9 mistakes that will kill the best data analyses
Can you mention a few more?
- Sampling or design of experiment not properly done
- Non robust cross-validation
- Poor communication of results to management or clients
- Poor data visualization
- Does not solve our business problems
- Database misses important data or fields
- Failure to leverage external data
- Can't make business data silos to "talk to each other"
- Developpers (production people) and designers speak "different languages"
Replies to This Discussion
-
Permalink Reply by Ed Fine on
-
Not getting all the relevant business constraints. What a way to look bad and lose the trust of your audience. "Oh, sorry, I did not realize that undermines our main client's contract and breaks the law. My Bad".
-
-
Not understanding the distribution of the population (i.e. assuming that the sample is from a normal population when it is not) and making inferences based on wrong assumptions.
-
-
Data computation and analysis takes a long time to react.
-
Permalink Reply by Kirk Fleming on
-
No. 5 applies to any project, of course, not just data analysis. While this mistake may seem like a 'no-brainer', my experience has been that it's common to finish an effort and not know if any real problem has been solved. While everyone will agree it has been, this is more a 'policy of success' than any actual measurable demonstration of it.
-
Permalink Reply by Anil Bhargava on
-
Client confusion as to what he/she needs out of Advanced Analytics to achieve for them.
-
-
Data quality is very important thingjn data analytics assignments. Not easily available. IT folks will kill you on given pain to them.
Resources
Videos
© 2021 TechTarget, Inc.
Powered by
Badges | Report an Issue | Privacy Policy | Terms of Service
