For a project to be successful, there are many different steps involved. Working on projects that use big data for decision-making is not different, but since these technologies are still fairly new, decision-makers, organisations, and project managers might not have first-hand experience implementing these types of projects.
Positium’s project managers have compiled five steps they consider crucial to a successful project using mobile positioning data (MPD), a type of big data with use cases in many sectors such as tourism, urban planning, public transportation and traffic planning, crisis management, population statistics, just to mention a few.
Positium has implemented close to 300 MPD projects in over 19 years, in countries like Oman, Indonesia, Lithuania, Estonia, Ghana, Brazil, and many others.
Wether you are a decision-maker trying to assess the cost-effectiveness of a new infrastructure project targeted at improving the experience of tourists, perhaps a new highway segment, or plan and redesign the public transport network of your city based on insights from big data, there has never been a better time to turn to MPD.
Successful use cases in different domains, a mature developed methodology and tools for processing the data, and quick access to it, guarantee a successful project. Keep reading to learn how to implement your next big data project with MPD.
Step One: What Do I Need?
First things first – the need to look for novel solutions comes from necessity, and therefore the best place to start is by analysing the current situation.
You should start by mapping what methods are currently used in your organisation to gather data to produce statistical insights. For tourism, this is usually administrative data coming from border checkpoints, border surveys, and data on nights spent in accommodation establishments, such as hotels. In urban planning or public transportation it could be surveys, traffic cameras, or ride tickets, among many others.
Once you have finished the analysis, the best way to gather insights into what solutions can complement, or even substitute, traditional data sources involving time-consuming practices, is by issuing a Request for Information (RFI). Researching the possibilities out there with the help of an RFI will help your organisation to get an accurate assessment of the marketplace or hold direct consultations with internationally renowned experts.
Although traditional data sources have major flaws in terms of representativity (sample size), bias, or other quality aspects, combining these sources with MPD adds a substantial and dynamic layer of information to get the most accurate results.
MPD is considered by far the best data source for understanding where people actually move, because mobile devices are almost always carried along, including when visiting a foreign country. You might forget your watch, credit card or even laptop, but never your phone.
To learn more about MPD, read the article ‘Mobile Positioning Data FAQ: MPD Basics’
Statistical insights are needed by many stakeholders to make decisions. In the tourism sector for instance, it is needed by tourism ministries to evaluate and plan investments, national statistics offices, or ministries of the economy to get insights on the expenditure and impact on the local economy, and local businesses to understand their potential and adjust their marketing campaigns.
After you have identified the need and gathered the stakeholders, the next phase is to assess the suitability of the indicators that MPD can provide you, and agree on a budget and the overall process of the MPD project.
To get MPD from mobile network operators (MNOs), there needs to be a clear understanding of the following: what conditions, and in what form the MNO can provide the mobile big data; if the MNO can provide quality raw data; if the local legal requirements allow MNOs to share the data; and who can provide the statistical calculations.
Step Two: Good Input Equals Good Output
After a successful RFI or consultations, a Request for Proposal or public procurement is announced to start the process of looking for a solution provider that can implement the MPD project.
The first phase of an MPD project is ensuring data access. This step can be quite time-consuming as, depending on the project scope, it includes negotiations and agreements with several different stakeholders (MNO(s), regulators, national statistics offices, ministries, local governments or similar). But as this step ensures the project quality and data access for the next steps, it is important to take the time needed to get the best possible solution.
Data access is one of the key challenges for government organisations. Mobile phone data is governed by privacy laws, telecommunication legislation, and MNOs’ internal rules. Ensuring access to the data is a time-consuming process. Hence, it is better to start as early as possible. Positium can manage this process based on the experience gained over more than 19 years in this field.
Here are four assessments that you need to conduct in this step:
- The assessment of the legal situation must clarify who can access the data and under what conditions they can do it.
- The assessment of the business model gives an understanding of the roles that different project parties will have.
- The assessment of the technical situation clears all the questions about data storage, hardware and software.
- The assessment of data quality will give a final approval that it is possible to utilise the raw mobile data provided by the mobile network operator(s) for statistical insights.
Step Three: Pilot Project
After the four assessments are completed, it’s time to move to the pilot project phase. In this step, MPD must pass through the next quality assessment process.
MNO data needs to be checked and improved if necessary, to make sure the input data meets the requirements of the pilot project. MNOs store the data in different forms, which means that this step varies in complexity.
As all countries have a unique context, Positium’s methodology for turning MPD data into statistical insights is adjusted accordingly. This stage involves an in-depth analysis of the possible indicators, creating test scenarios, running statistics calculations with several scenarios, methodology development, and analysing the results of these together with the stakeholders.
To learn more about our methodology for tourism, read our article on ‘The science behind producing tourism statistics from MPD’
Next, it is time to analyse the data received from the MNO(s) for the initial proof of concept. Usually, data for 6–12 months is needed to calculate key indicators. For instance, in the case of tourism statistics, indicators such as country of residence and trip length will be needed from the data, and to be able to calculare these, data for at least 6 months is needed.
Then comes the most interesting part – diving into the data to find patterns, detect changes and compare the result to existing reference data. This process is crucial to determine the most valuable insights for the specific country, and therefore it is important to include all stakeholders in this step.
After the calculations are done, there are several ways to get insights from the data. Looking at results tables is one option, but they can sometimes be too difficult to understand for people without prior data analytics experience. For our clients to best understand the results they are looking at, a report about the methodology and results is compiled. This has proven to be a very valuable tool in Positium’s previous projects.
To communicate trends, tourist flows and insights to a wider audience, an interactive dashboard or maps, animations and other data visualisation tools can be used. We aim to make the data easily understandable to all stakeholders.
Positium has developed interactive maps and dashboards for different national and local governments, public organisations, and tourism boards. Lithuania Travel, for instance, recently launched an interactive map that can be accessed via desktop web browsers. Another example of a similar dashboard developed by Positium was the one Oman’s National Center for Statistics and Information (NCSI) launched earlier this year. The dashboard can be accessed here.
Another option is to export the data tables to Business Intelligence software (Excel or Power BI) or GIS tools that your organisation is already using.Watch this video on how Lithuania Travel built a tourist map based on MPD
Step Four: Continuity Brings Success
Now the hard part is done – the method has been validated, and the pilot project was a success. To ensure that the most value can be retrieved from the data, it is wise not to let it be a one-time thing, but to ensure the regular production of MPD-based statistical indicators.
Data is only valuable when it is made available and used. The continued production of statistical indicators focuses on providing continuous data updates (e.g. monthly, quarterly statistics). It will include all steps related to data collection, processing, and calculation of the indicators to make sure that the input data provided by the MNO(s), methodological parameters and statistical indicators always have the agreed parameters and get updated promptly.
Currently, there are two countries in the world using MPD as their main source for official statistics in tourism regularly – Estonia and Indonesia. There are many more who use it for statistical insights and decision-making, such as the already mentioned examples of Oman and Lithuania.Learn more about how the Central Bank of Estonia has been using MPD for 14 year, read the case study: ‘Estonia Leads the Production of Tourism Statistics Using Mobile Positioning Data’
Step Five: Data Is Only Useful When It Is Used
The roadmap to using MPD for statistical insights varies from country to country and depending on the political and legal climate, it can take anywhere from a few months to a few years. At first, this might seem like a long time, but considering the alternative cost in time and money involved when using traditional methods, it is a worthwhile investment.
Now that you have a better picture of the steps needed for a successful MPD project, you can start the journey of using MPD to make the investment decisions your city or country needs. Data is most valuable when it is made available and used for the good of society.
If you want to learn more about how national and local governments around the world are using MPD to produce official tourism statistics, conduct mobility studies, create smart city solutions, and more: