Here is a collection of 10 harmful, but preventable, things you shouldn't be doing as a digital analyst.
Last week I learned the word iatrogenesis.
According to my good friend, Merriam-Webster, iatrogenesis is the "inadvertent and preventable induction of disease or complications by the medical treatment or procedures of a physician or surgeon."
This is not a bad doctor or an evil doctor or even an incompetent doctor. He didn't mean to leave the clamp in your abdomen. He didn't try to let you catch pneumonia from the patient in the next bed. He didn't want to break your arm while pushing you in the wheelchair for your broken leg. Sometimes these things just happen.
But, as inadvertent as they may be, they are preventable and that's the sad part. This is why the phrase Primum non nocere (First, do no harm) is so important to medicine.
Here then is a by-no-means-exhaustive list of iatrogenic things you may be doing and if you are... stop it!
We're understandably excited about the wonderful world of data, but don't go all One Direction asphyxiatedon us, OK?
Data is a tool, not a godsend. Data is a useful addition to the human decision-making toolkit; it is not the end-all and be-all. Humans have been making mostly good decisions since time immemorial without more than a few data points. Your petabytes are good, but they won't cure cancer... just yet.
Can't make a decision because you don't have enough data? You're doing it wrong.
Can't make a decision because you don't have good enough data? You're doing it wrong.
Humans have been making mostly good decisions since time immemorial without more than a few data points. You can, too. Be pragmatic rather than systematic.
Your job is to come up with (or help others) come up with new insights based on data. They do not need to know how hard your job is, how messy the data is, or how complex the models are. They have you to deal with all of that and are not nearly as interested in how the sausage is made as you.
Image source here.
Your pristine data isn't the goal. The goal is to reveal interesting things about people. One thing we know for sure is that people do the strangest things when you don't know what they're doing, so we use data to try and uncover their motives and predict their behavior.
The most elegant, crisp, comprehensive mathematical model is useless unless it is actually predictive of human behavior. Common sense beats algorithms all day.
Unless you want to be thought of as The Answering Machine, make sure you understand why people are asking you questions. That way, you can give them insights based on what they are trying to accomplish rather than simply feeding them numbers that may or may not be useful.
Give priority to questions when there is revenue is on the line or if your boss's boss's boss has asked for it. Otherwise, view each request for data as an invitation to a conversation.
You did not have 36 million, 285 thousand, 8 hundred and 27 people visit your Web site. You had somewhere between 34 and a half and 38 million with a 4 percent margin of error.
Communicate the fact that you are not an accountant. You do not deal in facts, you deal in probabilities. It will help everybody else feel more relaxed about the numbers.
Sometime, on your own time, sift through the data and see what it has to say. Sometimes poking it gently may surface an anomaly or a curiosity that make you tip your head and think, "I wonder..." This is good exercise and is the reason you got into this line of work anyway.
As David Weinberger likes to say, "The universe is analog, messy, complex, and subject to many interpretations." If you have some other information you can mix together in a reasonable timeframe, give it a shot. Data gets really interesting when you correlate it to something you hadn't thought of before.
While this is excellent advice to artists, it also applies to those who dabble in data.
Models are representations of the world at a certain point in time. Times change and your models should, too. When you make a model, give it a sell-by date and be sure to revisit it often. The economy, the competition, the weather, the consumer confidence index, and your kid's shoe size are all subject to change. Nothing you can do about it but be a little more flexible than you wish.
Do go out of your way to learn another new word tomorrow. It couldn't hurt.
Images via Shutterstock.
Join the Industry's Leading eCommerce & Direct Marketing Experts in Chicago
ClickZ Live Chicago (Nov 3-6) will deliver over 50 sessions across 4 days and 10 individual tracks, including Data-Driven Marketing, Social, Mobile, Display, Search and Email. Check out the full agenda and register by Friday, Oct 3 to take advantage of Early Bird Rates!
Jim Sterne is an international consultant who focuses on measuring the value of the Web as a medium for creating and strengthening customer relationships. Sterne has written eight books on using the Internet for marketing, is the founding president and current chairman of the Digital Analytics Association and produces the eMetrics Summit and the Media Analytics Summit.
IBM Social Analytics: The Science Behind Social Media Marketing
80% of internet users say they prefer to connect with brands via Facebook. 65% of social media users say they use it to learn more about brands, products and services. Learn about how to find more about customers' attitudes, preferences and buying habits from what they say on social media channels.
An Introduction to Marketing Attribution: Selecting the Right Model for Search, Display & Social Advertising
If you're considering implementing a marketing attribution model to measure and optimize your programs, this paper is a great introduction. It also includes real-life tips from marketers who have successfully implemented attribution in their organizations.
September 23, 2014
September 30, 2014
1:00pm ET/10:00am PT
October 23, 2014
1:00pm ET/10:00am PT