The easiest route - we'll conduct a 15 minute phone interview and write up the review for you.
Use our online form to submit your review. It's quick and you can post anonymously.
In my current project, the specific use case of Fivetran is bridging the gap between siloed data, allowing me to extract and have it in the same data warehouse. Fivetran works with its seamless connections to sources and destinations, helping to avoid reinventing extraction logic from scratch. Fivetran already has data models, so it pulls in data quickly, providing it all in the same data warehouse.
I've worked extensively with Fivetran, mainly used for extraction purposes, and I've worked with the transformation element in it as well. Fivetran not only has built-in connectors but also provides SDK connectors, allowing us to develop our own connectors in an easy manner. I don't have to write raw Python scripts or dumping scripts; it offers straightforward examples and guidelines, making it much simpler to develop custom connectors inside Fivetran. We've been able to develop many custom connectors as well, which is unique and beneficial for having everything centralized instead of having those connectors located elsewhere.
One of the best features by Fivetran is its clean, simple, and intuitive UI. It includes a transformation section where I can deploy my DBT queries and scripts. It also supplies good tracking capabilities for billing estimates and user permissions, allowing for customization to the desired level. The number of connectors it has remains a standout feature, and within connectors, the options available are very helpful. Although it sometimes appears static due to its built-in nature, it offers good flexibility for data transformation and caching, which I appreciate because it saves us extensive script-writing time.
The experience of using SDK connectors can be improved, as it lacks computational power, which is an area that needs enhancement. Additionally, for some connections, I want more flexibility during ingestion, specifically for transformations needed beforehand. When we tried to use a built-in Oracle connector, it didn't allow for the tweaks we needed, which led us to the SDK connector route and caused delays in development.
I've been working for almost four years in this field, and it has been a really great learning and challenging opportunity, allowing me to work on various different stacks and to learn and grow extensively.
In my experience, Fivetran is stable with very few instances of downtime or reliability issues, and overall, I am very happy with it.
Fivetran's scalability has been tested effectively, and it has been working well for our organization's growing data needs. However, the performance with SDK connectors could be improved, as it took around 10 to 12 days to conduct historic ingestion for just two years, which I find unsatisfactory.
Customer support from Fivetran is quite good; it's really nice and responsive. I am very happy with my experiences in reaching out to them.
Positive
Before Fivetran, we used Skyvia, which is a no-code ETL tool. I also considered Data Factory as an alternative. However, I found Skyvia lacking in maturity and Data Factory better suited for enterprise solutions but limited in connectors. Scripting solutions demanded more work and included additional costs, which led to my decision to switch to Fivetran.
My experience with pricing, setup costs, and licensing has been satisfactory; however, I notice that as row numbers increase significantly, it can get a bit more expensive. Overall, it's fairly reasonable.
We see fewer employees needed due to Fivetran's efficiency, which translates to money saved, as employee wages are much higher compared to the costs associated with the software itself.
Fivetran provides time savings, cost reductions, and improvements in data quality. For example, in working with Shopify, we faced numerous errors and challenges with in-house complicated connectors. The time directly correlates to cost; when the time investment decreases, fewer developers are needed, which means significant savings. Generally, we have seen very few instances of data quality issues since implementing Fivetran.
Before selecting Fivetran, I evaluated other options including Stitch, Rivery, Hevo, and Skyvia among others.
My advice for those looking to use Fivetran is to rely on its documentation, which is very straightforward and accurate. Avoid depending too much on AI for Fivetran issues; the dedicated documentation suffices for most inquiries. If problems arise, Fivetran's support team is quick to assist. I rate Fivetran 9 out of 10.
The usual use cases for Fivetran that I work with involve real-time replication from Oracle Fusion, Oracle DB, and some of the APIs. I also work with replication from S3.
The most valuable feature of Fivetran so far is the data replication. The real-time data replication is what I see best in the market where it reduces the overhead of customers needing to maintain the pipeline. Even when an error occurs, it automatically tries to solve the issue without manual interaction needed.
The effectiveness of Fivetran's pre-built connectors in enhancing my customer's productivity is significant because everything is easily configurable. The documentation is very clear, so there's no difficulty in finding the necessary steps. Everything is pre-documented. The pre-built connectors that are available are excellent. Most functionalities are covered, and in most cases, it works according to expectations.
If Fivetran can build in more transformations, that would be really helpful in my opinion.
I have been using Fivetran collectively based on project requirements. I work with AWS, GCP, Snowflake, and Fivetran as part of my experience. My experience with Fivetran specifically extends to about two to three years, depending on project requirements.
I have not experienced any challenges with deployment.
From my experience, Fivetran is very reliable. They have 99.9% accuracy on the data load and they maintain transparency. When the data is loaded, they clear it from their storage, keeping only metadata for MAR calculation and pricing purposes.
I often communicate with the technical support of Fivetran whenever we encounter issues. I would rate them a seven. If they could provide support more quickly, that would be great.
Positive
I usually participate in the initial setup and deployment of Fivetran. I handle the solutioning from start to end.
I don't use Fivetran's automated ETL feature, but after loading, we use dbt for the transformation.
I have not used Fivetran's Schema Drift Management feature.
The integration with data warehouses such as Snowflake and BigQuery has been successful. Fivetran has clear documentation on configuration for both warehouses.
Regarding pricing, Fivetran uses MAR pricing, which I would consider moderate. The pricing consideration ultimately depends on the project budget since the client is responsible for payment while I implement the solution.
On a scale of 1-10, I rate Fivetran a 9.
