The YouTube media company Linus Media Group has been making the news this week for its controversial decision to auction a prototype owned by tech startup Billet Labs, along with inaccuracies that led to an unfair negative review of the prototype.
An ongoing outcry by the community led to this interesting apology video:
On two occasions, you can spot the company’s executives blaming what is ultimately faulty data management that impacts both their external product and their internal processes.
Because any excuse is a good excuse to dive into data, let’s take a look.
Tech Tips for Linus
The first problem that is obvious to anyone watching is the almost complete lack of inventory management. The only inventory that appears to exist is a person being tasked with obtaining devices from suppliers and keeping track of the conditions attached with each item through e-mail messages with the suppliers. This is likely a method that worked when the company was smaller and handling relatively small numbers of items, but as the number of items grows, the harder it becomes to keep track of them all in a way that avoids human error.
This is especially critical for a company like LMG, which by its very nature has items that belong to other companies between their hands, exposing them to potential legal issues that could stem from inadvertently mishandling the items.
A proper inventory system to keep track of all the data points regarding each unique item along with a complete tagging strategy to ensure that all the possible data points required are appropriately tracked would have avoided much of this controversy.
How do you create such a strategy, you ask?
I’m glad you asked.
The first step is taking a typical object that would need to be inventoried and asking yourself, what might I need to know about this product? Off the top of my head, in a scenario like this, you might think of the following data points:
-Internal unique item ID
-Supplier
-Manufacturer
-Location in warehouse
-Date of reception
-Expected return date
-Actual return date (once returned)
-Owner
-How to treat the item after
-Sensitivity rating (is this publicly available or is it a private unreleased item?)
-Approximate value
-Contact info
-Item type
-Item description
-Comments field
Second, systematically track all this information in a centralized database as part of the procurement process.
Third, refer to this database whenever dealing with any item in the inventory.
Once this dataset is built, you can enter into the magical part of the process: automation and reducing or even eliminating the potential for human error. One might think of building alerts letting you know of approaching return dates or missed deadline, or a dashboard letting you know how many items you have in your possession at any given moment and which items are more sensitive and require a higher level of care.
Go figure…
The second problem they address is the recurring errors plaguing their reviews in recent months, including the same controversial review where a mistake led to a highly negative review based on a configuration that inconsistent with the way it was designed to be used.
On other occasions, fixes were mistakenly not applied to the data before publishing, or the charts published in the video belonged to the wrong product.
This problem is more difficult, but it could be addressed, of course, by putting more safeguards in place to ensure data quality. Isolated environments for working on different projects would be a first step, making it harder to mix data from a project with another. A more robust quality control process would help, with a checklist of quality points to look for in each video produced before it’s allowed to move on to publication.
The bad, the worse and the ugly
While most of these points can largely be attributed to an organization that grew slightly too fast for its own good and needs to shore up its processes to keep up with its newfound size, that doesn’t explain it all. A company can have the best data in the world and well-oiled processes, it still will encounter issues when its employees are overworked and harassed. A rotten culture that celebrates pressure and incredibly tight deadlines will always result in people cutting corners to meet management’s expectations.
Pressure culture? Now that’s a topic for another post…