Sometimes it’s hard not to find an electronic device in your home or office that does not connect and send data somewhere via the Internet. For companies that produce or maintain a service involving these products, this can be a good and a bad thing in terms of data.
On one hand, if you work at one of these companies, you have a great touchpoint with a user or customer that will likely enhance the lives of your customer as well as your service. On the other, you are likely generating an immense amount of data that may never be used, might be too costly, or be inaccessible. Many call this information “dark data” and it can really hurt your bottom line as an IoT focused company, if it hasn’t already.
According to the IDC, the amount of data created by connected IoT devices will see a compound annual growth rate (CAGR) of 28.7% from 2018 through 2025, generating a total of 79.4ZB of data by 2025! “Data is the new oil” is likely a phrase you have heard before but I’m going to introduce a new concept for you… “More Data, More Problems” and tell you three reasons why.
#1. Storage
Server storage is getting cheaper, especially cloud storage, and that’s a great thing for most businesses. Most companies are likely paying a few cents per GB per month on average (with some larger companies likely even paying a fraction of a cent) However, cheaper storage isn’t cheap anymore when you constantly need more of it. This is why it’s essential to establish a server monitoring performance process.
As your business scales, so will your data, and often you will have a tough time forecasting just how much you will need. More data generally also means that it will be harder to sift through and find nuggets that are useful… hence rendering a lot of that once precious data into “dark data” quickly.
#2. Security
CCPA, GDPR, and HIPAA – what do these three things have in common?
There is more focus than ever on all three of these topics by regulators and consumers in 2020 and beyond. Couple that with the fact a security vendor has detected that over 100 million attacks on IoT endpoints occurred in the first half of 2019 alone. That is enough to keep you up at night.
Even if data is largely unknown or inaccessible, that does not mean you still do not have to protect it (and spend time and money in the process). What is worse than not being able to use valuable data? Paying to protect it from malicious intent as well.
#3. Missed Opportunities
Exponential data growth and even a moderate team turnover rate can mean a lot of lost learnings and opportunities without being able to access underlying data. Meeting notes, chats, documents all often get lost in the shuffle and many teams are forced to re-learn the same things all over again.
Unfortunately, the missed opportunity value is hard to measure therefore making it hard to prioritize. Historical application telemetry data points, support cases and exception reports are data treasure troves that often get buried without much hope of seeing the light of day but often present the most value to product strategy.
So, What Can We Do?
If you are positioned to capitalize on the growth in the IoT space, you need to be aware of the risks mentioned above. Chances are if your company is more than a few years old, you already have some of these problems occurring. Unfortunately, there is no silver bullet solution to avoid the pitfalls in any of the areas above but to start the conversation, I will make these recommendations for each area:
- Storage – Extend to the cloud. If you have a lot of your data on-premises, there’s no better time than now to explore cloud storage offerings and potentially moving more to the cloud to capitalize on stretch tables, data lakes, and blob storage capabilities.
- Security – Discover and document your dark data. Doing inventory via data lineage and establishing a data dictionary is the first step in understanding what needs extra protection and what may not (if anything).
- Value – Keep track of lessons of the past. Long lost presentations, chat histories and email chains need structure. Put together a plan to lend structure to this data. The structure will provide an opening for analysis, and you can put this data to work for you rather than spending your time and money working for it.
John Kirkpatrick is a Marketing Manager at SentryOne, a leading provider of database performance monitoring and DataOps solutions on SQL Server, Azure SQL Database, and the Microsoft Data Platform. John is based in Charlotte, NC and enjoys following the latest in business and technology trends.