Automations

Automations allow you to schedule actions in Polinode to run automatically — for example, re-running a passive data query every Monday morning, emailing survey respondents who haven't yet completed a survey, or closing a survey at a set date and time. Instead of performing these tasks manually, you create an automation once and Polinode runs it for you on the schedule you choose.

Please note that the Automations module needs to be enabled for your organization before you will be able to use it. If you would like it enabled please contact your Account Manager or support@polinode.com. Once enabled, an Automations item will appear in the left-hand side menu. Any member of your organization can then create and manage automations, though the resources an automation works with (networks, surveys, queries and data pulls) still require the appropriate access — see Access and Permissions below.

The Automations Screen

Clicking on Automations in the left-hand side menu will show a list of all the automations for your organization. You can search for automations by name using the "Search automations" input and create a new automation by clicking on the New Automation button towards the top right of the screen. The list has the following columns:

Name: The name of the automation. If a description has been entered, hovering over the name will show it.
Action: The action the automation performs, e.g. "Run passive data query".
Schedule: A plain-English summary of when the automation runs, e.g. "Every Monday at 09:00 (Australia/Sydney)", "Once on Jul 21, 2026 09:00 (Australia/Sydney)" or "Manual only".
Last run: The status of the most recent run — Succeeded, Failed, Running or Skipped — or "Never run" if the automation hasn't run yet.
Modified: The date the automation was last modified.
Enabled: A toggle to enable or disable the automation. A disabled automation keeps its configuration but its schedule will not fire. For a once-off automation that has already run, this column shows "Completed" instead — a once-off automation cannot run a second time.

Each row also has four action buttons: Run now (runs the automation immediately, outside its schedule — disabled while the automation itself is disabled), Run history (opens the run history for the automation — see below), Edit and Delete. Deleting an automation cancels its schedule and cannot be undone.

Creating an Automation

Click on the New Automation button and a drawer will open on the right-hand side of the screen. First give your automation a name and, optionally, a description. Then select the Type of action the automation should perform. The available types are grouped by module and you will only see types for the modules that are enabled for your organization:

Combine networks (Networks)
Generate networks from survey (Surveys)
Email survey respondents (Surveys)
Open / close survey (Surveys)
Run passive data query (Passive Data)
Export passive data to S3 (Passive Data)
Run social data pull (Social Data)
Run research data pull (Research Data)

Once you have selected a type, the settings specific to that type will appear, followed by the schedule settings.

Combine Networks

Select two or more Source networks and a Target network. On every run the source networks are combined and the result replaces the contents of the target network. The target network's name, links and saved views are preserved — only its nodes and edges are replaced. You need Edit access to every network involved.

Generate Networks from Survey

Select a Survey and choose between two modes: Create new networks each run, which generates a fresh network (or networks) from the survey's responses on every run, or Update an existing network, which regenerates into the same network each time, preserving its name, links and saved views. The familiar options from generating networks manually are available as checkboxes — All relationship questions, Combine questions into one network, Exclude self selections, Exclude unselected nodes and Anonymize nodes. When updating an existing network the questions must be combined into a single network. This action is useful when you want a network to stay in sync with an open survey as responses come in.

Email Survey Respondents

Select a Survey, the respondent statuses to email (for example Not Sent and Not Started — the defaults — to chase non-responders), a From name, an email subject and an email body. The body supports the same mail merge tags as sending survey emails manually, including {firstName}, {fullName}, {email}, {url} and {optOutURL}. Emails are only sent while the survey is Collecting — once the survey is Completed, runs are skipped rather than failing, so it is safe to leave a recurring reminder automation in place until the survey closes.

Open / Close Survey

Select a Survey and the status to set it to: Collecting (open), Completed (closed) or Offline. Combined with once-off scheduling this lets you open a survey at its launch time or close it at its deadline automatically.

Run Passive Data Query

Select one of your organization's saved passive data queries. On each run the query is re-run against the latest collected data. You can choose to Create a new network each run or Update an existing network in place (preserving its name, links and saved views). Because running a passive data query re-pulls and re-aggregates your organization's data, automations of this type can run at most weekly.

Export Passive Data to S3

Exports your organization's passive data to the S3 bucket configured under Passive Data settings > Data Export. There is nothing to configure on the action itself — the destination bucket and path, the saved time period and the completion email settings all come from those settings. Each scheduled run exports interaction data incrementally (only the data added since the previous run) alongside fresh snapshots of the other files; see Scheduled exports for the full behaviour and file format. Because the incremental feed is shared across your organization, a single export automation is recommended — the editor warns if another one already exists. Automations of this type can run at most weekly. The export's own run history (under Passive Data settings) shows each run's outcome.

Select one of your saved data pulls from the Social Data module. You need Edit access to a data pull to use it in an automation, in the same way that you need Edit access to run it manually. You can then choose to Create a new network each run or Update an existing network in place (preserving its name, links and saved views). Each run starts the data pull just as if you had run it manually, including reserving Data Credits up to the pull's credit limit, and the outcome is reported through the data pull's own run history and completion email. Because every run spends Data Credits, automations of this type can run at most weekly.

Run Research Data Pull

Select one of your saved data pulls from the Research Data module. As with social data pulls, you need Edit access to a data pull to use it in an automation, and you can choose to Create a new network each run or Update an existing network in place (preserving its name, links and saved views). Access to the two modules is tracked separately, so Edit access to a Social Data pull does not authorize a Research Data pull, or vice versa. Each run starts the data pull just as if you had run it manually, including reserving Data Credits up to the pull's credit limit, and the outcome is reported through the data pull's own run history and completion email. Because every run spends Data Credits, automations of this type can run at most weekly.

Metrics from a Template View

Every action that produces a network — Combine Networks, Generate Networks from Survey, Run Passive Data Query, Run Social Data Pull and Run Research Data Pull — can also calculate metrics on what it produces. A Metrics section appears under the action's own settings with a single input, Apply metrics from a template view, which defaults to "None — do not calculate any metrics".

Select one of your template views and every metric in it will be calculated against each network that run produces. This is how you keep a scheduled network ready for analysis: rather than opening each new network and calculating Total Degree, Betweenness and Advanced Communities by hand, you save those metrics once into a template view and let the automation apply them.

Two things are worth understanding about how the results are stored:

The results are written into the network data as real attributes, exactly as if you had exported the network to Excel after calculating the metrics and reimported it. They do not need to be recalculated each time a view is opened.
No view is created on the result. Only the Calculate layers of the template view are used — any Filter, Color, Size or other layers in it, and any visual settings it carries, are ignored. The template view itself is not modified.

Metrics are calculated against the visible network, in the same way as when you calculate them yourself, so any nodes or edges hidden in the produced network's data are excluded. If a Calculate layer is hidden in the template view it is skipped. The dropdown shows how many metrics each template view will apply, so you can confirm you have picked the right one before saving.

Which Metrics Can Be Applied

An automation runs without a browser, so only metrics that Polinode can calculate on the server are eligible. This covers the great majority of them, including Total Degree, In Degree, Out Degree, Betweenness Centrality, Closeness Centrality, Eigenvector Centrality, Pagerank, Harmonic Centrality, Katz Centrality, Load Centrality, Current Flow Closeness Centrality, Average Neighbor Degree, Constraint, Effective Size, Efficiency, Brokerage, Clustering, Complex Centrality, Connected Components, Core Number, Core Periphery, Advanced Communities, Anti-Communities, K Clique Communities, HITS, Identify Influencers, External vs Internal, Informal Alignment, Informal Distance and Edge Betweenness.

The following cannot be applied by an automation:

Network-level metrics — Number of Nodes, Number of Edges, Density, Average Total Degree, Average Path Length and Diameter. These are a single value for the whole network rather than an attribute on each node, so there is nothing for the automation to write into the network data.
Communities. The basic Communities metric is calculated in the browser; use Advanced Communities instead, which produces a comparable result on the server.
Metrics that transform an attribute that is already there — Average, Percentrank, Top N Values, Binary Flag, Observed vs Expected, Rename Node Attribute and Rename Edge Attribute — along with the Connected To, Mutual and External edge metrics.

If a template view contains a layer that cannot be applied, the view is still listed in the dropdown but is greyed out with a short explanation underneath, for example "Layer 2 uses Communities — use Advanced Communities instead" or "Layer 1 (Betweenness) needs re-saving to record which edge attribute weights it". A template view with no Calculate layers at all is greyed out for that reason. See Apply Edge Weights for the re-saving case, which affects weighted metrics saved into a template view before automations could apply them.

Informal Alignment and Informal Distance are only available to organization accounts, so a template view using them can only be applied by an automation running as a user in your organization.

Where the Results Are Reported

For Combine Networks, Generate Networks from Survey and Run Passive Data Query the metric outcomes appear in the automation's own run history. Each metric is listed against the run with a tick if it was applied, a dash if it was skipped and a warning if it failed, including the reason.

The two data pull actions behave differently. As described above, these actions only start a pull, which then runs for minutes to hours after the automation has finished. The metrics are calculated at the end of that pull, so the results appear in the Social Data or Research Data module's own run history and completion email — not in the automation's run history, which only records that the pull was started. The automation editor shows a reminder of this when you select a template view for one of these actions.

If a metric fails, everything that succeeded is still written into the network and the run is marked as Failed with the failure recorded against the metric that caused it. A failed metric never leaves the network in an unreadable state.

Access and Template View Ownership

Template views sit at the user level rather than against a particular network, so the views offered are those belonging to the user the automation runs as. If you are editing an automation that runs as somebody else, the editor notes that the template views listed are theirs rather than yours.

The template view is re-checked at the start of every run as well as when you save the automation, because the two can be hours or months apart. If the template view has since been deleted, or that user has lost access to a network the metrics would be written to, the run reports that rather than writing a partial result.

Scheduling

Under When should this run? choose between Recurring and Once-off.

For a recurring automation, select a Frequency — Hourly, Daily, Weekly or Monthly — together with the relevant day and time inputs. Day of month is capped at 28 so that a monthly schedule fires every month. If you need something more specific there is also a Custom (cron) option that accepts a standard 5-field cron expression. No automation can run more often than hourly, and as noted above the passive data query, passive data export, social data pull and research data pull actions cannot run more often than weekly (for these types the Hourly and Daily options are hidden).

For a once-off automation, select the Date and Time for the single run. The time must be in the future. After a once-off automation has run it shows as Completed and will not run again.

Both kinds of schedule are interpreted in the Timezone you select, which defaults to your current timezone. Change it if the automation should run on another timezone's clock — for example, to email respondents at 9am their local time in another region.

Run History

Clicking the Run history button on an automation opens a timeline of its most recent runs (up to 50, with runs retained for 90 days). Each run shows its status — Running, Succeeded, Failed or Skipped — along with when it started, whether it was triggered by the schedule or manually, the result of the action performed and any error message. Skipped runs include the reason they were skipped, for example a survey no longer being in the Collecting status.

All automation activity — creation, edits, enabling/disabling, deletion and every run — is also recorded in your organization's Audit Log.

Access and Permissions

An automation always runs as the user who created it. Before each run Polinode re-checks that this user still has access to everything the automation touches. If the creator has left your organization or lost Edit access to a referenced network, survey or data pull, the run is skipped, the automation is automatically disabled, and an email is sent to the creator and your organization's admin users explaining why. Once access has been restored the automation can simply be re-enabled.

Runs happen in the background — after clicking Run now, check the run history a few minutes later for the outcome.