top of page
Understand your human limits, then exceed them.
Handling vast volumes of data quickly and correctly is simply beyond human capability.
Business intelligence is about understanding what's happening by unlocking 'insight' from business data.
Insight helps you make smarter decisions to maximise positive and avoid negative outcomes.
Collecting large volumes of data is a great start, but many companies fail to extract the maximum value.
There's dormant value waiting to be unlocked in every business.
Machine learning removes guesswork and human error from complex data analysis.
Computation automates data handling and simplifies repetitive analytics work.
Patterns reveal themselves in surprising ways. These patterns create forecasting models that predict the future.
Used well, statistical analysis will help you assign focus to the right business operations, to identify new opportunities and uncover organizational pain points.
Why Machine Learning?
Digitally simulate your business
Remove the barrier of 'time' from your analytics
A discrete simulation involves software that represents how elements move through your business, such as parts through a factory. You can run code to simulate what’s happening in real life.
This allows you to scroll backwards or forwards in time to review the root cause of a problem, accurately predict delivery or completion times, foresee maintenance requirements and capture congestion effects – such as where parts get backed up and create problems down the line.
It’s similar to congestion of traffic on a street: it looks very simple, but it’s not. There’s a complex wave that builds through the system.
If you can accurately simulate potential problems, then you can see what impact they would have on the system as a whole, giving you the chance to avoid them altogether.
“So, we’re using drones and neural network tech to take images of crops, using AI to make accurate estimations on the quality of produce, disease recognition, helping the farmer make price estimations.”
“In the kitchen, we have the knife. You can perform various operations with it, but you still need to know what operations to perform. You need to learn to read the data. What type of tools to use, what type of insights you can get.”
From the blog
Business Intelligence Capabilities
Applications of Artificial Intelligence, Applied to Business Intelligence
Make better decisions
Calculate organizational performance metrics and benchmarks.
Create complex dashboards that combine all sources of data to reveal macro, cross-business or inter-departmental insight.
Generate descriptive analytics from company reports with state of the art natural language processing.
Employ diagnostic analytics to reveal root causes.
Demand forecasting leads to inventory optimization, improved sourcing and supplier performance.
An optimization engine reduces the cost of outsourced resources, such as unnecessarily borrowed equipment.
Reduce financial borrowing requirements with more accurate predictions.
Supply chain forecasting
Run simulations, such as on the log files of factory lines, to accurately predict how materials and products move through the production line.
Use supply line foresight to course correct and maximize positive outcomes, or flag the potential impact of external or unusual negative circumstances.
Optimize transportation and logistics costs through better allocated warehousing and distribution channels.
- Optimise material and inventory costs.
React to future situations with predictive analytics, such as for maintenance cost prediction or more efficient inventory management.
Understand the full profitability of products including post service predictions of warranty cost/claims.
Understand the voice of your workforce with customized dashboards.
Analyse training effectiveness and include corresponding training recommendations for career progress.
Employee attrition analytics forecasts the likelihood of key talent loss so you can provide optimal incentive packages.
Text analytics captures the employee pulse, inferring sentiments from email and social media.
Every supplier has their own way of doing things, but this is what you can expect from a typical business intelligence adoption process.
Typical B.I. Process
Integrate data sources
Transform and clean
Explore and mine
Visualise and report
Gather business requirements and the intended use of the business intelligence solution. Every project begins by determining the key intelligence objectives and brainstorming the best approach, usually in the form of a joint workshop.
Collect and integrate data sources for unified access and collective manipulation. Our supplier will work closely with technical staff to collect transactional databases, file systems, flat files, etc, into a single big data file system (lake/warehouse/cloud).
Collected is transformed though a loading process and integrated into a database system (ETL). Scalable data processing scripts aggregate and clean into the appropriate format. Following this step, users are able to create SQL queries and perform ad-hoc analysis.
OLAP or Tabular solutions organize data and enable business users to define problems and draw conclusions through data exploration. Data mining techniques and provided tools will lead to greater insight about business operations.
To be useful, high levels of information need to be communicated clearly. Rich visualisations of data that address key business problems are presented in dashboards. Employees will save huge amounts of time by avoiding repetitive preperation.
Robotic Process Automation and 'Intelligent Automation'
Possibly the most obvious application of machine learning is the intelligent automation of repetitive tasks
Small, tedious tasks plague every business.
Automation of isolated tasks amount to small efficiencies. Yet, small efficiencies found across an organisation add up to exponential returns with scale.
“In about 60 percent of occupations, at least one-third of the constituent activities could be automated.”
Although many of these areas still require human involvement.
What the future of work will mean for jobs, skills, and wages: Jobs lost, jobs gained | McKinsey
Robotic Process Automation
There are three main reasons to adopt RPA technologies
Reducing work has a direct transfer saving of the substituted human's wage and time.
Manual activities are liable to human error.
Improves customer and employee experience, therefore increasing loyalty.
Human, Meet Robots
Two categories of robotic process automation
Un-attended Robotic Process Automations
Just let go
Unattended robots run on a server or back office computer in the background and can run 24/7.
These are usually well tested before being moved to production.
They perform work we don’t want to do, so we can be more creative. It's possible to restrict data access and design fail-safes that flag human attention.
Attended Robotic Process Automations
Attended robots sit on the same machine as a human agent.
They can speed up a process, validate something and perform administrative steps.
An agent can continue using their machine while a robot works in the background.
Four Stages of RPA Adoption
Uncover a positive ROI for a specific process (hours saved)
Short term focus is to ensure robots are effective in a specific environment.
This means reacting to the nuances of your organisation and incorporating as much business knowledge into processes as possible.
This process can take between a few weeks to a few months.
Demonstrate positive business unit outcomes
Now the aim is is to deliver multiple automated processes, often occurring in a single area and proves the business case.
The key factors to develop in this stage are intake strategy, governance, risk controls and change management.
Showcase strategic benefits of an automation pipeline
The operational learnings and structure from the previous stage is now defined and an approach is standardized for the rest of the business.
It could take a few years to convert other business units.
It's crucial (KPIs) are well defined and measured.
Full scale adoption and business transformation
With demonstrated success across multiple business units, it's now up to C-suite to create a company wide approach to RPA.
This can approaches to new RPA innovation, as well as the interlinking of robots from different departments.
The ultimate aim is improved customer experience at less cost.
Real Client Examples From the Collective