1. The Chief Data Officer (CDO) is not a trend
Over the past 20 years, responsibility for Business Analytics has shifted within organizations. First BI was placed with the IT manager, later it was given to a CIO or CFO. And sometimes it was given to the CMO (Chief Marketing Officer). Now the idea is that BI & analytics will end up with the Chief Data Officer (CDO). One of the research agencies predicted two years ago that 90% of larger organizations will have a CDO in 2019. An enquiry amongst our larger customers makes it clear that this percentage is not even close to being reached.
But more importantly; does the rise of the CDO not indicate the failure of the function of CIO? Because what responsibility would be left for him? And will a CIO allow a CDO alongside him?
The following developments stand out:
- The Chief Data Officer now often has a temporary goal which is on the one hand creating a ‘sense of urgency’ within an MT and the organization for newer areas such as data quality, a data-driven architecture and people, experimenting and innovating with data and finally translating business strategy to data strategy. The Chief Data Officer is more of a role than a function. Important here is for a MT to identify the core competences and invest in them firmly.
- Organizations continue to look for ways to quickly fit in new core competencies. This often leads to the tendency to fit new core competences into a single person, function or department. This inside-the-box thinking often turns out to be unproductive and outdated.
- In agile organizations in which teams have free access to a lot of data, you should actually ask the question: who in the organization is in reality not a data officer? And who does not make use of data science?
2. Open Data sources continue to strongly increase
The number of open data sources with large datasets is rapidly increasing. Most freely accessible datasets now cover the following areas or are offered by the following organizations:
- Politics, government, and government-related open data sources. Examples include data sets relating to elections, policy, health care, science, climate or, for example, education. Important sources are data.gov from the US government, the opendataportal from the EU or the open data from the English government . But this also includes the CIA world factbook, data from UNICEF, UNESCO and the World Health Organization.
- News sources and encyclopedia sites, including extensive data sets on climate, weather and sports. Nice examples of this are the sets from the New York Times dating back to 1851 or those from Wikipedia.
- Data sets made available by social networks such as Facebook or Instagram.
- Other big data sets made available by institutions and companies.
The government is now increasingly taking the lead in this, for example in relation to transport and mobility.
Startups are regularly established on the basis of more external data than internal data. It is only a matter of time before the moment comes when existing companies will be fully managed on the basis of external open data instead of internal company data. Will we soon be able to get rid of our KPIs and the data warehouse?
3. Agile organizations create a REST API strategy
Organizations increasingly opt for an agile architecture that moves quickly and smoothly with new information demands. This makes it possible to share data inside and outside the organization without human intervention or Business Analytics software. Via a REST-API, internal and external customers, suppliers and other (external) parties have access to a part of the data warehouse (DWH). As soon as these parties feel the need, through an anonymized call to the DWH, they can get automatic access to the data and use, exchange, enrich it by data mining, data science and algorithms, and share the outcomes. Many organizations are moving towards more transparent, more scalable and more flexible links. Often REST is used instead of SOAP because it is easy to use and understand and is therefore ideal for scalability and speed of sharing and adding new datasets.
4. An image speaks volumes
Communication in pictures and icons is still increasing, think of first emoticons and now emojis. Interpreting and converting to and from this ‘universal picture language’ offers on the one hand new data science possibilities and on the other hand new communication opportunities. In addition, billions of photographs and other visual material are made every week. These images can increasingly be used for data analysis and in the development of new revenue models. Organizations that excel in this area quickly gain an advantage over others.
Damage report via emoji
More and more companies are also responding to this trend, such as insurers. In some cases, it was already possible to report damage through WhatsApp and photos, and nowadays insurers even invite customers to portray the damage in emoji only. Car, sheep, ditch, splashes, sad face … it could be a lightning-fast alternative to the traditional claim form.
Diagnosis based on sensors and images
Secondly, images and photos allow remote analysis. Certainly in health care, the possibilities of diagnosing remotely on the basis of sensors and images are being considered. A good example is the SkinVision app which analyzes a photo. With this application on your phone, you take a photo of a birthmark with which the app analyzes whether this spot is skin cancer or not. Various media outlets as well as the Catherina Hospital in Eindhoven in the Netherlands investigated whether this app worked well. The outcome was that the results come close to the average ability of a dermatologist. But also think of applications in the insurance industry or real estate.
5. AI will delve
The algorithms were already there, but the large amounts of data were missing in the past. Because of this, Artificial Intelligence has still not impressed in broader applications. Now that more open data sources and more different types of data carriers (photos, videos, blogs, posts, log files and sensor data) are available, investments in AI create real added value for organizations.
AI with bankers and insurers
The impact of AI on our society, organizations and functions will increase. With an ever increasing regularity we see Passionned Academy clients participating in our BI training or Data Science courses. Especially from the financial and real estate sector. Bankers and appraisers who want to understand how AI can be used to create a distinctive advantage over competitors. But this is also includes examples like the Dutch Ministry of Infrastructure and Water Management and Dutch Water Boards, who using drones with infrared cameras and AI can quickly and carefully inspect dikes. The application of a drone in combination with AI is so effective because it provides quick insight and the data is immediately usable. The drone can also reach locations that are difficult to reach by foot.
Sensors and IOT devices
Many activities and jobs will have to deal with the deployment of AI solutions in combination with sensors and IOT devices in a radical way. Gartner thinks that we will already have 26 billion devices connected by 2020. Let alone when we also add more and more sensors. A wealth of information and at the same time a necessity to extract the actual information using AI.
6. The intelligent assistant
Natural Language Processing (NLP), AI and BI will become increasingly important in the home and at our workplace. Several ‘voice assistance’ solutions are already on the market. Apple has Siri, Amazon has Alexa, Microsoft has Cortana and Google works with Google Assistant. Often solutions based on the ingredients: voice recognition, interpreting your message with the help of NLP, comparing your request with existing requests and solutions and finally giving back results in spoken sentences. We are becoming more and more familiar with it through smartphones. There are now even solutions such as a robot that receives people in a town hall.
Answers lead to new searches
But as soon as answers lead to many new searches, there is often the need to search manually. In addition, it remains to be seen whether a manager will actually start to speak to his computer or device in the workplace in the presence of others because it could concern confidential or sensitive information. At the same time, the possibilities are unlimited when we think of the combination of AI and BI in particular. Think of a spoken message when, for example, a certain threshold value has been hit. Or requesting scenario calculations or asking how much time is left until the threshold value is reached.
Where smart NLP solutions are considered to be a full-fledged team member, the adaptation will go faster. Man and machine will only learn from and with each other only when it is possible to learn in all openness and without embarrassment. A learning machine requires a learning person and learning teams
7. Cloud and BI are not yet a happy marriage
Many BI solutions are nowadays very usable from the cloud. Many success stories about cloud computing are available such as Salesforce.com for CRM, Amazon.com and Google Apps. As business analytics tools from the cloud, Microsoft PowerBI and Tableau can be seen more often. Passionned Group has a lot of experience with supporting organizations in the choice for the best BI tools.
BI cloud solutions grow with difficulty
But we often see that BI cloud solutions can not yet grow sufficiently together with the maturity levels of organizations. In the start-up phase many solutions are fine, but as soon as more is required, problems arise. The limitations in the management and use of Cloud BI solutions are decreasing, but organizations do not seem to be lined up to entrust their mature BI system to the cloud. Certainly when dozens or hundreds of internal and external data sources have to be combined, the perception is apparently that the cloud is not yet the right place.
User-friendly BI tools
User-friendly BI tools are often used in the cloud if they cover a certain area (for example HR or Marketing), if quick and easy data analyzes must be generated or if the IT organization itself can not deliver fast enough. Often due to a lack of agility within the internal BI organization or BI architecture. In short, many organizations will still opt for in-house solutions, or in combination with the cloud (hybrid).
8. Realtime Augmented intelligence
Augmented reality, location intelligence and mobile BI: this combination is very interesting. All business activities where location is important will have to deal with the first steps towards integrating this information and the corresponding data analyzes into daily use. In the powerful combination of the AI, location intelligence and BI, data analyzes will be made available at a certain location based on what is observed – via a smartphone, glasses, drone or contact lens.
Location data analysis
For example, data about the progress of a building project that is being observed, current turnover figures at the location of a company, or the performance and maintenance of a machine in a production hall. But it can also be data for the diagnosis, monitoring and treatment of clinical pictures. Unprecedented opportunities that save time and stimulate real-time continuous improvement.