It’s not about how much data you have, but how you use it.
July 17, 2019 - Machine Learning
Data is at the heart of all that we do in business; not just us, but everyone! From the most simple metrics that measure a company’s performance, to the seemingly endless possibilities afforded by modern machine learning, data is everywhere and is the backbone of any company’s digital transformation efforts.
Everyone acknowledges that we now have more data than ever at our disposal. We hear about vast data lakes and trillions of bytes of data in the cloud; and companies lose sleep wondering what to do with their data to maximise productivity and profitability. Whatever the size of a business, data is changing the way we do business.
Back to Basics – Your data
So let’s start with the basics, Why is data so important to an organisation?
Without data, it’s impossible to know who your actual customers are or if they like and use your products, and how they’re using them.
Data analysis and tracking helps companies pinpoint performance challenges. This facilitates a better understanding of each part of the process and can help define which steps need to be optimized, identify key risks and capitalise on positive performance. Data analysis helps you understand and improve business processes to reduce wasted money and time.
Effective data visualisation is the new method of choice to interpret data performance metrics. Plotting data in an easy to use visual format is light years ahead of poring of reams of spreadsheets. We’ll talk more about that, later!
The challenge for most businesses is to truly understand customer behaviour. Capturing and processing data has become increasingly challenging. With data volume, veracity, and velocity all increasing, companies must pivot into new technologies to manage the scope of the data landscape.
Humans are notoriously good pattern matchers, however our brains are not good at correlating multi-billion sized data sets. Part of the challenge with being pattern matchers is we cannot derive new information. It’s really hard to find something you don’t know to look for. This is where the value of statistical models and visualization really comes in.
Using data-mining we can derive a lot of new information that otherwise may not have been possible. Correlating large amounts of data gives businesses more reliable understanding. With the right types of data and models we can get stakeholders new predictive and prescriptive information.
But it’s not just about having the data, it’s how you use it!Botros Gerges
Deriving meaningful insight and converting knowledge into action is a tall order – whittling data down to what’s most helpful is a great starting point.
Companies need access to their data to drive digital transformation and this is a challenge that every business encounters. Companies certainly have the raw data collected over decades, from a range of internal and external sources. The challenge is how it is leveraged and transitioned into a competitive advantage for each company.
Modern enterprises are starting to evolve from those previously siloed operations, where data and assets were locked away, to progressive businesses, where data flows freely and readily and can be used and interpreted by everyone.
Incredibly, a lot of businesses have no single person, or department responsible for data analysis (according to Dun and Bradstreet, 40% of businesses do not have anyone responsible for data in their organisation).
Utilizing Digital Platforms
Many companies are still using multiple spreadsheets, paper documents and manual entries. This results in high dependency on SMEs, manual work, high effort on low value tasks. It seems old-fashioned, to be frank.
If a company can harness data through user-centric digital platforms it enables those companies to make informed decisions, quickly.
As well as this old-fashioned paperwork, there is rarely a single source of truth (truly accurate, representative data). Disparate data, manual entries, and a high dependency on SMEs to interpret data results in poor data quality and inconsistent data governance.
This ultimately handcuffs the user from having high confidence in the information generated from the data.
Companies need to see data as an organizational asset that has value.Botros Gerges
Companies need to see data as an organizational asset that has value. They need to treat data as something that can be retrieved, ingested and integrated, even governed.
With cloud computing, everyone from the smallest start-up to a Fortune 500 will be able to couple big data technology with advanced data analytics. The power to uncover new operational and market insights, and untapped customer segments, is at their finger-tips.
Once we have clarity on these decisions, we can then pave the path to solving the core challenges by leveraging data.
Finger Food Advanced Technology Group™ works with clients in a few key areas:
● Identify the true source of their data
● Provision data on the cloud and work with partners to ingest and engineer data in such a way that the application can visualize the data – tailored for the user to drive decisions.
● Data cleansing and applying AI to identify insights that are not simply patterns.
● Build technical architectures and work to set up the environment to retrieve data.
● Visualize their data in a simplified format that anyone can view and interpret.
Data Visualization and Digital Twins
For most sophisticated companies, the answers to most challenging questions lie in their data. That said, super-intelligent analysts still rely on AI to collate salient, usable data, that they wouldn’t otherwise have access to.
Data visualization on each side of the spectrum is the combination of science and art. The science is driven by data engineers, data scientists, and developers. They get to ingest, store, clean, process, and implement the raw data.
On the design side, UX and UI designers, and 3D artists blend the aesthetics and create the experience.
Data visualisation when done correctly take large amounts of data and present it in a form that our brains are accustomed to ingesting. Data visualization is often misunderstood, while statistical graphs are easy for comparative analysis, data visualization is designed for many different users looking to gain different types of information.
The value from data visualization is interaction, with the ability for users to explore the data. Data visualization increases the effectiveness of data; The effectiveness comes from users lead interaction. Users lead engagement results in increased cognitive load, ultimately helping users retain and recall information during critical business decision making.
Data visualization has become a standard tool for many businesses. By using simple visual elements like charts, graphs, and maps, data visualization tools provide an accessible way to see and understand trends, outliers, and patterns in data. The next stage is to use state of the art holographic technology where users are fully immersed in a visually stimulating, interactive data world.
A digital twin, takes it one step further – they still rely on data, but it actually models and simulates a range of operating scenarios. It’s often viewed as a virtual representation of a physical product or process, used to understand and predict a live operation’s performance characteristics. Digital twins can simulate, predict, and optimize products or process before companies invest in physical prototypes.
A survey by Prysm revealed that 80% of organizations reported more accurate decision making when using data visualization tools, and 86% of companies said that data visualization enabled them to make decisions faster.
Data Storage and Integrity
Once we engage with a client and focus on their challenges, we have a pretty standard set of requirements that we detail around their data:
● We need to understand what data they actually have, how it is structured, and if it is deemed a trusted source of truth.
● We need to know where it is ‘kept’ – this can range from a bound ledger, to a filing cabinet, to the cloud!
● It’s important to also understand how it is currently being used or even visualized.
● We need to know what data is sensitive and how it must be protected, including relevant regulations and laws.
In a recent assignment with a Canadian energy transportation company, we worked closely with the company to quickly understand challenges in interpreting and communicating data that could inform critical decisions on pipeline safety.
We created a holographic visualization framework from raw data (they provided) to better allow experts to fully understand ground behaviour and pipe movement without the need to dig up physical pipes. It provided the company with a simple-to-use tool that quickly and easily showed a detailed 3D view of operations, based on recent historical data.
This holographic rendering of geological hazards could redefine the assessment process for a company. It would give pipe integrity experts extensive visibility, drive down costs, while reducing the threat of incidents on the pipeline.
Data was at the hub of the work and identifying the data type and source was foundational to the solution. This enabled us to work with the client and partners to provision data to the cloud and for us to build APIs for the application to access and visualize data. Data is also critical for us to develop simulation and optimization models that we can then scale system wide.
The next step in this process is to move closer and closer to visualizing real time data. The potential processing power of the cloud, may just speed things up, in the coming years.
We also worked closely with Goldcorp. Mining operations thrive on data, and making decisions based on visualizing a problem, can save millions. In this case, the challenge was to evaluate if new technology could improve shovel operator precision for material types moved from the face: ore versus waste.
The test was a success. Goldcorp learned a great deal about how this technology could potentially impact a mine’s productivity. Tests of the technology are on-going.
The message around data is clear. It’s impacting all that we do, and companies need to embrace this. Data driven digital transformation shapes how companies digitize every part of their business, as every part of the business needs efficient technology to operate. This is an era where everything is data driven and everything needs to connect. In turn, the way companies think about data and technology needs to progress – for companies large and small.
When it comes to data, you need to be able to count before you can multiply. Having the right infrastructure in place and knowing where your data is and having the right governance of the data is critical to ensuring companies can effectively use advanced technology such as AR/VR, ML, AI, etc to impact the success of their business. Otherwise it’s garbage in, garbage out!
For more information about the company, or if you have any further questions, please email our team email@example.com.