The post-COVID world will undergo a massive digital disruption that will impact business processes. With so many people working from home, digitizing assets, moving to the cloud, and digital transformation (connecting people where they are using various channels) are becoming key priorities to companies entering a disruptive 2021. It is interesting to know that McKinsey states that there will be a new operating model where Business and IT will massively collaborate and become one and the same.
While going agile, becoming an API driven organization, and delivering data products have become the new normal for organizations changing their mindset and culture, they still struggle to understand “data-driven” decision making through advanced analytics techniques such as machine learning and centralized data platforms. Companies’ digital strategies must take a data-driven approach to provide meaningful insights at the right time to the right person in the right platform. I am outlining 6 topics that will become a crucial part of analytic product teams in 2021. Over the coming blogs, I will break out each of these topics and explain them in detail.
Analytics Topics to Watch in 2021
- Analytics of Things (AoT):
In a survey conducted with 620 types of IoT platforms, AoT seemed to an essential character and value proposition, according to IoT analytics.com. The most common being a mobile phone, telematics, digital twins, power generator grids, health diagnostics devices, smart cars, among others. Each of these platforms generates so much data. Without AoT, this data is just a piece of “Dark Data” waiting to act upon.
- Virtual Analytic platforms (Single Pane of Glass):
Every customer I talk to have at least 3 different analytic tools (Tableau, Cognos, Qlik, MicroStrategy, Power BI, Looker…) and rightfully so. Analytical tools are now readily available as SaaS services in app stores and can be spun up in minutes by a non-technical user. However, what this means is the ability for multiple versions of the same dashboard in multiple technologies creating confusion and lack of trust for CXO’s on their organizational analytics. The answer can be IT overruling and coming up with a single BI product or to have a single entry point such as Theia, which will improve the data literacy of the organization with a single pane of glass. Also, these virtual layers can provide a catalog of dashboards and reports regardless of the underlying producer of analytic dashboards with appropriate governance.
- Self-service and Data Prep:
One of the most significant pain points, as stated by Tableau, is the lack of a full picture of the data that users want to consider to provide meaningful Insights. To avoid wastage of time, analytic companies such as Tableau and Power BI released their own data preparation tools such as Dataprep and Dataflow correspondingly. These tools allow the analytic user to consider doing simple data preparation activities, including standardization, data cleansing, and pattern recognition by connecting the user with hundreds of data sources. However, be wary of using the data preparation tools to any form of unstructured data as they are not mature enough in that space.
- Process Intelligence:
While AoT is the capability where large volumes of datasets are analyzed to be useful, there is no purpose for this analytics unless there is an action tied to the analytics. Take a simple example of a smart thermostat, such as Nest. Nest regularly mines the data for temperature changes, people in the room, and appropriately control the room temperature with maximum efficiency. Also, using a mobile phone app, a person can track the decision-making process and the parameters for those decisions. This level of process intelligence can be automated (Alteryx coined the term “Analytic Process Automation”), which will simplify the decision-making process.
- Business-critical Analytic Technologies:
When a marketing team is building a CRM system to track their closed-loop marketing spend, they would want to have analytics over their marketing process, including segmentation, Clickstream analytics, and customer intelligence. These simple yet powerful analytics does not require enterprise data warehouses and complicated data models. They need the ability to hook up data in motion and create ad hoc analytic models and dashboards by non-technical users. Besides, intentionally, these analytic technologies deliver a limited set of visualization capabilities as relevant to the analyst users. Examples would be Analytic dashboards from Tealium CDP, Dynamics CRM, and Salesforce.com.
- Cognitive Computing:
Named from “Cognitive Sciences”, this is the field of analytics that applies human-like intelligence for specific tasks relevant to vision, speech, and text (commonly called unstructured data). These are compelling technologies that can take inputs in the form of natural language or machine vision and provide outputs that are relevant to a “context.” Take Watson Analytics, for example, which combines AI and analytics and delivers predictive and prescriptive analytics for decision making. Cognitive Vision (CV) has become a typical implementation pattern used in driverless cars and large machines such as oil drills.
According to Ulster University research, 90 percent of the data floating around in the digital world today “has been created in the last two years.” This means organizations that invest in data products and think of creative ways of Infonomics using Analytics can “Uberize” industries and can disrupt traditional ways of doing business. Perficient has practice areas that can help get to it faster. Reach out to us.