When you're pulling data from a variety of less-popular data sources, it can be hard to find an ETL solution that works.
Meltano, one of the newest entrants into the world of ELT, has set out to change that. Meltano integrates with an existing library of connectors---meaning that in theory, it should be easy to connect lots of sources.
However, Meltano doesn't always work as planned. Many of these connectors need maintenance, and Meltano isn't the most user-friendly if you're not a technical user.
There are other ETL tools that offer the same (or even better) functionality, all with hassle-free hosting and maintenance.
Meltano started as an internal tool at GitLab. It spun off from its parent company in 2021 and has since raised $12.4 million in funding. But despite its independence, Meltano holds on to a lot of the same open-source principles as GitLab and maintains much of the transparent, developer-focused culture that's made GitLab popular.
Unlike other open-source ELT tools (Airbyte in particular), Meltano is built on the Singer protocol. It has created a team (Meltano Labs) charged with updating and maintaining existing Singer taps and targets and creating new ones. Building off Singer helped Meltano immediately offer a wide range of connectors.
But that dependence on Singer isn't all positive. Singer is notorious for underdeveloped and abandoned connectors, and Meltano can't maintain them all (and doesn't claim to). If you're using a less-common connector on Meltano, it could break without warning.
Meltano's developer-first mentality can also be a challenge if you're not a programmer. Right now, you need to host Meltano yourself. And there isn't a no-code interface. Meltano is built by developers, for developers---and not as friendly to those without technical skills.
Meltano only offers its free open-source platform at the time of writing. There are no costs for using the product, but you'll need to host it on your own. A paid version called Managed Meltano is in the works with a projected release date of 2023.
300+ connectors built on a library integrated with Singer
Open-source software using the MIT license
Public roadmap with upcoming features and bug fixes available for anyone to see
Programming best practices built-in, including Git version control, end-to-end CI testing, staging and development environments, and more
Native access to command line interface (CLI)
Comprehensive CDK for developing new connectors through Singer
Only self-hosted option currently available, so teams looking for a managed or cloud-based solution will need to look elsewhere
New software with a relatively small community (2,000+ Slack members) and no official support
Technical, code-first interface without a no-code or low-code option
Portable is a data integration tool that creates the long-tail connectors Meltano does not provide. It supports more than 250 data sources and creates custom integrations upon request, which are maintained by the Portable team.
For teams without the capacity to build difficult-to-find connections, or who don't want the ongoing maintenance requirements of Meltano and other platforms, Portable provides a simple solution to keep data tools working.
Manual: Free for unlimited data sources, destinations, and volumes
Scheduled: $200 per flow, with unlimited sources, destinations, and volumes
Custom: Tailored pricing for specific requirements
250+ data source connectors
Support for cloud-based data warehouses
Unlimited data sources, destinations, and volumes
Maintenance and development for free of new data sources
Personalized support on all plans
Portable provides 250+ prebuilt connectors for less-popular applications that the major competitors like Meltano don't support.
Portable can build as well as maintain all the data connectors you require at no cost. Development works quickly and new connectors can be available in days or hours.
Portable provides 24/7 customer support across all plans to resolve issues fast.
Transparent pricing that's not connected to monthly active rows or other volume metrics
You can try all the connectors for as long as you need with manual sync
Support for more than 250 long-tail connections you won't find with other competitors
Ongoing updates and fixes to connectors at no cost
Lacks support for the most common data sources (e.g., Salesforce or QuickBooks)
Lacks support for database connectors
Unavailable to customers based internationally
Businesses that require custom long-tail connectors, but that don't have the resources or technical knowledge to build and maintain them.
Airbyte is another open-source newcomer to data ingestion and ELT. It's seen rapid growth since its introduction in 2020. Unlike Meltano, Airbyte is not built on the Singer open source project which gives it more flexibility but also creates new limitations.
Open source: Free, since you host software yourself
Cloud: $2.50/credit (one million rows equals 6 credits; 1 GB equals 4 credits)
Cloud high volume: Pricing customized for customers who require more than 5,000 credits
170+ data connectors (not all of them available on cloud service)
Open-source solution that allows easy access to codebase
Change data capture support
Data warehouses as well as data lakes available as destinations
Credit-based pricing or the option to host the software for free
Warehouse-native data transformations
Airbyte currently has more than 170 connectors for its open source platform. Not all of them are accessible on its cloud service, but all will likely be available soon.
You can request custom connectors through the Airbyte website, however, they'll prioritize those they believe will benefit the most users. It's possible to create them yourself, but you'll need to master Airbyte protocols, configure the integration, and then maintain it by on your own.
Even though it's only a few years old, Airbyte has a robust community of more than 12,000 users. Paid plans include chat in-app support with one-hour response time.
Open-source platform that you can install on your own
Data engineer-focused platform with CDK to speed up development
Support for 20 destinations---more than many competitors
You'll be required to create as well as maintain long-tail connectors on your own
The credit-based pricing model can be difficult to predict
Cloud service is missing several well-known connectivity and destination options (e.g., WordPress, Google Analytics 4, Apache Kafka)
Engineering teams that are ready and capable of developing as well as maintaining customized connectors with Airbyte's CDK.
Stitch is an ETL tool owned by Talend that focuses on business intelligence. Stitch was the original creator of Singer, and integrates with it like Meltano. Stitch supports built-in data transformations with Java, SQL, Python, or a UI.
Standard: Begins at $100 per month (for 5 million active rows, one destination, and 10 sources)
Advanced: $1250/month (up to 100 million rows and three destinations)
Premium: $2500/month (up to 1 billion rows and five destinations)
Free 14-day trial available
130+ data source connectors
Integration with the Singer protocol for open connector development
Cost-effective solution for common data sources
Support for popular data lakes and warehouses
Pricing is based on rows per month, with prices increasing as the amount of destinations that are active
Stitch includes 130+ data sources. Some are classified as "Enterprise" and only available on the Premium and Advanced plans.
Stitch comes with a REST API of its own. It also works with Singer, the open-source data connection standard that Meltano is built with.
Stitch provides chat and email support to all customers during normal business hours. Certain customers might also be eligible for telephone support or a dedicated Global Customer Success agent.
Robust transformations that include the ability to parse nested JSON
Quick setup in a matter of minutes and helpful customer service
Compatible with the Talend suite of data tools
Pricing can be costly and difficult to forecast
Customer support is limited for users with the standard plan
Singer connectors may break without warning, and they aren't maintained by Stitch (unlike with Meltano, which has its own Singer maintenance team)
Teams that use the most popular information sources or possess the know-how to develop their own connectors with the Stitch API or Singer protocol.
Fivetran is among the oldest ETL tools, first established in 2012. It's a closed-source platform that's exclusively focused on ETL workloads (after a pivot from its original function as an analysis and visualization tool). Fivetran is a more expensive, enterprise-focused tool that emphasizes the largest and most popular connectors.
Standard select: Est. $60 per month (limited to 1 user and 500k active rows per month)
Starter: Est. $120/month (limited to 10 users)
Standard: Est. $180/month
Enterprise: Est. $240/month
Business Critical: Contact sales
A 14-day trial is available
Native warehouse transformations, including prebuilt models and orchestration through dbt Core in the Fivetran dashboard
Support for enterprise data sources and destinations
Low-impact data change capture for database replication
Fivetran provides more than 160 data source connections.
You can request bespoke data connectors through Fivetran, but the development team will prioritize those they expect to be most popular with the user base. You can develop your own connectors but need a Standard Select plan or higher to access Fivetran's REST API.
Fivetran offers support around-the-clock for all plans. Plans with higher prices have access to SLAs.
Established company with robust performance and integrations with applications in a modern data stack
Integration with cbt Core
Flexible data synchronization options
Usage-based pricing model can be very affordable for businesses with low volumes
There isn't much support for long-tail data connections, and the Fivetran team likely won't prioritize new connectors unless they're widely used
Fivetran is one of the most expensive solutions on the market
Pricing model can be difficult to predict
Fivetran is only able to support ELT workloads, which means that teams using ETL will need to look elsewhere
Companies with low data volumes or large budgets looking for an industry-leading ELT tool.
Hevo is a no-code bidirectional data pipeline that provides ETL, ELT, and Reverse ETL. It comes with built-in transformation capabilities and connects to the majority of major data sources.
Free: Up to 1 million events (limited to 50+ data sources)
Starter: Starting at $239/month
Business: Custom quote
Hevo also provides Hevo Activate for Reverse ETL:
Free: Up to 4 data warehouses and 3 SaaS targets
Starter: $199/month (4 data warehouses and 5 SaaS targets)
Business: Contact sales
150+ data sources (50+ available on free plan)
Real-time data migration
Built-in data transformations through Python
24/7 live assistance
Hevo Data offers 150+ data connectors (only 50+ are available for free).
The Hevo team does not offer the ability to create custom connectors. You'll have to utilize its Webhooks API, REST API source to connect to less common sources of data, or find alternative software for data integration.
Hevo provides 24/7 support, which is generally well-reviewed by customers.
Capabilities for ETL, ELT, and Reverse ETL workloads
Complete automations without code
Prices are affordable compared to other platforms
There isn't as extensive support for long-tail connectors
Manual mapping is required for certain transitions between different tools
There is less control over data ingestion and loading frequency
Small-scale businesses that make use of the most well-known data sources and require inexpensive ETL and ELT tools.
Rivery is a closed-source, cloud-based SaaS ELT solution built for nontechnical teams---making it the opposite of Meltano in many ways.
Rivery is a low-code platform that provides an interactive GUI for building data workflows, which it calls "rivers."
Starter: $0.75 per credit
Professional: $1.20 per credit
Enterprise: Custom pricing
One credit equals different measurements depending on the action taken
Free 14-day trial available
200+ built-in data source connectors
15+ data destinations
Data ingestion, orchestration, transformation, and DataOps management all included
Customizable Starter Kits to get integrations up and running quickly
24/7 customer support
Rivery provides 200+ pre-built connectors. Each plan has access to all connectors.
The Rivery Custom API feature lets you connect to data sources without built-in integrations. You can also request custom data sources from the Rivery team.
Rivery has leading customer support in ELT tools, according to G2. All plans have 24/7 global support, and Professional and Enterprise plans have additional onboarding and personalized attention. The Enterprise plan includes an SLA.
Time-saving starter kits for using the most common data workflows in a few clicks
Detailed development docs for the Rivery platform and API
Exceptional customer support
Credit-based pricing can be hard to calculate and even harder to predict month-to-month
Graphic interface can become unwieldy with more complex data workflows
No development environment which can be limiting for technical teams
Rivery is ideal for teams with a lack of technical knowledge that want a platform that's easy to set up quickly.
Precog takes a completely different approach to ETL connectors than its competitors.
Instead of using an open-source protocol (like Meltano) or creating connectors in-house (like Fivetran), Precog uses an AI engine to understand new data sources and develop workflows.
Request a demo or trial to get pricing (based on the number of data sources used).
10,000+ data sources built-in
Support for more than 100 destinations
SaaS as well as on-premise options are available
No-code platform that lets you create custom integrations without needing to program API connectors yourself
Precog provides over 10,000 data sources---more than Meltano will probably ever offer.
With Precog's AI software, you can design new data connectors in a matter of minutes or even hours without needing to code. The Precog team also develops connectors in 48 hours upon request.
Precog provides customer service based on your plan.
Innovative, no-code approach to interpreting and using new data sources
More data sources out of the box than any other platform
AI engine will automatically adapt to custom fields in data sources
Straightforward, no-surprises pricing based on sources used
Cloud, private cloud, and on-premise options available
Platform is fairly new and has a small user base
For teams with low data volume but a large number of sources, pricing can be more expensive that with other platforms
The no-code model can be restrictive for teams that want more technical control over data pipelines
Teams seeking a non-code solution to import data from a variety of data sources that are difficult to find.
CData is a closed-source data connectivity platform. It has a broad suite of tools to connect, replicate, integrate, transform, and analyze data.
Standard: $49/month (single data source)
Professional: $99/month (up to five data sources)
Enterprise: $199 per month (up to 10 data sources; contact sales for additional sources)
Additional tools in the CData suite are available separately
250+ connectors
A suite of tools that includes Sync, Arc, and many more
Speed-optimized connectors (which CData calls "drivers")
All-in-one solution for nearly all data needs
Options to hire the CData team to set up all your integrations
Support for ETL, ELT, and Reverse ETL workflows
Cloud-based and on-premise solutions available
CData includes 250+ prebuilt real-time data sources. CData calls these data connectors "drivers."
CData offers a Universal API driver which you can use to connect to different data sources. The CData team will also set up all connectors for you within 90 days upon request.
CData provides email support on all plans. Enterprise plans with 5+ users can add Premium Support which includes phone support and priority resolution.
Affordable plans for individual software tools, so you won't pay for more than you need
Lots of connectors built for the most well-known data sources
End-to-end pipeline tools in one platform
Option for white-glove setup, complete with custom integrations, in 90 days
Price for complex pipelines can add up since each tool is sold separately
Custom fields are not supported in data applications
Aside from official support, the community isn't large, and some users say documentation can be hard to understand
Uses proprietary column names which can lock you into CData's infrastructure down the road
Businesses looking for an all-in-one platform with SQL-based connectivity.
Matillion is a reputable data pipeline company that focuses on enterprise customers.
Although the majority of ETL platforms are SaaS (hosted on the cloud) or open-source (available to be deployed using your own), Matillion offers an on-premise solution that can be localized but is backed by closed-source software.
Free: Up to 1 million rows per month
Basic: $2.00/credit
Advanced: $2.50/credit
Enterprise: $2.70/credit
110+ data source connectors
On-premise deployments available
Cloud data transformation using a graphic user interface (GUI)
Consumption-based pricing model
Software compatible with Matillion data stack
Matillion offers 110+ connectors. Because the Matillion data platform is made of two tools---ETL and Data Loader---connectors may be different across the products.
Matillion doesn't support long-tail data sources. Customers can utilize the GUI to build simple API integrations, but the user is responsible for maintaining the integration.
Matillion offers a help site for clients. Integrating everything with Matillion's ecosystem can streamline support.
Strong data transformation capabilities built-in
On-premise option available
Because Matillion handles data loading and transformation it can be easier to control data governance
Matillion's GUI-based transformations may be difficult to master
There are fewer data connections than competitors like Meltano
A smaller number of partnerships with transformation tools due to a focus on Matillion's tools
Large companies searching for a complete tool to handle the process of transformation, particularly companies seeking a solution on-premise.
Segment is a platform built to ingest, transform, and use customer data.
Segment offers several customer-focused features you won't find with Meltano (or other competitors), including merging personas, dividing by stages in the customer journey, and more.
Free: Up to 1,000 monthly visitors and two data sources
Team: $120/month for up to 10,000 monthly users
Business: Custom pricing
300+ connectors available on all plans, including free option
Data management and debugging
Support for pixel embedding, user profiles, and audience segmentation
User data is exported to marketing tools and CRMs
Segment includes 300+ connectors built-in, all of which are available across all plans.
Segment comes with a Functions tool that lets users create their own connections. While it only requires a few lines for each source, Segment is not a no-code tool.
Segment offers Standard support on all plans. Teams spending at least $60,000 can upgrade to Advanced, Premium, and Premium Plus support options.
Customer-centric features that integrate with features you won't find with other platforms
Large number of built-in integrations with the most popular data sources
Affordable prices and a free plan for small-sized companies
Focus on customer data can make the platform less effective for other use cases
Monthly tracked user costs can be expensive for customers with large numbers of visitors, especially if many are anonymous visitors counted multiple times
Data governance, transformations, and other critical features are only available on the most expensive Business plan
Segment is best for data teams that focus solely on customer data.
Picking the right data ingestion tool can limit how effectively you process and learn from your company's big data. Finding an option that "just works," even as APIs and data sources evolve, is vital to protecting your business.
Meltano is a new but fast-growing tool for ELT workloads, but isn't perfect. It's built for people with a programming mindset---to the point there isn't a no-code option.
You'll need to host it yourself, and there is no official support. And Meltano's dependence on Singer offers big advantages, but the serious downside is that connectors can break and leave your data stranded.