How Phiona Works
Last updated
Last updated
Phiona is an on-demand data lakehouse platform that can serve as an organized place for all of your different types of data files. We can ingest files that were previously local to your computer, as well as connect to different applications like Google Sheets, Airtable or Dropbox, data tables in MySQL or PostgreSQL, or data lakes like Amazon S3.
Once you have your data in Phiona, you can use it as database and/or API platform to power your business intelligence initiatives, internal applications, or external client reporting. Since we're no-code first, you don't need to know any SQL, Python, or how to configure your cloud database for best performance.
When you log into Phiona for the first time, you'll be asked to set up a project for your data.
A project is similar to a folder on your desktop, where you can bring many different types of data that may all refer to the same topic, like "2020 Marketing Ad Clicks" or "Atlanta Warehouse Daily Inventory." When you first open Phiona, you will be able to create and name your first project, and when you log in in the future, you'll be able to see all of your projects on this main homepage.
Once you've created and named a project, you'll be able to add different data sources to it. Phiona has a few different options for adding data. For more information on connecting specific data sources, click on one of the data sources in the Data Sources section.
The project page contains all of the datasets and data connections that you've chosen to include as part of the virtual data hub. Within a project, different actions can be taken to either:
Add new datasets or data connections
Manipulate the datasets themselves
Save and use previously defined manipulation steps as workflows
Schedule workflows to be run on a certain date and time
View activity of datasets added, deleted, manipulated, or exported
Phiona represents each unique data file or connection as a tile, where multiple pieces of information is included about the nature of the dataset as well as options for manipulating that data through cleaning, combining, or transformation steps.
Data Sets - All of your data sets and connections will be housed in this project section. This data can be raw data that was uploaded, data connected from cloud applications or a database, or datasets manipulated using a Phiona blueprint.
Blueprints - Blueprints are the steps that are taken to manipulate a dataset from start to finish. Think of them in this way- if your raw dataset are the ingredients and the finished dataset is your meal, the blueprint is the recipe. These recipes are attached to specific datasets so that others can see the actual steps performed to get the dataset prepared in a certain way, or so that Phiona can schedule these steps to be run in the future when the underlying data has been updated.
Workflows - This is the section where you can create recurring workflows for your datasets. If the underlying data in, for example, a Google Sheet, changes every day, a scheduled daily task can be created to refresh the Google Sheet, perform a certain blueprint that you previously created to clean or transform the data, and either keep it in Phiona, send it to another Google Sheet, or export in a specific format and send to someone else. You can also set up a webhook-based trigger, which will perform a workflow as soon as an outside application or program accesses the webhook URL.
Activity Log - This is a list of all of the tasks performed within the project for audit and documentation purposes.
Add New Data Set - Upload or connect a new dataset to the project.
Data Tiles - A dataset or data connection.
Account Settings - Clicking on this button will allow you to view your subscription and billing settings, the API keys that you've generated, manage your application authentications (Google Sheets, Dropbox, etc), view help documentation, and log out of Phiona.
Data Type - This symbol signifies where your data originates. The logo for specific databases or applications will populate here. In this case, the file originates from a Google Sheet.
Export - By clicking this link, you can export and send your file to different places.
Delete - Delete the current dataset. Please be aware that this action will only delete your dataset from Phiona, and not your local device, Google Sheet folder, or the table in your database.
Dataset Name - The current name of the dataset. This can be changed later after a workflow is created to clean, combine, or transform the data.
Date Last Modified - The date and time in which the file was last manipulated, either directly by the user or via a scheduled workflow.
Blueprints Applied - Once you've created a blueprint of actions, the updated dataset will have an attached blueprint that shows the data lineage. Click into this area to see the details of each action.
Data Cleaning - When you connect a dataset to Phiona, the application automatically runs an algorithm to determine if there are any cleaning issues that may need to be performed to standardize your data and make it better for future analysis or use. Clicking this button takes you to the data cleaning co-pilot, which automatically identifies cleaning issues for you to take action on to better prepare your data for analysis.
Transformations - When the dataset is manipulated by any transformation from the Transform options, this green text will show up to denote that this is no longer a raw dataset. Clicking this button takes you to the Transform functions, which are manipulations made on the dataset to change the overall shape or form of the data.
REST API - Clicking this link opens a modal window where you can enable this dataset to be exposed a fully RESTful API endpoint for external applications to access.
Data Preview - Clicking the tile itself will take you to a data preview page, where you can see how Phiona has interpreted your data column types, your data itself, as well as show metrics around the number of unique values and any missing values, for example.
Once you've clicked into a dataset, whether to preview it, clean it, transform it, or combine it with another dataset, you'll enter the dataset view. Phiona's dataset view looks very similar to a spreadsheet, but is meant for potentially millions of rows of data, so it is focused on actions that affect more than one cell at a time.
We group different types of primary actions into the main navigation screen (like data cleaning), but for standard data tasks, you can use the header dropdown menus that make it easier to perform certain actions like filtering data, sorting data, or deleting columns. For larger actions that require more complex user input, Phiona has a focus area to the left that shows the guided visual development tasks.
Navigate to Explorer - To see high-level information about your data source, click on the Explorer section.
Navigate to Cleaning - To see all of the cleaning recommendations for your dataset, click on the Cleaning section. This navigation button will have an indicator to show you how many potential cleaning issues were automatically identified, and Phiona will guide you through the recommendations.
Navigate to Transformations - Phiona allows for complex data transformations, like combining datasets, performing math on number columns, enriching your dataset from an API, and many others. Navigate here to see the transformation actions populate in the focus area.
Navigate to Exports - After you're done manipulating your data, you can send it to many different places, like Google Sheets, or a Postgres database.
Focus Area - This is where Phiona presents information that the user can utilize to both understand and manipulate their dataset. The Focus Area can show cleaning recommendations and options for fixing issues, as well as different transformation functions.
Data Overview - This area shows multiple pieces of information like the dataset name, number of columns, number of rows, and the number of rows Phiona is currently showing on the screen.
Undo/Redo Buttons - As you work with your dataset, you may perform an action that wasn't intended. You can undo the action at any time to see the dataset before the action was performed, or redo the action if needed. One thing to note is that Phiona's undo feature is similar to time travel- if you undo multiple actions and forge a new path, your old path is no longer available for you to take.
Blueprint Dropdown - Phiona automatically translates your data actions to code, and we list out all of these data actions in a blueprint so that when you want to finalize your changes, we can run the changes in sequence. These blueprints list out all of the specific information, as well as allows you to comment on the step to document why the change was made.
Column Header Menu - The column header will show you the column name and the type of data Phiona has identified (see available types below). Clicking on the column header on the Data Preview screen will show you key metrics about the data column. Clicking on the column header on all screens will show a dropdown menu with different options depending on the data type. Holding and moving the column header will move the position of the column to another area of the dataset.
Data Cells - This is where the data sits in the main spreadsheet view. Most of the time the sample data will be highlighted in gray and white. There are a few ways that some of the cells may be highlighted:
For certain cleaning steps, a column, row, or cell may be highlighted in red to denote a potential issue that needs to be fixed. Once a selection is made in the Focus Area to fix the issue, the change will be previewed in blue.
For certain transformation steps, as you fill in the transformation details, the column, row, or cell will highlight in blue to show you a preview of what the transformation will looks like.
