Translations life cycle management must be considered seriously, as soon as we decide to translate Power BI reports. Translating a report into several languages takes some time, when we do it for the first time. But we expect that the maintenance of translations will be as easy as possible when the report will be modified or when a new language will be added.
Some of our colleagues or customers reject reports in a language they are not used to. So, when we want to reach a larger audience, our reports need to be translated.
Even if the report has been designed in the users’ language, we might need to transform an IT format to a business format. For example, “TotalSalesAmount” -> “Total sales amount” or “Dim_Date” -> “Calendar”. This way, IT and businesspeople will be comfortable with terms they are used to. Both of them will be efficient without effort.
You may have a look at this interesting blog post from Chris Webb : https://blog.crossjoin.co.uk/2020/06/28/naming-tables-columns-and-measures-in-power-bi/
Here are the topics which need to be analyzed carefully:
Depending on the language quantity, we might need several people to help translating or validating translations. Some of those people will be business users located in different places, in different business units and not working at the same time. For these users, we need to find a comfortable way to manage translations allowing :
- copy and paste
- automatic translation
- easy identification of new labels
- easy identification of the labels needing a review after modification
- splitting translations between translators with an appropriate level of security
The most popular tool to manage data for business users is definitely Excel. An Excel file can be hosted on the network or on OneDrive and shared among multiple users. No need to train them to a new tool and support will be minimal.
- Model
- Tables name & description
- Columns name, display folder & description
- Measures name, display folder & description
- Hierarchy name & description
- Perspective name & description
- Report
- Visuals title, legend name & header tooltip
- Page name
- Textbox content
- Data to be translated outside of the database
For each of these entities, we need to detect any creation, modification and deletion to maintain translations and, eventually, to activate translators.
Initialization
- Create 3 sheets in the Excel file : Model, Visuals, Data
- Load entities to translate in these sheets. Color the cells in yellow to identify a new entity was appended
- Add 1 column by language
- Initialize translations for each language with an automatic translator. Color these cells in yellow
- Share the file with translators asking them to validate their languages. For security purposes, it is possible to split the file in as many files as translators
- Once the translations are validated, remove the yellow color
- Once there is no more yellow in the file(s), it is time to publish. Use the file(s) as a data source in Power BI
Evolutions
Each time an entity is added or modified we need to update translations. From time to time, we need to clean the file, removing entities no longer used.
- Load entities into the file(s)
- Color new or modified entities in yellow and color removed entities in grey
- Optionally remove lines with a grey color
- Share the file(s) with translators asking them to validate only yellow cells for their languages
- Once the translations are validated, remove the yellow color
- Publish
Do you feel it is hard ?
Don’t worry, in the blog posts down below, I share with you automatic tools to accomplish these tasks with minimal effort.
Report translation is very important for data culture adoption. The initialization might be a boring task without a proper methodology and tool, but the maintenance might be even worse. So, translations life cycle management must be considered properly before starting the task.
You might begin with this post : Visuals labels translations in Power BI reports
See also :
Be the first to comment