Custom Connectors 101

Learn about Funnel's Custom Connectors: custom-made integrations from your unique data source into Funnel.

Zarah Fager avatar
Written by Zarah Fager
Updated over a week ago

Summary


Introduction

In addition to Funnel's extensive library of Core Connectors, which are a core part of the Funnel platform, users can also leverage Custom Connectors to, in many cases, import data from any platforms or systems not already supported out-of-the-box.

This allows you to connect your unique data sources to Funnel through a Custom Connector - built by us, exclusively for you.
For more information on how to work with our Integration Engineers, go here.

Custom Connectors are available for the following plans: Business and Enterprise. See our Pricing page for more details.


Benefits of a Custom Connector

  • Connect to virtually any marketing app or platform with a connector built exclusively for you

  • Fully managed connectors that deliver the same high reliability as Core Connectors


Custom Connector Request Process

How do I make a Custom Connector request?

Start by checking out this article How do I request new data in Funnel?

Please include as much information with your request as possible! The more we know, the better we can help you. It’s always helpful to include an example report with your Custom connector request. This is true even for cases where the Custom Connector already has been built.

IMPORTANT: In cases where you are sending the data to Funnel, making changes to the report being sent before the data source is reconfigured may cause the data source to break. Once the data source breaks, an Integration Engineer will need to resolve the issue before future files can be ingested. For some common troubleshooting tips, go here.


Edit Configuration/Schema Change

When a Custom Connector type needs to be reconfigured to ingest a report with new or different fields, the Integration Engineers will need to know the following:

  • What changes need to be applied?

    Attaching an example of the updated report(s) that will be sent is the easiest way to understand the necessary changes. If the data source is instead using a query to get the data, then please provide the updated query instead of/in addition to the report.

  • Which source(s) do these changes need to be applied to?

    An easy way to identify a source is by its Data Source ID. You can find this information as a Dimension in the Funnel Data Explorer, underneath the Data Source drop-down menu.


Additional Data Sources of Existing Custom Connector Source Type

If additional sources of an existing custom source type are needed, then a Custom Connector Request is necessary to start the process. The Integration Engineers will need to know the following information:

  • What existing source type do you need additional sources of?

    The easiest way to identify a source type is by its Data Source Type. This can be found as a Dimension in the Funnel Data Explorer, underneath the Data Source drop-down menu.

  • The number of sources required; in cases where it’s more than 1.

  • In which Funnel workspace the new data source(s) should be connected.
    NOTE: If unspecified, by default the Integration Engineers will connect any new sources in the same Funnel workspace you submitted the Custom Connector Request from.

For more information on the next steps and the different Custom Connector request stages, go here


How can I make my Custom Connector request as fast and smooth as possible?

  • Provide as much detail as possible in the initial Custom Connector request that you submit.

  • Have a list of your data requirements (e.g. tables, fields, metrics, measures, columns).

  • Have a list of the potential methods to extract the data from the desired platform. If you don’t know what methods the platform offers, reach out to them.

  • Provide us with example reports of the data requested.

  • Provide documentation for the platform as it relates to the transport method (e.g. API documentation).

  • Be prepared to share platform credentials with us - but DON’T submit them in an email.(To protect you and your security, we’ll ask you to rotate the credentials if you email them).


Potential limitations

  • While we have a very high win rate for Custom Connector builds against an ever growing number of platforms, we can’t guarantee we can build everything!

  • Some platforms limit the data provided in exports versus what you can see inside the platform.

  • Data volumes need to be within Funnel's Fair usage policy

  • We’re bound by platform Terms of Services and so sometimes cannot offer certain types of data.


Using Custom Connectors & Data Sources

Custom Connectors are used to import data from platforms and applications not yet supported in Funnel via a Core Connector. They can also, in some advanced use cases, be built to complement data being imported from a Core Connector. Once a Custom Connector is developed, you can then have one or more Data Sources that receive data from it, much like any Core Connector.

A Custom Connector source receives data in one of two ways; "push" or "pull".

  • Push means that either the user or the platform sends data to Funnel from somewhere outside of Funnel, usually in a CSV/XLSX file. (e.g. via scheduled email reports, files pushed to an sFTP hosted by Funnel, files pushed to a Funnel owned Amazon S3 bucket).

  • Pull means that our team of Data Integration Engineers automate an extraction of data from a platform or system, typically via an API. Other options include, but are not limited to, external sFTP servers, cloud storages, SQL databases.

While both ways of getting data to Funnel are dependable, pushing data to a Data Source has some consequences that are important to keep in mind. Please refer to this article for tips and tricks on how to troubleshoot them.

If you want to import data through a method not mentioned here, get in touch with us! We are always happy to investigate new solutions to bring in data into Funnel.


Custom Connector Capabilities

While the Core Connectors are built for scalability and solving common use cases, Custom Connectors have additional capabilities. For example, you can:

  • Customise a flexible data refresh window

  • Import data that isn't primarily keyed on a date dimension (dateless data)

  • Handle unconventional file structures

Did this answer your question?