Today's ELT tools work in mostly the same way they worked a decade ago. You build manual connectors, sync to your data source, and extract what you need.
But Precog has set out to completely change how ELT works. Rather than using manual connectors, Precog relies on an AI engine to interpret data and apply schema.
Using artificial intelligence to ingest and transform data is innovative, to be sure, but comes with a few key disadvantages.
Today, we'll compare Precog with the top competitors in the ETL space so you can decide what's best for you.
Precog was built on a technical solution to the barrier that's kept other ELT from integrating artificial intelligence: nested JSON.
Precog is based on a new set of algorithms called multidimensional relational algebra that allows AI to interpret data relationships even in deeply nested JSON fields.
This shift means Precog can collect, interpret, and apply schema to almost any data stream.
With some human testing, a new data connector is ready to use for a variety of applications including business analytics, data science, e-commerce, marketing, machine learning, and more. The innovation has let Precog develop thousands more connectors than any other ELT tool.
But Precog isn't the right option for every team. The AI-driven model sets some limitations, so let's explore them right now.
Precog currently only offers custom pricing. It's priced per data source, not volume, so the more sources you need to support the more you'll pay.
10,000+ ready-to-use data connectors and 100+ destinations
Cloud and on-premise options available
No-code interface that automatically builds connectors to new data sources
Straightforward pricing based on data source
No-code user experience can be limiting for technical teams looking for more control over connector development
New platform with limited documentation and community
No public pricing information can be frustrating for teams comparing options
Portable creates the long-tail connectors you won't find with other platforms. Unlike Precog, which relies on AI to generate schema for data connectors, every data source that goes through Portable is coded by hand. It's also maintained by the Portable team indefinitely, so you don't need to worry about connectors breaking unexpectedly.
Portable was founded in 2020 and already supports 250+ long-tail connectors. The team creates new connectors on request and develops them fast, often providing an integration for new data sources within an hour.
Manual: Unlimited use of data sources and destinations with manual sync
Scheduled data flows: $200 per data flow (unlimited volume)
Custom: Pricing tailored to your needs
250+ data source connectors
Major cloud data warehouses supported as destinations
Generous plans (including free plan) without limits on sources, destinations, or volumes
Creation and maintenance of custom data sources at no additional cost
Hands-on customer support
Portable comes with 250+ data connectors out-of-the box. All 250+ connectors are available on all plans---Portable doesn't restrict some connectors to Enterprise customers like some competitors.
Portable is built around developing and maintaining custom long-tail connectors for customers. Turnaround time can be as fast as an hour in some cases.
Portable offers premium white-glove to all customers, no upgrade required.
250+ connectors, all built around long-tail data sources major competitors don't support
Straightforward source-based pricing that's easy to understand and predict month-to-month
Free plan that lets you test all connectors for free for as long as you like
Custom connector development and maintenance at no extra charge
Portable only develops connectors for long-tail data sources ignored by competitors, so you won't find support for the biggest enterprise sources like Salesforce or QuickBooks
No support for databases as connectors
Only available for businesses based in the United States
Portable is best for companies that need long-tail data sources, but don't have the resources for the development and maintenance of these connectors on their own.
Rivery is one of the closest competitors to Precog in its no-code philosophy. Rivery is a data integration tool designed to help customers ingest from data sources as fast as possible without using code, instead offering starter templates with prebuilt data workflows (called "rivers") that work on a GUI or with native Python.
Starter: $0.75 per credit
Professional: $1.20 per credit
Enterprise: Custom pricing
One credit equals 1 API pipeline execution, 100 MB of data replication, or one logic or transformation execution
14-day free trial
200+ data sources
15+ supported data destinations
24/7 customer support
Support for ELT, Reverse ETL, and transformations
Plug-and-play starter kits with prebuilt "rivers" to connect popular data sources and destinations
Rivery has 200+ data source connectors.
You can connect new data sources using the Rivery Custom API. You can also submit requests for new data connectors to the Rivery team, though it's unclear how these are prioritized.
G2 consistently recognizes Rivery for the best support in its category, with a current score of 9.8/10 compared to an industry average of 8.5/10 for ELT Tools. Rivery offers different support levels for its plans.
Rivery's starter kits make it easy to get up and running quickly
No-code "rivers" and GUI are easy to understand for nontechnical users
Highly rated customer support
Pricing is complex, even compared with other competitors that price on volume, and can be hard to understand or predict month-to-month
While the GUI makes simple connections easier to understand, it can become confusing for large and complex data pipelines
Users have mentioned the error messages and alert system can be ambiguous and hard to interpret
Rivery is a great choice for teams looking for a no-code option that's fast to set up.
CData offers a comprehensive set of data tools. CData offers a complete done-for-you setup service that will create custom connectors and integrate your entire data pipeline in 90 days. Its suite of tools works with ELT, ETL, data visualization, and more.
Standard: $49/month (one data source)
Professional: $99/month (limited to five data sources)
Enterprise: $199/month (limited to 10 data sources; custom pricing for more than 10 sources)
Each tool includes separate pricing and plans
250+ connectors (CData calls them "drivers") optimized for fast transfer
Integrates with CData set of software such as Arc, Sync, and Connect Cloud
SQL-based integrations
Broad support of data pipelines, including ELT, ETL, and Reverse ETL
Cloud-based SaaS or on-premises offerings
CData provides more than 250 built-in data connectors (called "drivers").
CData has a unique offer to program your entire data pipeline, including custom connectors, in 90 days. You can also create custom connectors not supported by CData using its Universal API driver, but you'll need to maintain these connections yourself.
CData offers two levels of support. All plans include email support, and Enterprise customers can add Premium support which includes phone assistance and guaranteed 24-hour response time.
Straightforward monthly pricing model
White-glove setup service that includes creating custom connectors
Wide set of built-in data connectors to the most popular sources
Comprehensive suite, with software for almost every use case
Price is for a single piece of software, meaning complex data sets that require several tools can get expensive quickly
Not all connectors import custom fields in data, meaning some data will be unusable
Documentation can be hard to understand and use
CData is an ideal choice for teams looking for several types of data tools all in a single platform.
Stitch is a business intelligence ETL tool. In its early days, Stitch founded Singer, an open-source framework to facilitate creating new connectors for data sources and destinations. Stitch still integrates with Singer but after its acquisition by Talend in 2018 has focused on building proprietary connectors.
Standard: Starts at $100/month (limited to five million active monthly rows, one destination, and 10 sources)
Advanced: $1250/month (limited to 100 million rows and three destinations)
Premium: $2500/month (limited to one billion rows and five destinations)
14-day free trial
130+ data sources and integration with Singer protocol
Support for popular data lakes and warehouses as destinations
Alerts and monitoring automation tools
Built-in transformation capabilities with Java, SQL, Python, or native UI
Stitch includes 130+ data sources. Some of these are considered "Enterprise" connectors and are only available on Advanced and Premium plans.
Stitch is compatible with Singer for creating new connectors. Singer's flexibility means you can create new "taps" (for data sources) or "targets" (for destinations) that are automatically compatible with other taps and targets built with Singer.
Stitch provides chat and email support during work hours. Higher-tier plans are eligible for phone support and dedicated representatives.
Built-in transformations using several programming languages as well as through the Stitch transformation UI
Automation tools to save time and manual work
Connection to Talend suite of data tools
Cost-effective plans for small businesses or teams without large data needs
Pricing based on monthly active rows can be hard to predict each month
Connectors built through Singer aren't maintained by Stitch and are prone to break or stop working unexpectedly
Limited customer support on less expensive plans
Stitch is best for teams that use common data sources or who are familiar with Singer for creating long-tail connectors.
Hevo is a data pipeline tool for ELT, ETL, and Reverse ETL workflows. Like Precog, Hevo is built with a no-code UI and has built-in data transformations. But it has a fairly limited selection of data sources that's even more restricted on the free plan.
Free: One million events (limited to 50+ data connectors)
Starter: Starts at $239/month
Business: Custom quote
For Reverse ETL workflows, businesses will need to use Hevo Activate:
Free: Access to 4 data warehouses and 3 SaaS targets
Starter: $199/month (limited to 4 data warehouses and 4 SaaS targets)
Business: Contact sales
Flexible pricing
150+ data sources (free plan only gives access to 50+)
Real-time data sync available
Round-the-clock live support
Built-in data transformation capabilities with Python
Hevo has 150+ data sources. Only 50+ are available on the free plan.
There are two main ways to create new data connectors with Hevo. The first option is to use its webhooks API to import data points from the application of your choice. The second method is to integrate your own connector using Hevo's REST API. In both cases, you'll be responsible for ongoing maintenance.
Hevo provides 24/7 customer support for all its plans. The level of support depends on the plan---Starter includes live chat, and Business includes a dedicated account manager.
No-code platform is easy to understand for nontechnical data teams.
Broad range of supported data workflows including ETL, ELT, and Reverse ETL
For teams that only need the supported 50+ data connectors, the free plan has generous usage limits
Usage-based pricing can be expensive for large data loads and is hard to predict in advance
Transferring data between tools can require manual mapping
Free plan only includes a very limited number of data connectors
Hevo is best for nontechnical teams looking for an easy-to-use, no-code tool that only need support for the most common data sources.
Matillion is a data workflow program with an option to host on-premise. Matillion's main focus is on the data needs of larger companies.
Free: Limited to 1 million rows per month
Basic: $2.00/credit
Advanced: $2.50/credit
Enterprise: $2.70/credit
110+ data source connectors
GUI-based cloud data transformations
Cloud-based SaaS and on-premise deployments available
GUI-based data transformations for the cloud
Integration with Matillion Data Loader and Matillion ETL
Matillion comes with 110+ prebuilt data connectors. Matillion is divided into two separate programs---Matillion Data Loader and Matillion ETL---and not all connectors from the ETL tool are available with Data Loader.
Of the Precog competitors on this list, Matillion has the fewest connectors. That means it's even more likely that you'll need to create a custom connector to work with long-tail data sources. Matillion includes a GUI for connecting to other APIs, but development and maintenance falls on the individual user.
Matillion's integrated ETL and Data Loader platforms could need less support overall compared to a plan made of different products. Matillion's support is the same across all plans but Enterprise customers can add on Mission Critical Support.
Built-in data transformations
Flexibility with both on-premise and cloud-based deployment options
Improve data governance control with integrated platform
Just 110+ data connectors, fewer than any other Precog competitor
Deep integration with other Matillion tools can mean it's difficult to connect to a third-party tool not part of the Matillion ecosystem
GUI-based data transformations can have a steep learning curve
Matillion is best for larger businesses looking for an all-in-one platform, especially those looking for an on-premise solution.
Segment is a platform for customer data. Segment's focus on customer data gives it a clear focus in the market, but also means it's probably not the best Precog alternative for other use cases. Twilio purchased Segment in 2020.
Free: 1,000 visitors per month and two data sources
Team: $120/month for up to 10,000 monthly users
Business: Custom pricing
300+ built-in data sources
Tools for data governance and debugging
Unique customer-focused features like user profiles, segmentation, and embedding pixels
Functionality to export data to CRMs and marketing platforms
Segment includes 300+ connectors, all of which are included in all plans.
Segment Functions is a built-in tool for developing custom connectors. While the tool makes it easier to develop new data connectors, your team will still be responsible for ongoing maintenance.
Every plan comes with Standard support. For teams spending $60,000 or more, Segment offers four levels of paid support add-ons.
Free plans for teams with fewer data needs
Unique customer-centric features you won't find on any other platform, including Precog
Intuitive interface without much of a learning curve
Segment's user focus means it's likely not a good fit for other big data use cases
Pricing based on monthly users can result in duplicate charges for businesses with large amounts of anonymous users
Not all connectors support custom fields, meaning even some supported data sources may require additional customization
Segment is best for teams that only handle user data and want a platform tailored to that use case.
Fivetran is one of the oldest and most established ELT tools around. It's a closed-source platform that offers managed solutions geared toward enterprise customers and data sources like Oracle, SAP, and Workday.
Standard Select: Est. $60/month (limited to one user and 500k active monthly rows)
Starter: Est. $120/month (limited to 10 users)
Standard: Est. $180/month
Enterprise: Est. $240/month
Business Critical: Contact sales
A 14-day free trial is available
160+ data sources
Native warehouse data transformation
Database replication with data change capture
Focused on the premium needs of large enterprises
Fivetran offers 160+ built-in data connectors.
You can request custom data connectors through Fivetran, but the team will decide which to develop. Access to Fivetran's REST API requires the Standard Select plan or higher.
Fivetran provides 24/7 support on all its plans. Higher-priced plans offer SLAs.
Near real-time or real-time data sync options
Support for the most important enterprise data sources and destinations
Robust features and reliable platform trusted by the world's biggest companies
Higher prices than nearly every other tool on the market
Usage-based pricing model that's opaque and hard to predict from one month to the next
Little support for long-tail data sources and not a high priority for future development
Only supports ELT workloads
Fivetran is best for large enterprises looking for a solution tailored to their needs.
Airbyte is an open-source ELT data platform. It's a relative newcomer---founded in 2020---but has built a passionate customer base in a very short time.
Open source: Free, as you'll be hosting the software yourself
Cloud: $2.50/credit (one million rows equals 6 credits; 1 GB equals 4 credits)
Cloud high volume: Custom pricing for users who need more than 5,000 credits
170+ data source connectors (limited selection on the cloud service)
Open-source solution with broad community support
Data change capture support for databases
Data warehouses and data lakes as destinations
Paid managed solution and self-hosted option
Warehouse-native data transformations
Airbyte has 170+ connectors in total. The cloud platform currently doesn't have access to all of these but support is expected in the future.
You can request connectors through the Airbyte website, but there's no guarantee the team or community will work on it. You can also code a new connector yourself, but if the API changes at any point you'll be responsible for updating it.
Although it is a new solution, Airbyte has a robust community of over 12,000 users. Paid plans also come with in-app chat support with one-hour response times.
CDK to help with custom connector development
Open-source platform with broad community support
Self-hosted option as well as SaaS cloud-based plans
Support for 20+ destinations
Credit-based pricing model can be hard to understand and predict month-to-month
You'll have to create and maintain connectors to all long-tail data sources
Cloud service is missing many of the integrations included in the open source version, including major sources and destinations like Google Analytics, Apache Kafka, WordPress, and more
Engineering teams who have the capacity and resources to build custom connectors with Airbyte's CDK.
Meltano is an open-source ELT tool and the newest Precog competitor on this list, spinning off from GitLab in 2021. Meltano and Precog lie at opposite ends of the spectrum, with Meltano offering an open-source, developer-first platform that's built on manual connectors created with Singer.
Meltano currently only offers a self-hosted option that's free. A managed cloud platform is expected to launch in 2023.
300+ data sources
Native integration with Singer open-source data connector protocol
CDK for creating custom connectors
CI testing, version control, and additional development features built-in
Meltano is compatible with 300+ data sources. Meltano integrates with Singer and has its own team responsible for maintaining these connectors.
You can create a new connector using Meltano's CDK and Singer. It's likely your team will be responsible for maintaining this connector if its API ever changes.
Meltano currently doesn't offer official support, though this will likely be added once the managed cloud solution launches in 2023. Meltano's Slack community has 2,000+ members.
Open-source, free platform
Best practices for developers built-in
Singer integration makes it easy to connect with new data sources and destinations
Meltano is a platform built by developers, for developers---it doesn't have a no-code option for nontechnical users
No guaranteed maintenance meaning connectors could break without warning
No managed plan available, only self-hosted open source
Smaller user base and no official support
Meltano is ideal for teams with a high degree of technical proficiency who want complete control over the data ingestion process.
Precog is an innovative and attractive ELT tool, but it comes with its flaws.
Its AI interface makes it easy to connect to long-tail data sources, but can also limit your team's ability to adapt those connectors.
And experimenting with a new program before its user community has gone mainstream poses a challenge for any business.