Data Scientists – They Don’t Add Value

  1. Mike Gagnon says:

    Bingo.

  2. Uzair M. says:

    Most of what you said is True, but only because it is obvious. Of course in the near future the transaction between the databases will advance and most of the work will be done much quickly than it is being done at this time and will be more accurate. In-fact there are some tools that are already available that only a few people know of to make the data processing faster with much higher accuracy rate. What infomatica should be concerned the most is how to do the transformation more securely without any hiccups.

    You did not provide any solid information for that I give you 1 star.

  3. Monica says:

    @Uzair thanks for your star.

    I do take your point about transactions & databases getting faster, and of course new tools – this is clear. However I do not believe it is all about speed of moving data. What I am more concerned about is the speed of the ‘complete analytical transaction’ from hypothesis concept, through data gathering, analysis and final conclusion. This complete transaction includes the ability to quickly identify and integrate a data source into the tool of your choice. With current approaches and technologies, this is still estimated at 80% of time spent – which is a pity given how much attention & investment analytics has received recently. My point behind a balanced focus on all aspects of data management is that unless you address this 80% bucket, the fastest tools in the world can only make you go 20% faster.

    As for accuracy, I am curious what you mean by this. I trust today that all data integration technology can accurately replicate a value from one system to another. But for analytics, context is key and I role this into my idea of high quality data. Our DQ and MDM tools contribute to data quality beyond a simple accuracy check, overlaying context and highlighting potentially poor data held in the source system.

  4. Umer Mushtaq says:

    As for accuracy, I am curious what you mean by this. I trust today that all data integration technology can accurately replicate a value from one system to another. But for analytics, context is key and I role this into my idea of high quality data.

Leave a Reply

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

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>