Resources · Make integration

BigML Make integrationAutomate BigML with Make.

Can your BigML models score new data without anyone opening the dashboard? The BigML Make integration gives you 23 modules: one scheduled trigger, 18 actions and 4 searches. This guide connects your account and walks you through a first working scenario.

Try the BigML integration on MakeSee the 23 modules

Verified Trustpilot reviews · AI, automation & growth agency

What you can automate

What does the BigML Make integration let you automate?

The BigML Make integration links your BigML account to Make so predictions, anomaly scores, centroids and topic distributions can be created, read, updated and deleted inside a scenario, the automation you build in Make as a chain of modules. You pick the bricks, Make runs them in order.

Flag odd records before they spread. Send each new row from your store or your sensors to Create an Anomaly Score, then pass only the unusual ones to the next module. Operations teams get an alert instead of a surprise.

Tag every new lead with its closest segment. Create a Centroid tells you which cluster a record resembles most, and the result can travel to your CRM in the same run.

Sort incoming text by subject. Create a Topic Distribution reads a document and returns its topic probabilities, which lets a support desk route tickets without reading them first.

Know the edges: the integration works with resources that already exist in BigML, and the documentation lists no numbered limit. Anything the 23 modules skip goes through Make an API Call, which reuses the same connection. If you are weighing tools, the n8n vs Make comparison lays out the criteria, Make pricing explains how operations are counted, and Make training helps a team build its first scenarios.

Connect

How do you connect BigML to Make?

  1. 01

    Get your API key from BigML

    Sign in to your BigML account and copy your API key. Make asks for this value when it creates the connection, which is your BigML account linked to Make once and reused by every BigML module.

  2. 02

    Add a BigML module and open the connection form

    In your scenario, add any BigML module and click Create a connection. You can give the connection a name, which helps when several accounts or teammates share the same Make organization.

  3. 03

    Paste the key and save

    Paste the value into the API Key field and click Save. Make can now read your BigML resources, and every module you add afterwards offers this connection in its list.

First scenario

Your first scenario with BigML and Make

GoalWhen a new resource is created in BigML, Make fetches its full record so the next module can use it.

  1. 01

    Create the scenario

    Open the Scenarios page, create a new scenario and click the big + to place the first module on the canvas.

  2. 02

    Add Watch Resources

    Search for BigML, choose Watch Resources and select the connection you just created. This module is a scheduled trigger: at each check, if it finds a new resource, it starts the scenario with it.

  3. 03

    Chain Get a Resource

    Click the + on the right of the trigger, pick Get a Resource and map the resource ID coming from Watch Resources. Each new item arrives as a bundle, one element passed from module to module.

  4. 04

    Test with Run once

    Create a resource in BigML, then click Run once. Open the bubble above each module to read the bundles it received and check that the details look right.

  5. 05

    Schedule and switch it on

    Click the clock on the first module to set the schedule, every 15 minutes by default, then turn the scenario on. From now on Make checks BigML on that rhythm without you.

Triggers

What starts a BigML scenario?

1 module starts a scenario when something happens in BigML. A scheduled trigger is checked by Make on the scenario's schedule; an instant one is called by BigML as soon as it happens.

BigML1Watch Resources

Watch Resources

Scheduled trigger

In Make“Triggers when a new resource is created.”

Watch Resources notices every resource newly created in your BigML account and starts the scenario with it, so the next modules can react without you checking the dashboard.

How it fires
scheduled, Make checks the app on the scenario's schedule. The default is every 15 minutes, and the Free plan cannot go below that; paid plans go down to 1 minute.
When to use it
whenever a teammate or another scenario creates predictions or scores and you want them logged, shared or enriched right after.
Watch out
if the bundle from the trigger lacks what your next step needs, add Get a Resource to fetch the resource by its ID.
Modules

What can Make do in BigML?

BigML gives you 22 modules in 6 groups. For each one: what it does for you, when to reach for it, and what to watch out for.

Anomaly Scores

5 modules
BigML1

List Anomaly Scores

Search

In Make“Lists all anomaly scores.”

Brings back all the anomaly scores stored in your account, each one as its own bundle. An anomaly score measures how unusual a data instance looks compared with the rest.

When to use it
you want a table of recent scores in a spreadsheet so an analyst can review them in one place.
Watch out
this is a search, so every module after it runs once per score and counts that many operations.
BigML2

Get an Anomaly Score

Action

In Make“Gets a specified anomaly score.”

Fetches a single anomaly score by its ID and hands its content to the next module. It is the precise tool, where the list module is the broad sweep.

When to use it
an earlier step already gave you the ID and you only need that one result to decide what happens next.
Watch out
map the ID from a previous module instead of typing it, or the scenario keeps reading the same score.
BigML3

Create an Anomaly Score

Action

In Make“Calculates the anomaly score of a data instance.”

Sends one data instance to BigML and returns its anomaly score, so a record that looks strange gets flagged while it is still moving through the scenario.

When to use it
a new order or a new sensor reading arrives from another app and you want to know whether it stands out before it lands in your reporting.
Watch out
every record you score is one operation on your Make plan.
BigML4

Update an Anomaly Score

Action

In Make“Updates an existing anomaly score.”

Edits an anomaly score that already exists rather than producing a second copy, which keeps your BigML account easy to search.

When to use it
after a human review, you want the stored score to carry the outcome your team agreed on.
Watch out
Make needs the score's ID; grab it first with List Anomaly Scores or Get an Anomaly Score.
BigML5

Delete an Anomaly Score

Action

In Make“Deletes a specified anomaly score.”

Removes an anomaly score from BigML for good, which is useful once the result has been copied to the place your team actually reads.

When to use it
as the last step of a scenario that archives scores elsewhere and wants the account left clean.
Watch out
a deletion cannot be undone, so run your first test with Run once on a score you can afford to lose.

Centroids

5 modules
BigML6

List Centroids

Search

In Make“Lists all centroids.”

Returns every centroid in your account as separate bundles. In BigML, a centroid is the answer to one question: which cluster is this data instance closest to?

When to use it
you want to export segment assignments in bulk to a CRM or a shared sheet.
Watch out
a long list means many bundles, and each following module spends one operation per bundle.
BigML7

Get a Centroid

Action

In Make“Gets a specified centroid.”

Opens one specific centroid and passes its details along, handy when an earlier module handed you only its ID.

When to use it
a customer record in another tool references a centroid, and you need the cluster behind it to route that customer to the right campaign.
Watch out
if the ID comes from an old record, the centroid may already have been deleted, so test with a fresh one.
BigML8

Create a Centroid

Action

In Make“Finds out the closest cluster to your data instance.”

Finds the cluster a new data instance resembles most and gives the answer back to Make as a bundle, ready to be written into another app.

When to use it
someone signs up through a form and you want to tag the contact with the segment it matches before your sales team sees it.
Watch out
each contact you classify counts as one operation, so filter out the contacts that do not need a segment before this module.
BigML9

Update a Centroid

Action

In Make“Updates an existing centroid.”

Modifies a centroid already stored in BigML so the record stays aligned with what happened downstream, without creating a duplicate.

When to use it
a salesperson confirms or corrects a segment and you want BigML to keep track of that decision.
Watch out
the module works on one centroid at a time, identified by its ID, so pair it with List Centroids for several.
BigML10

Delete a Centroid

Action

In Make“Deletes a specified centroid.”

Erases one centroid from your account permanently, a simple way to stop old classifications from piling up.

When to use it
once a contact has left your CRM, and the cluster result attached to it no longer serves any report.
Watch out
there is no recycle bin; check the ID in a Run once test before you switch the scenario on.

Topic Distributions

5 modules
BigML11

List Topic Distributions

Search

In Make“Lists all topic distributions.”

Collects all the topic distributions of your account, one bundle each, so you can see how your documents were classified by subject.

When to use it
you want an overview in a spreadsheet of which subjects your support tickets or articles fall under, fed from what BigML already computed.
Watch out
each bundle triggers one run of the modules that follow, which adds up in operations.
BigML12

Get a Topic Distribution

Action

In Make“Gets a specified topic distribution.”

Retrieves one topic distribution by ID, with its topic probabilities, and passes it to the rest of the scenario.

When to use it
a ticket stored in your help desk keeps the ID of its distribution, and you want to reread the subjects before escalating it.
Watch out
map the ID dynamically; a typed value makes the scenario fetch the same document every time.
BigML13

Create a Topic Distribution

Action

In Make“Calculates all topic probabilities for a given document.”

Reads a document and returns the probability of each topic, turning a block of text into something a scenario can sort on.

When to use it
a support email or a customer review lands in another app and you want it routed to the right person according to its subject.
Watch out
none of the 23 modules builds a topic model, so the one you score against has to exist in BigML first.
BigML14

Update a Topic Distribution

Action

In Make“Updates an existing topic distribution.”

Changes an existing topic distribution in place, so the stored version reflects later information without a second copy.

When to use it
an agent reclassifies a ticket by hand and you want that correction recorded next to the original analysis in BigML.
Watch out
you need the distribution's ID; List Topic Distributions or Get a Topic Distribution gives it to you first.
BigML15

Delete a Topic Distribution

Action

In Make“Deletes a specified topic distribution.”

Removes a topic distribution from BigML for good, keeping your account limited to the documents you still care about.

When to use it
when a ticket is closed and archived in your help desk, and its analysis has no reason to stay in BigML.
Watch out
nothing brings a deleted distribution back, so keep a copy elsewhere if a report may need it.

Predictions

5 modules
BigML16

List Predictions

Search

In Make“Lists all predictions.”

Pulls every prediction stored in your BigML account into the scenario as individual bundles, ready to be filtered, counted or exported.

When to use it
finance or sales wants the predictions gathered in one sheet to compare them with what actually happened.
Watch out
a search can return zero, one or many bundles; plan the next modules for all three cases.
BigML17

Get a Prediction

Action

In Make“Gets a specified prediction.”

Looks up one prediction by its ID and delivers its result to the next module, without scanning the whole list.

When to use it
a deal in your CRM keeps the ID of the prediction made for it, and you want to show that result in a message to the account owner.
Watch out
an ID that points to a deleted prediction returns nothing useful, so test with a recent one.
BigML18

Create a Prediction

Action

In Make“Creates a prediction using a model, logistic regression, deepnets etc.”

Asks one of your BigML models (a model, a logistic regression, a deepnet and more) for a prediction on new data, and returns the answer as a bundle.

When to use it
a quote request comes in and you want a predicted outcome attached to it before the team picks it up.
Watch out
the model must already exist in BigML, and every prediction counts as one operation in Make.
BigML19

Update a Prediction

Action

In Make“Updates an existing prediction.”

Adjusts a prediction already saved in BigML, useful to keep one record per case instead of a trail of near duplicates.

When to use it
the real outcome of a deal is known and you want the prediction record to note it for later comparison.
Watch out
the module edits one prediction per bundle, identified by its ID.
BigML20

Delete a Prediction

Action

In Make“Deletes a specified prediction.”

Permanently deletes one prediction from your BigML account, a quick way to clear results that no longer serve any analysis.

When to use it
at the end of a scenario that has already copied the prediction to your CRM or your data warehouse.
Watch out
Make cannot restore it; try the scenario with Run once on a test prediction before activating it.

Resources

1 module
BigML21

Get a Resource

Action

In Make“Gets a resource.”

Fetches any BigML resource from its ID, which makes it the natural partner of Watch Resources in a scenario.

When to use it
the trigger told you something new exists and you now need its complete content to send a summary to your team.
Watch out
the ID has to be mapped from the trigger or another module, not typed by hand.

Other

1 module
BigML22

Make an API Call

Action

In Make“Performs an arbitrary authorized API call.”

Opens the rest of the BigML API to your scenario: you reach any part of the BigML API that the listed modules do not cover, with the connection you already set up.

When to use it
you need an operation on a kind of resource that has no dedicated module among the 23.
Watch out
you write the request yourself, so keep BigML's API documentation open while you fill the module.
Need help

Need help automating BigML with Make?

A person reads every message.

FAQ

Questions about the BigML Make integration

01Is the BigML Make integration free?
Yes. BigML is a standard app in Make, so it works from the Free plan. That plan allows 2 active scenarios, a minimum of 15 minutes between two scheduled runs, 5 minutes of execution per run, files up to 5 MB and 512 MB of data transfer. Paid plans (Core, Pro, Teams and Enterprise) lower the interval to 1 minute, remove the cap on active scenarios and count usage in operations per month. Your BigML subscription is a separate matter that this page does not cover.
02What do you need to connect BigML to Make?
A BigML account and its API key. In Make, add any BigML module, click Create a connection, optionally name it, paste the key into the API Key field and save. The connection is then reused by every BigML module of your scenarios, so you only do this once per account. If the key changes, edit that connection rather than adding a new one, or older modules will keep using the outdated value.
03Does the BigML trigger in Make react in real time?
No. Watch Resources is a scheduled trigger: Make checks BigML on the scenario's schedule rather than being notified by BigML. A new scenario runs every 15 minutes by default, which is also the minimum on the Free plan, while paid plans can go down to 1 minute. If that delay matters, the plan and the schedule setting are what to change, not the module. Set the clock on the first module before switching the scenario on.
04What if a BigML module is missing in Make?
Use Make an API Call. It sends any request to the BigML API with the connection you already created, so a resource type or an operation without a dedicated module is still within reach. The BigML documentation in Make lists no numbered limit for these modules. The tradeoff is that you build the request yourself, so keep BigML's API reference at hand and test the module with Run once before switching the scenario on.
05Should you use Make or n8n for BigML?
It depends on what your team already runs. On Make, BigML comes with 23 ready modules covering predictions, anomaly scores, centroids and topic distributions, one scheduled trigger and Make an API Call for the rest, all available from the Free plan. The same app often has an n8n node too, so check its coverage before choosing. Compare the modules each tool offers for your exact use case, then pick the one where your other apps are already connected.
Same platform

Other docs you might need on this stack.

Browse all integrations