Get started as a Data Partner

See how to get started as a Data Partner and integrate your statistical data with Ducky Footprints

Onboarding process for Data Partners

The onboarding of Data Partners is a result of collaboration between Ducky and Tietoevry, a trusted partner and service supplier for Ducky. The onboarding process consists of the following steps:

1. First approach by potential data partner
2. Non-disclosure agreement and test sample
3. Contract and Data Processing Agreement
4. Data transferred to Tietoevry Solutions and linked through data hub
5. Aggregated data integrated into Ducky's systems and made available

Several steps in this process can happen in parallel, and can be amended in order to streamline the work flow and accommodate for customer needs, infrastructure, regulatory frameworks, etc.

1. First approach by potential Data Partner 

As a prospective Data Partner you reach out to Ducky through for instance support@ducky.eco, an existing customer relationship with Ducky or through our network in Tietoevry. You will be assigned a Customer Success person in Ducky as your point of contact. Together you will look into what sort of data you might contribute, what you can get out of this and assess the legal basis for data processing. Here are some resources to get started:

2. Non-disclosure agreement and test sample

Before any data or metadata is shared, Ducky will sign a non-disclosure agreement. 

Next a channel for data delivery is established. A data sample is sent over to verify that the data is useful and can be inserted into Ducky’s data pipeline. After an initial review, Tietoevry will inform Ducky on what value we might be able to extract from the raw dataset. Ducky will then decide on what data to accept and agree with Tietoevry on a format for the processed dataset.

3. Contract and Data Processing Agreement

As a Data Partner you have two options for how to proceed: You can either create and store enriched data sets within your own organization (preferably a Microsoft Azure Data Lake), or you can make use of Tietoevry's secure Data Vault and processing services in the case you wish to outsource the next steps.

A Data Partner contract and Data Processing Agreement (DPA) will then to be agreed upon. The DPA essentially states that on behalf of the private actor providing the data, Ducky and/or Tietoevry can adjust the data based on our needs. It also establishes that we are only the data processor.

4. Data transferred to Tietoevry Solutions

Once you have agreed to a method of storage and processing, a Data Lake or secure Data Vault is established for you to transfer the data into. As a Data Partner you retain ownership of your raw data in any case. Depending on the transfer method, an update period needs to be established, defining how often data will be delivered. After this, data transfers can begin.

Data sets must then be enriched with location, anonymized and aggregated to neighborhoods. You either perform these steps within your own organization, or you ask for the processing to happen on your behalf (inside of your secure Data Vault). Tietoevry can for instance query the National Population Register on behalf of a Data Partner or according to other agreements. Finally the data is anonymized and only aggregates from neighborhoods with more than 5 people will be made available for Ducky. Click below for an overview of the system architecture:

As a Data Partner you have access to your enriched data and can use it for other purposes. Please refer to your main point of contact to evaluate possible solutions like direct access in the formats supported by Tietoevry Azure services or access through Ducky Insights APIs.

5. Data integrated into Ducky's systems

At this point Ducky will integrate the processed dataset into our data processing pipeline. This will require a rewrite of scripts in the Ducky Footprints calculator (already completed during previous steps), and subsequent testing with the development team. 

Once the data has been properly integrated, a new version of the footprint calculations will need to be run and the usual error tests will need to be performed. If the resulting footprint passes these tests the map and all subsequent extra features will be updated. This data now includes actual user behavior like the energy consumption of a particular household. The data is made available through Ducky Footprints and Ducky Insights APIs.