The biggest misunderstanding about data is the belief that people understand data.
The current talk of numbers in the context of coronavirus exposes this fact.
Currently there isn’t any data on infection rates in the general population (unbiased samples). It’s not there on mortality, or on magnitude of outcome improvement for patients receiving treatment either.
Despite of this lack of data, the majority of the narratives are discussing the infection rates of various countries and comparing their healthcare capacities.
Basically, the majority of the narrative is founded on an incorrect understanding of how the numbers map to an actual real life process of data collection.
How is it possible, that respectable, peer reviewed expert publications include such crass mistakes?
Regardless if the incorrect narratives were produced intentionally or not, the popularity of these headlines exposes the misunderstanding of data in the general population.
On the bright side, there are ongoing efforts to collect this unbiased data. Sadly, some of them were biased in their data collection.
What does this tell us about data literacy in companies?
The fact that all these analyses can just go through reveals that the vast majority of professionals, even those working closely with data, do not have the education necessary to draw correct conclusions.
Collecting and interpreting data is no easy feat. Given that most decisions require data, this means there is a lot of data work to be done.
Many companies are finding that even when 10% of their employees are working as data professionals, hiring more still brings benefits to the bottom line.
Empowering operational employees to take better decisions and automating tasks (and decisions), has the benefit of optimizing each process in depth. Such optimisations compound, often leading to significant impacts on profit.
Stop wasting time and get on the data train.
Sure, you can build up the data stack DIY, with time. But how long does it take you to iterate through all the things before you settle on an algorithmic automation? Do you have the talent to take you straight to the finish line?
In my experience, a natural grassroots evolution only gets you so far.
If you do not focus on data in time, you will get stuck managing departments that are created just to cope with insufficiencies in technology.
Additionally, rediscovering the wheel (the flat tyre) tends to be costly and will often discourage businesses from applying data further.
Data is not just an asset, but a frame of mind, with a focus on fact and efficiency.
How much data is enough? As much as is needed for the decisions you can make (but not more).
Start with management.
Data literacy for management is key for creating organisational change.
Get them a data coach that can assess their situation and teach them any missing skills. This can be someone internal or external, depending on the company structure. Internally, this role is often filled by the embedded analysts.
Data warehouses and literacy training for the operational employees.
Operational employees consume data differently than management. Here a detailed process view requires more data and different analysis skills.
With the help of a well designed data warehouse and a little tool training, we reduce the complexity of analysis to the level of a pivot table, facilitating what is known as self-service.
Data professionals are there to provide data automations and help with complex analyses, to avoid any wrong conclusions. Complex experiments can be notoriously tricky to perform correctly.
You can build your own data team, or hire one externally.
This data team helps businesses develop and implement their data strategy to ensure a good return on investment.
Data team provides data solutions that ensure good availability of data, and the simplicity that facilitates usage.
Get in touch at email@example.com .