This a growing trend, not because people can spell "regression" but because of automation. What does it mean for data and analytics leaders? Not everyone likes or agrees with the term because of the expectations it sets. Let's dig deeper.
The "data analyst" is comparable to a country's middle-class, the workhorse persona that, through critical mass, is a substantial body to understand and focus on. In a country, they have the votes and whether they work or save money has a direct impact on the whole economy. The data analyst has the same implications for analytics success.
Just a few years ago, self-service, internal user communities, and training were the go-to tools to grow and enable this vital user segment. Vendors innovated to (other than to make money) move business people into a group called "citizen data analysts" by automating data analysis, data loading and other tasks that used to require more knowledge. These citizens are not quite as technically-savvy as a true-blue data analyst, but with the right tool, it doesn't matter anymore.
At the same time, data science became hugely important but finding an actual real data scientist was like finding a warm spot in a swimming pool. Insurance companies had actuaries, but everyone else was left out to dry. Guess what? Vendors are doing it again. We now have "citizen data scientists," and vendors are trying to help us democratize data science. Yes, it is a good thing.
Anyone who has an access tool id and password today can be part of that group. Many of the data access tools today have simple data science analyses built-in. You will recognize these as natural-language queries, narration, augmented and automated data preparation, automated advanced analytics, and visual-based data discovery capabilities. These capabilities have been drivers of new purchases lately. One example that keeps impressing is Power-BI, with its crazy-paced innovations and mind-boggling price point.
One strategic assumption is that a lot of business value will be derived from these types of capabilities - more than through traditional model creation. Organizations that are serious about monetizing their data cannot afford to ignore this trend. If anyone capable of typing in a question in plain English can do some data science, this enables a whole organization, not just a few quants. The data scientists can focus on more complex problems.
Anyone who has graduated from college or university in the past ten years will be tech-savvy. Organizations can, therefore, implement a Data Strategy that leverages them by giving them three tools (these may eventually merge into one platform, and we see evidence of that in the market already): a cloud-based, modern data access platform; a data preparation platform (a "citizen ETL tool"); and a data catalog. This is where the market is going. Coach business leaders and decision makers about the potential transformational impact that such augmented analytics can have if used by many. However, also stress the need for responsible use and governance on the analytics produced and avoid potential unintended consequences, perhaps by developing guidelines for appropriate use of augmented analytics tools and capabilities and put an emphasis on people and process.
Here is a link to an article that argues against the term. I am always amazed at how people think about real problems in the form of impediments to delivering value from data...and they come up with great ideas or technologies. Understanding processes and how people work is essential - all technology does is "implement or simplify" some process(es) for a particular group of people. Someone shared this with me this morning, and I am sharing it back. I am not sure I agree entirely with the opinion presented (another debate), but the explanations given contain excellent nuggets. Notice the discussion of PODS and the DATA SCIENCE CANVAS. Super!