Setting Up

This tutorial will show you how to create and invite people to teams in Graphext. Teams are a fundamental part of working in Graphext and are where you collect datasets and projects. We will create two new teams, invite someone to a team and share some resources between our teams.

To start working in a team, we need a dataset to use. We will upload this to the workspace of a team where we can create projects to analyze the data inside of it. We'll be working with a dataset called 'Fictional-Employees.csv', which contains information about the performance and characteristics of a set of 100 imaginary employees. Download the dataset here.

‍

"None of us is as smart as all of us."

- Ken Blanchard

‍

Data science is a team sport. Your Graphext workspace is organised into teams to make it easier to collaborate on investigations. Datasets and projects are local to your teams. This means that any dataset you upload or project you create inside a team will only be accessible to members of that team. For instance, uploading Fictional-Employees.csv to your Personal team would mean that members of a team called Marketing wouldn't be able to see or use it.

‍

Step 1. Create a Team

First, let's create a team. Start from the Datasets panel of your Graphext workspace. Notice that on the left-hand side of the screen there is a thin sidebar containing icons alongside some text like Personal and Public. These are your teams. Your personal team is just for you. The public team is for all Graphext users to explore.

To provide a workspace for you and your colleagues to collaborate, you should create a new team. Let's call this team Ants because if any animal knows how to collaborate well, it is ants.

From the left sidebar, click the New icon to create a new team. Enter Ants as your team name and click Create Team.

That's it! You should see Ants appear in your left sidebar. Click on the Ants icon to navigate to the Ants workspace. Inside it, you should see that there aren't any projects and there aren't any datasets. Any work that you've done in other teams won't be visible from the Ants workspace.

‍

‍


Step 2. Upload a Dataset

Let's upload a dataset to the Ants team so that we can start working inside of it. If you haven't already, download the Fictional-Employees.csv dataset to your computer.

‍

Download the dataset here.

‍

To learn more about uploading datasets, see our Intro to Datasets guide.

‍

Now, from inside of the Datasets panel of the Ants workspace, click New Dataset. Upload the Fictional-Employees.csv file using the Adding Dataset window. Either browse for the file on your computer or drag and drop it into the window.

It will take a few moments for Graphext to process the dataset but you should immediately see that it has appeared inside of the Datasets panel of the Ants workspace. Navigate to the Datasets panel of your Personal team. Notice that there is no sign of the Fictional Employees dataset.

‍

‍


Step 3. Create a Project

In the same way that datasets you upload to specific teams will only be visible from inside that team, projects you create inside teams won't be visible to Graphext users outside of that team.

Let's create a simple project inside of the Ants team. Working from inside the Datasets panel of your Ants team, select the Fictional Employees dataset, bringing up the project setup wizard on the right-hand side of your screen.

‍

For a more detailed overview on creating projects, see our Intro to Projects guide.

‍

Choose Employees and then Cluster as your analysis type. Then, inside of the Select Target and Factors tab, choose Performance level from your variables list and send it to the list of targets. Add Coachability, Curiosity, Work Ethic and the additional numerical characteristics from the other variables list to your list of factors - you should have 10 factors in total. Click Next then name your project something like Ants: Fictional Employees and execute it.

Head back to the Projects panel of the Ants workspace and you should see the project being created. This might take a few minutes. Navigate to the Projects panel of your Personal team and notice that the Ants: Fictional Employees project isn't there!

‍

‍


Step 4. Invite Members

A team isn't much good if you are the only Graphext user inside it. Adding new team members to the Ants team means that these people will be able to see or edit and create datasets and projects within the Ants workspace.

‍

A Note on Team Roles

There are three kinds of roles within a team. Each role comes with a different set of permissions. When you invite a new member to a team, they are automatically given a Member role. Changing the role of a team member gives them a different set of permissions. Each team member can be given any role as there are no limits on the number of members assigned to one type of role.

Admins can add integrations, create projects and change the team settings.

Members can create projects but cannot add integrations or change the team settings.

Viewers can only view projects.

‍

Let's add a new member to the Ants team. Click on the Account and Teams Settings icon from the bottom of your team's sidebar. This brings up a window where you can manage the members of your team. Click Add member under the list of team members and enter the email address of another Graphext user.

This will automatically add the user as a Member of the Ants team. You should see that they have been added to the list of team members. Change their role to Viewer by clicking on the dropdown icon next to the text indicating that they are a Member. Select Viewer from the menu list so that team member can only view the projects inside of Ants.

Now you've invited a new person to Ants, the team will be visible from the left sidebar in their Graphext workspace.

‍

‍


Step 5. Share a Project

You can share the work you do inside of a team without having to invite new members. Projects can be shared with other Graphext users as well as other teams so that the analysis you've done on a specific project can be seen, explored or added to by others.

To share a project with another team lets first create a new team called Wolves. Do this by clicking on the new team icon from the bottom of your list of teams.

Navigate to the Projects panel of the Wolves team workspace and notice again that there are no projects here. When you share a project from the Ants team with the Wolves team, it will appear here.

Head back to the Ants team and find the Ants: Fictional Employees project that you created earlier. Click the 3 vertical dots on the right side of the project card and choose Share from the menu list. This brings up a Share settings window where you can share the project with other Graphext users or teams.

Click inside of the Share with field and scroll to the bottom of the dropdown list, where you will find a list of your teams. Select Wolves and then click Share. Close the Share settings window and head back over to the Wolves team's Projects panel.

Voila. You should see the Ants: Fictional Employees project listed as a project inside of the Wolves team. Notice that the Fictional Employees dataset isn't inside of the Wolves Datasets panel. This is because you shared the project and not the dataset. Any changes you make to the project will be visible from both the Ants and Wolves team.

‍

‍


Step 6. Remove a Team

The Wolves team is pretty pointless. It hasn't got any members apart from yourself and only contains one project, which you can control from the Ants team anyway. Let's remove it.

Before we remove the Wolves team itself we need to remove the project that we've just shared with the Wolves team. Click the three dots inside the Ants: Fictional Employee Clustering project card and choose Remove from the menu list.

Then, from inside of the Wolves team, select the icon representing Account and Teams Settings from the bottom of your left sidebar. Next to the list of team members, click on the option to Remove team then click Accept in the confirmation window that appears.

Done. Wolves no longer exists as a team inside Graphext.

‍

‍

Need Something Different?

We know that data isn't always clean and simple.
Have a look through these topics if you can't see what you are looking for.