When we built Icon Map, we made a deliberate architectural decision that all client data should always remain within the customer’s Power BI tenant. This is a critical principle, as it ensures organisations remain compliant with their data residency and governance requirements.
It also simplifies the security model considerably, making it much easier for security teams to assess and approve Icon Map as a trusted solution. In addition, this approach ensures that Tekantis remains purely a software provider, without the need to process or store any client data on Tekantis-owned infrastructure.
For more detail, please refer to our security whitepaper.
Why geocoding is not built into the visual
One consequence of this design is that we do not include built-in geocoding services within the visual itself. Geocoding typically requires data to be sent to external services, which conflicts with our principle of keeping data within the client environment.
For comparison, some out-of-the-box visuals, such as Azure Maps and Esri, perform geocoding behind the scenes. This can introduce uncertainty around when data is sent externally and how it is processed.
We also believe that geocoding does not belong in the visual layer from a performance and architectural perspective. It is far more efficient to handle geocoding as part of the data processing layer, for example within a data platform using Spark, or as a one-off preparation step. Many datasets, such as locations of fire stations or retail outlets, do not change frequently, so there is little value in recalculating coordinates every time a report is refreshed.
Tools for the steps before the map
That said, we recognise that spatial data preparation can still be a barrier for many users. We have therefore introduced a set of tools within the Icon Map Customer Portal to simplify these common tasks.
Rather than forcing everything into Power BI, these tools focus on the steps that typically happen just before data reaches a report.
In line with our principle of minimising data movement, these tools are designed to perform as much processing as possible locally within the user’s browser. No data is sent to Tekantis servers. Instead, the browser interacts directly with your own Azure Maps or Google Maps instances to perform operations such as geocoding.
Converting addresses into coordinates
One of the most common challenges in spatial analysis is that data often starts as addresses or postcodes rather than latitude and longitude. Before it can be visualised, this information must be converted into coordinates.
The Customer Portal includes a geocoding tool that performs this step quickly and produces results that can be used directly in Icon Map or Power BI.
We have designed the workflow to be as simple and efficient as possible.
You can follow the step-by-step process in our documentation.
To help manage usage costs, the tool only geocodes new data. For example, if you upload a CSV file, then later add additional rows and upload it again, only the new rows will be processed. Existing records are left unchanged, helping to avoid unnecessary consumption of Azure or Google credits.
Travel time and catchment areas
Another common requirement is understanding travel distance and coverage. Organisations often need to know how far customers are willing to travel, how long deliveries might take, or what area a facility can realistically serve.
The portal includes a tool for generating travel time catchments, also known as isochrones. Users can upload a list of locations and create realistic drive-time or walking areas based on actual road networks.
See documentation for more information
This is just the start
These tools are the first step in expanding the capabilities of the Customer Portal. Over the coming months, we plan to introduce additional features such as route generation and PMTiles creation.
If there are other spatial data preparation tasks that take up significant time in your workflow, we would welcome your feedback. Please get in touch at support@icon-map.com.
You may also be interested in our related blogs covering geocoding within Power Query.