5 Reasons Data Engineers Choose CData Sync Over Fivetran

ETL/ELT software can seem invariable, with each tool seemingly akin to any other—a tool that moves data from Point A to Point B. But organizational infrastructure and data sets aren’t homogenous, and in reality, neither is ETL software.

Beyond features, a software vendor’s approach to data movement and connector integrations can differ from others. When there's a mismatch in organizational needs and vendor capability, companies often discover they need a new partner in data movement. With the recent changes in how Fivetran calculates unit costs per million rows, more users are rethinking their data pipeline options.

Here are the top five reasons data engineers choose CData Sync over Fivetran.

1. Flexibility to deploy anywhere

Organizational IT infrastructure is rarely neat and tidy. There’s no one-size-fits-all approach. If you have a diverse architecture hosted locally and in public and private clouds, a SaaS version of ETL (extract, transform, load) software hampers your options to deploy in the most optimal way to meet your company’s security, compliance, and performance requirements.

Fivetran’s product was built as a SaaS offering, and only within the company’s top-tier plans are hybrid or self-hosted options available. Fivetran’s primary method for connecting to on-premises data is via a locally hosted agent. An agent-based approach is a deterrent to many concerned about introducing a new potential point of failure, added security risk, and possible performance issues.

Sync offers self-hosted and SaaS versions of its software at all pricing tiers with a uniform license cost wherever you choose to implement. If you need it behind your firewall, you can do that. If your Sync instance needs to be completely disconnected from the internet, you can do that, too. Same for clustering instances to meet performance criteria.

2. Connect to all your data – at a deeper level

Just as deployment options matter to data engineers, they also want to access all their data sources and destinations. Additionally, they need to trust that the data will be accurate and complete.

Unless you’re on Fivetran’s Enterprise or Business Critical plan, you’re restricted to cloud-only connectors. That’s a serious limitation for accounts with hybrid environments. Sync, however, provides access to on-prem and cloud connections at every plan level.

The breadth of connectivity options is important, but not all data connections are the same. Fivetran claims more than 650 integrations, but at least 450 of those are “lite” API connections rather than fully managed driver-based connections. CData’s business was built on driver-based connections, and Sync includes more than 270 driver-based connections in addition to hundreds of API connections. That results in meaningfully deeper connectivity across a wider range of connectors.

One Sync customer moved away from Fivetran because its Jira connector couldn’t identify and replicate the customer’s hundreds of custom Jira fields. At the same time, the connector required replicating unwanted historical data, which incurred additional per-row charges. With Sync, the customer can find all their custom fields and can customize precisely what columns are moved into its data warehouse.

Sync unlocks the full potential of your data by seamlessly integrating with both on-prem and cloud data warehouses. Unlike Fivetran limiting most plans to cloud-only destinations, Sync supports a broad spectrum of databases for all customers—32 in total versus Fivetran’s 19. This includes SQL Server, Oracle, DB2, MySQL, PostgreSQL, Snowflake, Databricks, and more. This flexibility empowers enterprises to migrate, transform, and analyze data across hybrid infrastructures without vendor lock-in.

3. Predictable pricing: connection-based vs. row/usage-based

Usage-based pricing is pervasive across the software industry. You pay for compute in self-hosted clouds. You pay for the volume of data ingested into Snowflake or Databricks. But you have options with Sync.

If an organization has low data volumes, a usage-based pricing structure for ETL would make sense. We at CData hear from a lot of customers who started off with relatively low volumes of data running through Fivetran, but as row counts grew, so did their bill.

An annual Fivetran contract to replicate 100 million monthly rows—using only a single connector—is $32,986 on its Standard plan.

Sync customers pay based on the number of data connections they need, which is easier to budget and scale. Sync’s Standard plan is only $8,000 per year and allows for five connections and up to 100 million rows of replicated data each month. That’s a huge difference, and for many is a no-brainer.

4. One tool for ETL/ELT, reverse ETL, and custom transformations

Simplifying software sprawl is on every IT wish list. There are good specific-use tools that can piece together robust data pipelines. Many organizations, however, don’t want the administrative and cost overhead of bucketing together separate software when it can be avoided.

Fivetran’s focus is solely on ELT (extract, load, transform), with transformations only as a post-flight process, largely outsourced to dbt, which it integrates into the platform. Fivetran also has no capacity to perform reverse ETLbringing enriched warehouse data back into operational systems like CRMs. This makes running separate reverse ETL-only software necessary, which exponentially compounds costs and complexity.

Conversely, Sync is built for ELT and ETL, allowing transformations in-flight or post-flight. Sync also features ever-expanding reverse ETL functionality, allowing users to bring SQL Server, Snowflake, and Redshift data back into Salesforce or Dynamics 365.

While like Fivetran, Sync integrates with dbt Core and dbt Cloud to facilitate data transformations, Sync also offers Custom SQL transformations – which can handle custom scripts to enable sophisticated data manipulation and analysis.

5. Granularity of data security, governance, and control

Every software purchase requires tradeoff costs. No tool is going to perfectly match what every organization would design in-house given limitless resources.

When deciding to implement ETL software, an IT team wants the most efficient, secure, and accessible way to move data to be readily used for trusted analysis. CData built Sync to encourage rather than restrict security and control.

Only CData offers a private cloud

Sync’s SaaS offering provides each customer with a private cloud instance, which is separate from any other customer's Sync deployment. This contrasts with Fivetran’s shared cloud, a model more susceptible to data breaches because the same database is accessed by multiple accounts.

Full control of your data

Unlike Fivetran, which applies a black-box data normalization process that lacks transparency on how data is altered, Sync ensures that no changes occur unless explicitly defined by the customer. Additionally, Sync uses end-to-end encryption, and because it does not store customer data, the software has no access to your information.

Synchronize your data as often as you need to

Control over pipeline job scheduling is also a sticking point, as users want to avoid delays in accessing updated information. For batch replication jobs, Fivetran restricts scheduling frequency to once an hour on its Starter plan and once every 15 minutes on Standard. Only at the Enterprise and Business Critical plans can users schedule 1-minute data syncs.

With Sync, every plan includes 1-minute scheduling, giving you real-time control over your data pipeline without upgrade barriers—ensuring your data is always fresh.

Easy and compliant CDC access

Customers find that Fivetran’s requirement of elevated admin database permission levels to configure change data capture (CDC) isn’t practical from a security and compliance standpoint. Sync doesn’t need enhanced database access to configure CDC, making setup and performance more efficient and less risky.

CData Sync wins over Fivetran

See how Sync compares to Fivetran to help you quickly deploy robust data replication pipelines between any data source and any database or data warehouse.

Compare Sync and Fivetran