Take a mental stock of your pantry and the items you use to cook your meals. Chances are, you’re using a small fraction of all your ingredients and kitchen tools to whip up a limited range of dishes. A company’s relationship with data is a lot like that.
When Seagate and IDC surveyed executives and IT leaders of medium to large-sized companies across 12 countries, they found that enterprises use only one-third of the data they have. The rest of the data languishes in a data lake or warehouse, like a condiment sitting ignored and unused in the dark corner of a kitchen cabinet. The data could potentially give you insights that open up vast opportunities, such as boosting revenue and winning back customers who are about to churn. But data is only valuable if you actually see and use it.
To solve the challenge of activation, developers and data engineers have built modern data stacks that make data more accessible to business users. For instance, customer data platforms (CDPs) remove silos by unifying customer data from multiple sources. But having a single, centralized source of truth for your data is only half the solution. You need to be able to activate your data in the business tools you use every day. And that’s where reverse ETL comes in.
What is reverse ETL?
Reverse ETL is the process of copying data from a warehouse into business applications like CRM, analytics, and marketing automation software. You perform this process by using a reverse ETL tool that integrates with your data source and your business SaaS tools. Some CDPs can also activate data in those apps. (We’ll talk about the similarities and differences between reverse ETL tools and CDPs later.)
The term “reverse ETL” comes from “extract, transform, and load,” which is the process of taking data from a source, cleaning and structuring it, and sending it to a data warehouse or data lake. If you load data before transforming it, you’re performing ELT.
In short, ETL/ELT and reverse ETL sit on opposite sides of a data pipeline. One fulfills data integration, the other, data activation.
Why businesses need reverse ETL
A reverse ETL tool is a powerful component of the modern data stack. All business teams and departments stand to benefit from implementing reverse ETL.
Operationalize your data
“Operationalization” doesn’t just mean using something. In scientific research, it means making an abstract concept concretely measurable.
In the same way, data that sits in a warehouse has a vague potential to contribute value to your business. But when you use it in business apps, you turn it into a concrete, measurable component of activities like marketing campaigns or product development. Compare the results you achieve when you design data-driven campaigns versus those based on limited customer information.
All departments in a company can operationalize data. For example:
Finance creates a custom payment plan for B2B customers who have been lagging behind in payments and sends automated follow-up emails using an invoice and accounting software like Xero.
Customer support prioritizes the queries of customers marked as “VIP” on a ticketing system like Zendesk.
Marketing runs a re-engagement campaign using an email tool like Mailchimp whenever a customer abandons a shopping cart and fails to complete checkout within 24 hours.
Prevent data silos
Reverse ETL tools let you access data across different departments. For example, sales reps are not confined to sales data. You’re limited only by the restrictions your company sets for privacy and security reasons.
As a result, data silos break down, and you no longer have to keep begging another team or a data analyst to create a list or report for you. You can load the relevant data into the app you’re using. For example:
Product gets a list of high-value customers and gives them early access to a new feature you’re rolling out on your SaaS app. The data is finance-related but isn’t siloed within that department.
Sales invites people who downloaded a case study to view a product demo. No need to ask the marketing team to send them details whenever someone downloads it.
Accounting discovers that some customers with late payments also complained about your product in the last 30 days but didn’t get their issues resolved. The accounting staff works together with customer support to resolve the problems before following up again on the payment.
Easily integrate and scale your analytics
Integrate a reverse ETL tool with operational analytics software to get concrete answers to questions like:
What is the best predictor of customer churn?
What traits and behavioral patterns do customers with the highest lifetime value have in common?
How does the user onboarding experience affect customer loyalty?
What does the customer journey of a certain audience segment look like?
Which communication channel has higher engagement?
Does our product recommendation algorithm lead to larger basket sizes?
To answer questions like these, you need data from multiple channels and departments. And as your business grows, you want to continue asking and answering data-driven questions without having to set up new analytics workflows. Reverse ETL tools make that possible as you only need to integrate them once with your business intelligence and analytics software.
Give data teams more time to focus on higher-value work
Once you connect a reverse ETL tool to your data warehouse, you reduce the need for data analysts to manually extract and prepare data. Analysts face many requests like these every day, which means they can spend hours performing a relatively simple task over and over again.
Implementing a reverse ETL tool saves data teams a lot of time and lets them focus on solving more complex data problems, such as maintaining a high quality of data, implementing security and privacy practices, and identifying the most useful metrics and information for your business goals and problems.
Where reverse ETL fits into your data infrastructure
Reverse ETL tools are nascent, but they’re adding more software integrations by the day. That means you won’t have to overhaul your current data infrastructure and business tech stack to implement reverse ETL.
Reverse ETL and data warehouses
Reverse ETL tools connect to widely used cloud data warehouses like Snowflake, Google BigQuery, and Redshift. They can also use spreadsheets as sources.
Once you’ve linked to your source, choose the app where you want to activate and sync specific data tables or data sets. If you can’t find one of your business apps on the list, send the reverse ETL tool provider a request for integration.
You take the above steps within the interface of the reverse ETL tool. It’s typically a point-and-click process that doesn’t require you to write SQL.
Reverse ETL and CDPs
Some CDPs have a function similar to reverse ETL. Segment, for instance, lets you send data directly from the CDP database to marketing, sales, product, eCommerce, and customer support tools, like A/B testing, behavioral analytics, social media ads, CRM, and more.
But a reverse ETL tool and a CDP can also work together. The CDP collects data from multiple sources, cleans it, unifies it to create customer profiles, and then feeds the transformed data into warehouses. The reverse ETL tool then takes data from warehouses to business apps.
The value of combining a reverse ETL tool and a CDP lies in having enriched customer profiles and customer segments from the CDP. CDPs perform identity resolution, which means they can identify the same customer across different interactions.
For example, a CDP might find that the name and email address of a new customer match those of a person who viewed a product demo months ago. Or that the Instagram user who liked your post about your anniversary sale also bought products from you during that sale. The CDP stitches these disparate pieces of information together, creating a profile of a customer’s interactions with your business over time and across multiple channels. The CDP also lets you create customer segments based on customer profiles.
3 best reverse ETL tools
Airtable, Asana, Notion (productivity and collaboration)
Amazon, Facebook, Google, LinkedIn, TikTok (ad platforms)
Appcues (user onboarding)
dbt (data transformation)
HubSpot, Intercom, Marketo, Mailchimp (marketing automation)
Freshdesk, Zendesk (support)
Both Hightouch and Census are user-friendly for non-technical staff, like marketing and customer support teams. They charge based on the number of destinations and/or destination fields, access control features, and service-level agreement terms.
For developer-driven projects and data-mature companies, Grouparoo, an open-source reverse ETL framework, is a good choice as it gives you greater control over data modeling and user segmentation. As of this writing, it has much fewer integrations than either Hightouch or Census.