1. Data cleaning
Small variations in data and its labelling can make interpreting and extrapolating difficult. Data cleaning removes misspellings and differences, no matter how small, to make your data easy to export into different formats. For large datasets, cleaning and organisation is a time-saver - various departments will need to use the data, so being able to present it to them clearly and on-demand is essential.
2. Visualisation applications
Data and information can have limited use without the ability to see the results pictorially. Visualisation tools turn data into charts, graphs or diagrams of your choice, comparing multiple streams of data all at once. Many tools are quick and easy to use (often with drag-and-drop functionality), so appropriate for beginners outside your IT department when making reports.
3. Geographic Information Systems (GIS)
Location-based data can be one of the most important for certain business types, particularly when they operate in different markets around the world. GIS creates maps using data, showing colour-coded differences, and allowing the user to zoom in for more details. Other data types can be overlayed, plug-ins installed, and maps exported to other formats.
4. Temporal data analysis
Timeline visualisations filter and sort information for you, giving you various ways to interact with it. For organisations that are mainly interested in how data changes over time, temporal data analysis shows timelines of data and visible changes.
5. Data prediction
Analysing past data is ultimately to inform future decisions. Data prediction analyses the numbers to find consistent patterns and relationships, builds models using this data, and then generates predictions. A 1997 piece in Statistical Computing and Graphics referred to it as “a blend of statistics, AI, and database research”.
6. Data scraping prevention
Malicious tools can obtain data from your site or database by using scripts and bots that automatically crawl it. Unfortunately, many firewalls don’t prevent sophisticated tools, but user pattern analysis can at least highlight them. Websites like Yell.com use it (they’re particularly at risk from data scrapers) to detect what’s natural human usage and what’s clearly a bot.
7. Risk management
Some tools can use data to predict financial risk, issues within the supply chain, and where cost needs to be allocated. Educated human insights will always be valuable in business, but some tools can make far more accurate predictions based on real data. The more data it processes, the more accurate and intelligent its observations.
8. A centralised database
Business-critical data needs to be managed from one comprehensive database. It can give permissions to staff who need different data for different functions, saves a significant amount of time day-to-day, and makes tracking usage simpler. All of this will be essential after GDPR is introduced in May 2018, and businesses that don’t comply may be fined.
9. Collaborative tools
Some data tools need to work for the whole department. Collaborative data science platforms help a team build, prototype, test and deliver data products, whether they’re a proficient coder or not. If a problem needs solving, or a question remains unanswered, your department can use collaborative tools to build the answers.
10. Natural language explanations
To make data analysis easier for people regardless of their area of expertise, natural language generation translates data into easy-to-understand interpretations. Data is only as valuable as the insights it provides, and those insights aren’t possible if the reader is unsure how to interpret what they see.