Data Integration: Hyperscalers Lead in Closed Ecosystems

By riding the cloud migration trend, CSPs are gaining visibility into data integration.

Gartner had noted this at the end of 2024 in its Magic Quadrant synthesis dedicated to these solutions. It highlighted that this increased visibility translated into a notable gain in market share.

A year later, the finding still holds. In parallel, another vendor typology is standing out for its growth: niche players offering more specialized—or more cost-effective—products.

Without Informatica, SAP is no longer a “Leader”

Gartner bases its assessment on two axes. One is forward-looking (“vision”), focusing on strategies (industry-specific, go-to-market, marketing, product…). The other concerns execution—the ability to actually meet demand (customer experience, pre-sales performance, quality of products and services…).

The situation on the “execution” axis:

On the “vision” axis:

Nine of last year’s top ten “leaders” remain in place. Alphabetically they are: Ab Initio, Denodo, Google, IBM, Informatica, Microsoft, Oracle, and Qlik. SAP slips to the “visionaries” due to a retreat on the Execution axis. Gartner did not consider Informatica’s acquisition, completed on December 8, 2025.

Ab Initio lauded for automation and agentic capabilities…

The product under consideration is Ab Initio Data Platform.

Last year, Ab Initio was praised for handling complex data-management use cases within large enterprises. Gartner also valued the customer experience, driven by a direct engagement approach. It also highlighted the company’s knowledge-graph approach that linked business indicators to data models.

This year, Ab Initio is lauded for the stability of its leadership team and its longstanding customer base. It is also praised for its support and the resilience of its platform. It scores well for its metadata-driven automation approach and templates, as well as for its AI Central agentic framework (data understanding, pipeline creation, natural-language interaction…).

… but not the UI, nor the pricing

Configuration and upgrades can be time-consuming, in addition to a steep learning curve for technical teams, Gartner noted last year. Customers tend to find pricing steep and license management challenging, it added, while also noting weak penetration for simple use cases like autonomous ETL.

The point about the learning curve remains current. Gartner adds a UI that is not intuitive and minimal community support. Ab Initio broadly lacks visibility relative to other “leaders” (particularly when it comes to content production). Its pricing is complex, and on-prem deployments lack flexibility.

AWS narrows the gap with the competition…

Most of the services Gartner grouped under its evaluation—Glue, Kinesis, Athena, etc.—are included in the Amazon data-management platform SageMaker.

Last year, Amazon stood out for ecosystem thinking—from the “zero-ETL” integration between S3, Redshift, and Aurora to Glue-SageMaker connections, plus Data Zone for metadata management. Gartner also appreciated the ability to manage multiple user profiles (Glue combines notebooks, GUI, spreadsheet interface, and NLP with Amazon Q). The serverless architecture, welcomed by customers, especially for autoscaling efficiency, was also highlighted.

This year again, Gartner notes the level of integration with the rest of AWS—emphasizing shared governance. It also praises the robustness of the data-prep offering for GenAI use cases. It notes that AWS has managed to close the gap with competitors on streaming data and advanced transformations.

… but remains centered on its ecosystem

Last year, Gartner noted that Glue could incur high costs, especially with large data volumes. And despite the ability to connect to external databases, it did not offer the depth of integration of pure-plays—further, it cannot be deployed on other public clouds. Another limitation: the complexity of usage for advanced data-engineering scenarios requiring code (for example, integration with Apache Iceberg and managing Spark jobs).

This year, the constraint of “AWS-centric” is framed more generally: the catalogue of connectors to other destinations is limited. On the one hand, the sources are relatively rich; on the other, the configuration often lacks flexibility. Add to that pricing that is perceived as high, with price hikes that can be unpredictable and cost-management tools that customers want improved. Also watch the maintenance of pipelines, which is often complex and time-consuming, and automated remediation is limited.

Denodo remains distinguished for data virtualization…

The product considered is Denodo Platform.

Last year, Denodo stood out for its brand recognition in data-virtualization, Gartner noted, as well as its growth that outpaced the market and its expanding partner network. It also credited user experience, especially on the functional side.

This year, Denodo’s data-virtualization reputation remains strong. Granular access control and the evolution of its data catalog into a data marketplace for products earn it more points. Gartner adds the Denodo Assistant bricks (automatic description and labeling of sensitive data) and DeepQuery (answering business questions with a reasoning model).

… but remains underused for certain types of integrations

Denodo’s products are rarely used for bulk/batch or replication integrations, especially when performance SLAs are in play, Gartner noted last year. It also mentioned the absence of accelerators or industry templates, and the difficulty of optimizing and maintaining distributed deployments.

The first observation still applies (and also to streaming data integrations). Gartner notes the frequent need for complementary tools to cover all types of integration and complex use cases. It also cites the difficulty in resolving issues when integrating third-party software and in configuring SSO on complex deployments; plus the lack of native monitoring.

A Google that is distinctly AI-focused…

Gartner considered Cloud Data Fusion (visual pipelines), Datastream (replication), Dataflow (streaming data), Cloud Composer (orchestration), and BigQuery Data Engineering Agent (enrichment and automation of pipelines in BigQuery).

Last year, Google stood out for the level of Gemini integration within its offering. Another strength was governance capabilities at scale (automatic discovery, lineage, metadata exploitation…). Gartner also noted that the products were easier to use than average for data engineers—with exhaustive documentation.

This year, the note on Gemini shifts to a note on the ability to cover AI-use cases, supported by integration with Vertex AI. Gartner also appreciates adaptation to multiple profiles (visual pipelines, notebooks, code…) and the BigQuery data-engineering agent’s capabilities (pipeline creation, problem-solving…), even if it does not extend to pipelines implemented with Google’s other data-integration tools.

… but it is also Google-centric

Last year, Gartner observed a Google-centric offering; it urged caution for anyone not fully engaged in this ecosystem. It also pointed to a lack of portfolio unification (those needing multiple integration modes may require several tools).

The Google-centric vision remains: products are designed and sold primarily for use within the Google Cloud ecosystem. The portfolio remains fragmented—ten tools in this instance—with user experience and functional capabilities varying significantly.

Unstructured data and hybrid deployments, IBM’s strengths…

The product considered is watsonx.data integration, which includes DataStage (bulk/batch), Data Replication, and StreamSets (streaming data), all delivered within the watsonx.data platform.

Last year, Gartner praised IBM’s broad vision—from managing hybrid deployments to active metadata exploitation and the involvement of watsonx AI. It also highlighted its geographic reach and partner network. The acquisition of StreamSets had enhanced its ability to manage complex pipelines in multi-cloud environments.

This year, one of the strong points is the architecture, which decouples pipeline design from the integration style, the latter selectable at runtime and deployable in hybrid/multi-cloud. Another strength is the handling of unstructured data, supported by Granite and Slate models. Gartner also notes the level of integration with the watsonx.data intelligence component, which includes catalog, lineage, and data-product management.

… but pricing remains a sticking point

For comparable use cases, IBM’s solutions are more expensive than the competition, Gartner noted. The firm also highlighted that activating elasticity and governance capabilities could involve complex configuration. It pointed out a lack of clarity around DataStage license portability and best practices for migrating to Cloud Pak for Data.

This year again, IBM is pricier than the competition—or at least perceived as such. The resource-unit model contributes to this perception. The offering rarely appears on shortlists and in modern data-architecture projects, except for those already using DataStage. It is also less considered by organizations seeking specialized tools not part of an integrated suite (data virtualization or data replication, for example), especially when the source is not a mainframe or an IBM database.

Informatica, once again praised for offering maturity…

The considered offering is Cloud Data Integration, which Informatica distributes within its IDMC platform (Intelligent Data Management Cloud).

Last year, Informatica earned praise for its CLAIRE AI engine and its data-fabric vision. It was also noted for its approach to preparing data for AI use cases, and more broadly for the maturity of its offering—the breadth of connectors, the range of use cases covered, and the integration styles supported.

This year again, Gartner mentions a “clear AI vision,” spanning unstructured data handling, the Agent Engineering brick, and CLAIRE Copilot and CLAIRE GPT modules for pipeline management. Its brand recognition, partner ecosystem, and talent pool are other strengths. The maturity of the offering remains, for the same reasons as last year.

… but still under pressure

Last year, Gartner explained how the rise of CSPs was a “challenge” to Informatica’s growth. It also noted that pay-as-you-go pricing might not be advantageous for those who would use only portions of the product. It emphasized that a large part of the customer base remained on PowerCenter, with migration to IDMC potentially costly and time-consuming.

This observation remains valid. This year, it is accompanied by another finding: customers are “exploring alternative solutions”… Informatica’s market share is also declining, primarily to the benefit of hyperscalers. There will be continued vigilance regarding the roadmap and pricing now that Informatica is part of Salesforce.

The data-fabric vision still resonates with Microsoft…

Gartner considered Data Factory (included in Microsoft Fabric), as well as Azure Data Factory, SQL Server Integration Services, Power Query, and Azure Synapse Link.

Last year, Microsoft earned a strong note for the significant adoption of Fabric, among both new and existing customers. Gartner also praised the level of integration with the rest of the Azure cloud and the injection of Copilot capabilities.

This year again, the adoption of Microsoft Fabric—and the Data Factory component—gets favorable mentions, as does the partner ecosystem and the pace of feature development. The Real-Time Intelligence module (stream processing) is praised for its accessibility and intuitiveness.

… but it, too, centers on its ecosystem

Like other hyperscalers, Microsoft has a product focused on its ecosystem, Gartner noted last year. It also pointed to relatively low user satisfaction with support. It stated that replication and virtualization capabilities still lacked maturity, which led some to consider the offering mainly for simple deployments.

The lack of maturity of the offering remains evident this year. Features may not suit production use, Gartner says (examples include CI/CD updates and data copying). On-prem capabilities are limited; the focus is clearly on the cloud. And the offering remains Microsoft-centric—effective mainly for those already in Azure or Microsoft Fabric.

Oracle preserves the GoldenGate advantage…

The main offerings considered are GoldenGate and OCI Data Integration. Gartner also considered Oracle Data Integrator and Oracle Autonomous Database Data Studio.

Last year, Gartner highlighted Oracle’s “agnostic” approach, with OCI acting as a hub across CSPs (metadata sharing, FinOps…). It also appreciated GoldenGate’s replication and streaming capabilities, and its strong support for complex scenarios, including hybrid environments.

This remains true, with emphasis on on-prem deployments support. The same goes for GoldenGate’s replication and streaming capabilities. Gartner adds AI features, led by an agentic framework.

… but it is generating less and less interest

Oracle tends to appear less frequently on shortlists than other players in this market, Gartner noted last year. Its solutions are perceived as expensive, he added. Despite the connector catalog, they are more often considered when Oracle databases are the source or destination of the integrations.

Robust on operational data integration, GoldenGate often eclipses the portfolio touching on analytical data, Gartner states. Pricing remains perceived as high, especially given the lack of transparency. Oracle, more broadly, draws less interest than in prior years, and its customer-retention rate sits below the market average.

Several robust bricks at Qlik…

Three solutions were considered: Qlik Talend Cloud (the main one), Talend Data Fabric, and Qlik Replicate.

Last year, Gartner credited Qlik for the robustness of its replication and data-preparation bricks. It did the same for its connector catalog and for governance, strengthened by the Talend acquisition.

This year, the firm maintains that the replication brick is “one of the best on the market.” It also commends a holistic view of data management, driven by an emphasis on governance and a commitment to a lakehouse architecture following the Upsolver acquisition. Another strength is the product’s robustness in bulk/batch processing and data transformations.

… but with a slowdown in R&D since the Talend acquisition

The Talend acquisition potentially weighed on R&D, Gartner noted last year. It had also stated that Qlik could mature in data-virtualization and that it had said little about price hikes.

This last point remains valid; it surprised some customers, with the lack of public pricing fueling frustration. As for the R&D slowdown, it has been confirmed, creating uncertainty about Qlik’s ability to keep up with market pace. Also beware of limited automation capabilities, both for pipeline design and for optimizing data transformations.