What’s the big deal with reverse ETL? Why do you need to include this in your company’s data stack?
First, consider this scenario. You have a 360-degree view of your customers inside your central data warehouse. That’s great! But how do you use it to send an offer to your best 10,000 freemium members to go premium?
Let’s face it. Analytical charts and reports coming from the data warehouse are cool. But your sales team is using Salesforce. So, how can they act on those 10,000 members? It’s going to be a significant productivity boost if those members are copied to Salesforce. To decide it, implement a reverse ETL scenario.
In this article, you’ll learn what reverse ETL is – its meaning, use cases, and examples. So, dive in and find out how this could help your company.
Table of contents
- What Is Reverse ETL?
- Reverse ETL vs. ETL/ELT
- Reverse ETL Use Cases
- When Is Better to Use Reverse ETL?
- Reverse ETL Tools
- Reverse ETL vs. CDP
- Reverse ETL Example, Using Skyvia
- Conclusion
What Is Reverse ETL?
Data warehouses include data from various data sources. And aside from data analysis, it’s supposed to reduce, if not eliminate, data silos. But in the end, the data warehouse becomes a data silo itself. Though users know its value, it’s hard for them to act on the insights within it. It’s not in the workflow within the apps they use.
Of course, your data professionals designed it to address users’ pain points. But it’s a step close and yet far. So, if it’s not in their workflow, why not push it there? That’s what reverse ETL is for.
It is an ETL data pipeline, but of course, reversed. Instead of having the data warehouse as the target, it becomes the source. And the insights coming from it will be formatted and pushed to the target third-party apps. Figure 1 below illustrates that point.
So, instead of visiting a page or app to get insights from customers, they see it from their work apps. Since it’s now in their workflow, they can act on it. For example, they can send personalized emails to the right customers at the right time. Sounds good? Definitely!
So, here’s the definition: Reverse ETL is Extracting insights from the data warehouse, Transforming them, and Loading them to operational systems.
Reverse ETL vs. ETL/ELT
Still confused? This table will fill the gaps for you. Let’s compare it also with Extract, Load, Transform (ELT).
As you will see later, reverse ETL solutions exist like their ETL and ELT cousins. This shows that their use cases are different also. And that’s the focus of the next section.
Reverse ETL Use Cases
By using reverse ETL, you’re making your insights actionable in your users’ day-to-day work. This is better known as operational analytics. It is all automated, so it improves productivity. And eventually, it helps increase sales and profits.
And it’s not just your company but your customers benefit from the improved value of your services.
So, what are the use cases? Most of them involve operational analytics. I’ll cite some of them here and let your creative imagination do the rest for your needs. Here they are:
- Extract subscribers who downloaded a free e-book as leads into Salesforce.
- Make very personalized email campaigns based on app usage and user location for marketing teams.
- Provide customer app usage history to Zendesk for support companies.
- Get granular customer segmentation for Facebook Ads from a Snowflake data warehouse, etc.
When Is Better to Use Reverse ETL?
After reading some use cases above, you probably thought of applying them in your company. Check out the following high-level business tasks if this applies to you:
- Personalized marketing campaigns;
- Proactive customer feedback;
- Data-driven lead scoring;
- Smart ad targeting;
- Other data-driven decision-making.
Reverse ETL Tools
Next, check out the following tools you can use today and analyze which one is the most suitable for your business needs.
Skyvia
Skyvia is a universal SaaS (Software as a Service) data platform for code-free data integration, cloud backup and data management. Its Data Integration product combines ELT, ETL and reverse ETL functionality. It is a completely online solution, you need only a web browser to use it. Among Skyvia’s clients are NASA, Panasonic, Telenor, AOL, Pepsico, etc.
Skyvia offers free plans for each product. Paid plans start from as low as $15 per month. Pricing details can be found here.
PROS
- Around 80 connectors and growing;
- Free plans available;
- Free trials of paid plans available;
- Convenient UI;
- Lots of integration features;
- Failure alerts and detailed logs;
- Scheduled runs;
- No code or low code querying of data sources;
- High customer satisfaction rate on G2Crowd and Gartner.
CONS
- Limits of free usage;
- No phone support for lower pricing plans.
Census
Census is an operational analytics platform. It is also one of the early startups that serves reverse ETL as a service. Their clients include Canva, DigitalOcean, Loom, among others.
You can start for free. Payment starts at $300/month for more than 10 destination fields and 50,000 rows. Pricing details can be found here.
PROS
- Start free with 1 data warehouse and 10 destination fields;
- Free trial of paid plans available;
- Fast processing and syncing;
- Near real-time syncs;
- Sync failure alerts and detailed logs;
- Scheduled runs.
CONS
- Querying sources requires SQL knowledge;
- 1 data warehouse only for free, Growth ($300/month), and Business Plans ($800/month);
- Fewer number of sources and destinations but growing (49 at the time of writing).
Grouparoo
Grouparoo is the only open-source tool on this list. Their clients include Sendoso and Chipax, among others.
You only pay for the cloud-hosted solution. They also charge by volumes of data and it starts at $150/month for 100,000 sync records. Pricing details can be found here.
PROS
- Open-source;
- Allows local and private cloud deployment for testing and production;
- FREE forever for private cloud/community edition;
- Allows other sources aside from a data warehouse;
- With a free trial in the cloud;
- Unlimited number of sources and destinations;
- No code or low code querying of data sources;
- Scheduled runs.
CONS
- Ticket-based support only for the free Community edition;
- You pay for Enterprise UI for a user-friendly configuration. Otherwise, your option is the Command Line Interface (CLI);
- Fewer number of sources and destinations but growing (29 at the time of writing).
Hightouch
Hightouch is another SaaS solution for reverse ETL. What’s unique about Hightouch is their Audience Builder. It turns your data warehouse into a Customer Data Platform (CDP). You can check what that means in the next section. They are trusted by Kong, Springboard, and Buffer, among others.
Hightouch charges by the number of destinations. One destination is free. You pay for more and it starts at $350/month. Pricing details can be found here.
PROS
- 93 connectors and growing;
- Start free with 1 destination and an unlimited number of sync fields;
- Free trial of paid plans available;
- Near real-time syncs;
- Sync failure alerts;
- Scheduled runs;
- Live debugger;
- Audience Builder.
CONS
- Querying sources requires SQL knowledge;
- Email support only for free plan;
- Limited destinations depending on the plan.
Polytomic
Polytomic is another reverse ETL tool built for business users because of its user-friendly UI. Their customers include Vanta, ShipBob, and SourceGraph, among others.
Price starts at $500/month. Pricing details can be found here.
PROS
- Near real-time syncs;
- Sync failure alerts;
- Scheduled runs;
- No code or low code querying of data sources;
- Minimal API usage using smart job queue;
- Allow Webhooks destination for sending data to custom apps you built.
CONS
- Fewer number of sources and destinations but growing (40+ at the time of writing);
- No free trial or details of basic plan on their website.
First, let’s define what a CDP is. Customer Data Platform (CDP) is a system that collects customer data from different sources, cleans, and combines them into a single customer view. The insights also become actionable by making the combined data available to other marketing systems.
Reverse ETL vs. CDP
Sounds familiar? You’re right because that’s what reverse ETL does to data warehouses.
Both reverse ETL and CDP are big names in data trends. They also accomplish the same end goal: To make data actionable. But it’s like comparing apples and oranges. Here are their differences:
- CDP is a system but reverse ETL is a process.
- CDP’s scope is limited to customer data. Meanwhile, reverse ETL is much more diverse.
- The target users of CDPs are marketers. But reverse ETL users are data specialists and power users.
- CDP starts data collection from raw customer data. But reverse ETL starts from the data warehouse.
If you already have a data warehouse, reverse ETL will make sense for you. But if you start from nothing and you only need customer data, CDP may be a logical choice. A trial of various tools from both camps will help you decide which one better suits your company.
Reverse ETL Example Using Skyvia
As mentioned above, Skyvia is a cloud data platform that combines both ETL and Reverse ETL tools. With Skyvia, you can both automatically replicate all business data to a central data warehouse to connect this data, build analytics and make visualization, using BI tools, and return the required actionable data back to the operational system. What’s convenient is that Skyvia is a code-free solution, you don’t need to install it to use.
Let us imagine a situation when a company has already connected Salesforce CRM data with the product data by replicating and further building a data model in the Snowflake data warehouse. This can be easily implemented using the ELT functionality of Skyvia Replication product. Such centralized data can be further used for reporting and data visualization, using BI tools, but data cannot be moved back to operational systems.
However, the sales team might need to get Sending Product Qualified Leads (PQLs) and customer product usage data back to Salesforce CRM to be able to prioritize leads. It is not as easy as it might seem. But Skyvia can help you take on the Snowflake to Salesforce integration challenge. To start, first sign up to Skyvia. It is free.
Here are these three simple steps:
1. Create a connection to Snowflake and Salesforce. If you have already used them for replication, connections should be already created and available among your Skyvia objects.
2. Create an import package to migrate data from Snowflake to Salesforce:
- Select Snowflake as a source connection and Salesforce as a target one in the package;
- Add a task to the package, in which you specify the Snowflake table to be loaded to Salesforce. For example, we select the Snowflake Contact object as source, SF Lead as target;
- Columns with the same names in Snowflake and Salesforce will be mapped automatically. Map all other required source columns to target ones, using expressions, lookups, etc. and save a task.
3. Schedule your package to run automatically at a specific time.
The package settings are ready and you can start the process. Skyvia has the email notifications feature that allows you to get notified about errors in integration packages if any occur. Simply enable email notifications in your Skyvia profile, and you will be notified about everything. You can read more about Import product here.
Conclusion
So, do you want actionable insights coming from your data warehouse? I’m sure you do. This will empower different teams right in the tools they use every day. The disconnect between insights and user productivity becomes a thing of the past with reverse ETL.
Did you find this article helpful? Then please share it with your friends and followers on your favorite social media platforms.
