# Dataddo - Complete Reference > The data backbone for AI: real-time, governed data movement for AI, analytics, and the operational systems teams act in. Dataddo operates data movement as managed infrastructure - orchestration is centralized while data movement executes wherever the data lives, in the customer's cloud, on-premises, or both, so regulated and hybrid environments can use a managed service without data leaving the perimeter. --- ## Company Overview Architecturally, a single Control Plane orchestrates Data Planes that run where the data lives - in the customer's cloud, on-premises, or both. Founded in 2019, headquartered in Prague, Czech Republic. Used by enterprises in regulated industries (banking, insurance, healthcare, public sector) as well as digital-native companies at scale. Key differentiators: - Control plane / data plane architecture - management runs in the cloud; data movement executes inside the customer's environment when required - Hybrid and on-prem deployment supported natively - sensitive workloads never traverse public infrastructure - 400+ actively maintained connectors; Dataddo owns the connector contract end-to-end (API changes, schema drift, and maintenance are Dataddo's responsibility, not the customer's) - All transport patterns in one platform: ETL, ELT, log-based CDC, streaming, Reverse ETL, batch file delivery, direct delivery over REST / Apache Arrow - Four interfaces, one governance layer: UI, REST API, CLI, and MCP server - EU data residency available; the data plane can run entirely inside the customer's own environment - SOC 2 Type II certified, ISO 27001 certified, GDPR aligned, PCI DSS compliant Positioning note: "the data backbone for AI" is the current top-level frame. "Control plane" describes the architecture, not the positioning. ## Architecture: Single Control Plane, Multiple Data Planes - Manage the entire data integration operation from a single Control Plane while running Data Planes wherever the data lives - cloud, on-premises, or both. - One Control Plane can orchestrate multiple Data Plane deployments simultaneously: centralized governance without sacrificing architectural flexibility. - Sensitive data can be processed exclusively within a private, on-premises Data Plane, while less sensitive workloads run in the cloud - all under a single management interface. - Data Planes deploy on all major hyperscalers (AWS, Azure, GCP), regional, local and sovereign cloud providers, and on-premises container platforms including Kubernetes, Red Hat OpenShift, and VMware Tanzu. - Connection patterns supported across cloud and hybrid: cloud-to-cloud, on-prem-to-cloud, cloud-to-on-prem, and on-prem-to-on-prem. Reference: /platform/architecture, /platform/control-plane, /platform/scalability ## All Data Transport Types One platform, every movement pattern - no separate tool just because the delivery pattern changed: - ETL & ELT - classic extract-transform-load, or load-first with in-warehouse transformation - Change Data Capture (CDC) - log-based, low-latency replication that tracks row-level changes as they happen, including deletes - Data Streaming - continuous, event-driven pipelines for time-sensitive and AI-ready workloads; no data is cached in transit - Reverse ETL - push curated warehouse data back into business tools, CRMs, and operational systems (insert, update, upsert) - Batch File Delivery - structured delivery of datasets to S3, SFTP, or any storage target - Direct delivery / zero-copy with Apache Arrow - serve data to models, agents, and applications over REST or Apache Arrow with no interim storage layer to operate Reference: /platform/data-transport ## Four Interfaces, One Governance Layer Dataddo is fully operable through four interfaces, so each kind of operator works the way they work best - and every action available in the UI is available headlessly via API or CLI: - UI - web interface for business teams to build and monitor pipelines without filing tickets to engineering - REST API - programmatic control and automation; embed Dataddo into internal developer portals or existing orchestration tooling - CLI - for engineers in the terminal: scripting, CI/CD, GitOps, and mass deployment of thousands of pipelines - MCP server - native protocol for AI agents (Claude, ChatGPT, Cursor, LangChain, and any MCP-compatible framework) Pipeline configuration can be version-controlled as YAML, templated, and deployed as code - suitable for internal developer portals, multi-tenant managed-service offerings, and config-as-code workflows. Reference: /platform/interfaces ## We Own the Data Contract, Not Just the Connector Building a pipeline is day one; keeping it running is every day after. Source APIs change, schemas evolve, vendors deprecate endpoints. With Dataddo that is Dataddo's problem, not the customer's. - Connector ownership end to end - Dataddo engineers design, build, and continuously maintain every connector; when an upstream source changes its API, auth, or data model, Dataddo updates the connector and restores the pipeline. - Custom connector development - if a needed source or destination isn't covered, Dataddo builds and maintains it under the same ownership model. - Proactive pipeline monitoring - continuous health monitoring detects anomalies, delivery delays, and data-quality deviations before they surface downstream. - Schema drift detection - structural changes are detected automatically and handled by configurable rules (propagate, alert, or quarantine). - Data Quality Firewall - configurable validation that blocks or alerts on bad data before it reaches a destination. - Observability - run histories, payload inspection, error drill-down, source-to-destination traceability, and immutable audit logging. Reference: /platform/data-contract ## Data Foundation for AI Dataddo keeps the storage layer behind AI systems current, validated, and PII-safe. It does not train or host models and does not generate embeddings, so where a vector database is part of the stack it is populated by the customer's own embedding pipeline, reading from storage Dataddo keeps current. Dataddo does serve data directly to models and agents itself, over MCP, REST, or Apache Arrow. Three documented delivery patterns, chosen by how fresh the data must be and how much history the AI needs: - Warehouse, data lake, or object storage - the default. Holds long time-series ranges, so it suits RAG corpora, fine-tuning datasets, and retrieval that reasons over history. Destinations are the systems AI tooling already consumes: BigQuery, Snowflake, Databricks, Amazon S3, Azure Blob Storage. - CDC-replicated operational mirror - for latency-sensitive AI. Log-based CDC keeps a continuously updated replica of the production database that an AI system can query with near-instant freshness, without loading production. Typical case: a support chatbot needing the current state of orders, tickets, or accounts. - Direct retrieval via SmartCache - extracted data is held in SmartCache and retrieved through a REST interface (JSON or CSV) or Apache Arrow, integrated with the MCP server. No storage layer to operate. Trade-off: SmartCache holds recent extractions, not long time-series ranges. The three combine: a chatbot can answer order-status questions from the CDC replica while its knowledge base is re-embedded from the warehouse. Governance applied before data reaches the model: - PII exclusion and hashing at the extraction step, before data leaves the source. - Data Quality Firewall column-level rules (nulls, zeros, anomalies). In blocking mode, failing records never land in the table an embedding job reads, so a broken upstream export cannot silently poison an index. - Row-level technical metadata (a stable natural key, extraction timestamp) plus connector-level metadata - dataset descriptions, field descriptions, and PII/sensitivity flags - giving retrieval and text-to-SQL agents semantic grounding instead of guesswork. - RBAC and immutable audit logging, so AI systems meet the same compliance standards as the rest of the data infrastructure. Stack-agnostic by design: adopt new model providers or AI platforms, or run multiple models in parallel, without rebuilding pipelines. The MCP server gives agents one governed connection rather than a separate MCP server per source. Reference: https://docs.dataddo.com/docs/data-delivery-to-ai-agents Reverse ETL is a separate, live capability covering warehouse-to-application delivery (CRMs, ERPs, ad platforms, operational systems). Dataddo should not be described as routing AI-generated outputs back into operational systems as a productized capability. Reference: /solutions/ai-data-foundation, /solutions/agentic-data-streaming, /platform/interfaces ## Solutions ### AI Data Foundation (/solutions/ai-data-foundation) The ingestion and delivery layer AI depends on. Keeps RAG, agents, and models grounded in fresh, governed, multi-source data. Three delivery modes - warehouse or object storage, real-time CDC replica, or direct retrieval via MCP - with a comparison of freshness, history depth, and storage you operate for each. ### Real-Time CDC (/solutions/real-time-cdc) Log-based change data capture from PostgreSQL, MySQL, SQL Server, Oracle, and MongoDB. Sub-second latency, deletes captured, managed end to end - no CDC infrastructure to build or keep on-call. ### Agentic Data Streaming (/solutions/agentic-data-streaming) Committed data changes delivered to AI agents as they happen. Log-based CDC into event delivery, with no streaming platform to stand up and governance applied before data reaches the agent. ### Database Replication (/solutions/database-replication) Replicate PostgreSQL, MySQL, SQL Server, Oracle, and MariaDB into Snowflake, BigQuery, and other destinations. Fast parallel full loads followed by sync tuned to how each table actually changes. Published benchmarks at docs.dataddo.com/docs/high-performance-batch-database-replication. ### SaaS Data Integration (/solutions/saas-data-integration) 400+ managed connectors from SaaS tools (Salesforce, HubSpot, Google Ads, Stripe, and more) into the warehouse. Multi-account extraction, pagination and adaptive sizing, auto-retry, schema-drift handling, nested-JSON flattening, and scheduling down to one minute. Vendor API changes are Dataddo's responsibility. ### Data Activation / Reverse ETL (/solutions/data-activation) Push curated warehouse data into the tools teams act in - CRMs, ERPs, ad platforms, marketing automation. Insert, update, and upsert, governed and fully managed. ### Single Ingestion Layer for Enterprises (/solutions/managed-data-sync) Batch, real-time CDC, streaming, and direct delivery run from one control plane, governed under one audit trail, wherever the data must stay. Positioned as the consolidation play for enterprises running several point tools. ### SAP Data Replication (/solutions/sap-data-replication) SAP S/4HANA into a warehouse, lake, or AI stack through SAP's own OData services. Read-only over HTTPS, never RFC, never direct HANA access, nothing installed on SAP hosts - compliant with SAP Note 3255746. Supports CDS views and hard deletes. ### Legacy Modernization (/solutions/legacy-modernization) Connect Informix, Oracle, and other foundational systems to a modern data stack without migration or disruption. Includes a large-scale insurer reference (~600M rows, Informix to Azure). ### Legacy URLs still live Earlier packaging predating the current Platform / Solutions structure remains reachable and should be treated as secondary: /use-cases (etl-elt, reverse-etl, database-replication, powering-analytics, headless-data-products, headless-custom-workloads) and /products (data-anywhere, data-to-dashboards, headless-data-integration). ## Comparison: Dataddo vs Azure Data Factory (ADF) Azure Data Factory is a pipeline-building and orchestration framework you operate yourself. Dataddo is a fully managed data integration service that runs and maintains the pipelines for you. How they differ: - Operating model: Dataddo is a managed service (connectors, scheduling, and API-change handling are Dataddo's responsibility); ADF is a framework you author and operate. - Connectors: Dataddo offers 400+ managed connectors plus on-demand connector build and customization, strong on Oracle, SAP, Informix, and business SaaS, with connectors that self-heal after source or schema changes; ADF has about 90 built-in, well-suited to Azure-native services, and connector breakage is the user's to fix. - Reverse ETL: Dataddo writes data back into operational apps (insert, update, upsert); ADF's SaaS connectors are largely read-only. - Speed and scale: Dataddo uses parallel Mesh Ingestion for sub-second CDC and terabyte-scale loads; ADF copy scales with paid Data Integration Units and runs transforms on Spark clusters. - Multi-cloud: Dataddo is cloud-neutral, running on AWS, GCP, Azure, and sovereign clouds such as StackIT, Exoscale, and Hetzner; ADF is Azure-anchored, with its control plane in Azure. - Governance and observability: Dataddo has built-in data quality checks, monitoring, alerting, and an audit trail, and integrates natively with governance and catalog platforms such as Dawiso and Collibra; ADF assembles observability and governance from separate Azure services such as Azure Monitor and Purview. - Pricing: Dataddo is a managed subscription; ADF is metered across activity runs, DIU-hours, and compute you size. When ADF is the right choice: teams that are all-in on Azure, need Azure-native orchestration of complex workflows, have a dedicated Azure engineering team, or are standardizing on Microsoft Fabric. Working together: Dataddo and ADF are complementary. Dataddo can land data from SaaS, databases, and non-Azure clouds into Azure Blob, Data Lake Storage, Azure SQL, Synapse, or Microsoft Fabric, while ADF orchestrates the downstream transformation. Full comparison: /vs/azure-data-factory Other published comparisons: /vs/fivetran, /vs/airbyte, /vs/hevo, /vs/supermetrics, /vs/funnel. Index: /vs/competitors ## FAQ: General ### What use cases does Dataddo serve? Dataddo supports ETL, ELT, reverse ETL, database replication, event-based integrations, and end-to-end integration of online sources with dashboarding apps. It also offers access to a full REST API, so any of Dataddo's data integration functionality can be deployed in a headless scenario. For more details on specific use cases, see our other use case pages: ETL & ELT Reverse ETL (Data Activation) Database Replication Powering Analytics (SaaS apps to dashboarding tools) ### Is Dataddo deployable within the Google Cloud Platform/AWS/Microsoft Azure ecosystems? Absolutely. Dataddo subscriptions can be managed through accounts with any of the three major cloud providers — and the data plane can also run in European sovereign clouds (STACKIT, OVHcloud, Scaleway, Hetzner, Exoscale) or on-premises when data residency requirements demand it. See our: GCP marketplace page AWS marketplace page Azure marketplace page ### How does Dataddo ensure data quality? Dataddo includes several built-in mechanisms for data quality that apply across every workload: The Data Quality Firewall (rule-based, configurable per column) Detailed monitoring and notifications Format harmonization, so data from disparate sources is analytics-ready Data blending/union The ability to exclude personally identifiable information (PII) from extractions Specific workloads add their own controls. Reverse ETL adds flexible write modes, easy data mapping, and syncs as frequent as every 5 minutes. Database replication adds the Truncate Insert write mode, detailed flow logs, and automatic data type conversion. See our documentation for more information about how Dataddo approaches data quality. ### Where can I learn more about Dataddo's SOC 2 Type II certificate? Consult the Dataddo SOC 2 Type II Certificate . ## FAQ: Platform ### What is Dataddo? Dataddo is an enterprise data movement platform. A single control plane orchestrates data planes that run where your data lives - in your cloud, on-premises, or both - so data moves as managed infrastructure without being routed through a vendor's cloud. Dataddo was founded in 2019 and is headquartered in Prague, Czech Republic. Explore the Dataddo platform . ### What is the difference between Dataddo's control plane and data plane? Management - scheduling, configuration, and monitoring - runs in Dataddo's cloud control plane. The data movement itself runs in a data plane. By default, Dataddo operates as a fully managed cloud-to-cloud service, moving data directly between cloud sources and destinations. When data must stay in place, the data plane can instead execute inside your own environment - your cloud, on-premises, or both - so sensitive workloads never traverse public infrastructure. The same control plane orchestrates either model. Learn more about a single control plane with multiple data planes . ### What types of data movement does Dataddo support? Dataddo supports ETL, ELT, CDC, streaming, reverse ETL, batch file delivery, and zero-copy Apache Arrow - all in one platform. See all data transport types . ### Can I use Dataddo through a UI, API, CLI, or MCP? All four, with one governance layer. Every action available in the web UI is also available via REST API and CLI, so pipelines can be version-controlled as YAML and deployed as code (CI/CD, GitOps). An MCP server gives AI agents native access. See UI, API, CLI, and MCP . ### How many connectors does Dataddo have, and what if mine isn't supported? Dataddo maintains 400+ connectors across SaaS APIs, databases, files, and object storage. Dataddo owns the connector contract end-to-end, so API changes, schema drift, and connector maintenance are Dataddo's responsibility, not yours. If you need a connector that does not exist yet, Dataddo builds custom connectors, typically within about four weeks. Browse the connector catalog . ### How does Dataddo get data into AI systems? Dataddo keeps the storage your AI systems read from current, validated, and PII-safe. There are three patterns: a warehouse, lake, or object storage for RAG corpora and fine-tuning; a real-time CDC replica for latency-sensitive cases, such as a support chatbot that needs live order state without loading your production database; or direct retrieval through the MCP server, REST, or Apache Arrow, where an agent gets the latest extracted data with no storage layer to operate. Governance is applied before data lands - PII is excluded or hashed at extraction, and the Data Quality Firewall can block failing records outright. Dataddo does not train or host models and does not generate embeddings; where a vector database is part of your stack, your own embedding pipeline populates it from the storage Dataddo keeps current. See how Dataddo is operated via UI, API, CLI, and MCP . ## FAQ: Evaluation ### What should I evaluate when choosing a data integration platform? A useful evaluation goes beyond the connector count. The questions that separate platforms in practice are: Coverage across source types. Can one platform handle SaaS APIs, databases (with CDC), files, and object storage, or will you need separate tools that each fail differently? Who maintains the connectors. When a source API changes or a schema drifts, is that the vendor's responsibility or yours? Unmaintained connectors are the hidden cost of most "self-serve" tools. Deployment model. Does your data have to transit the vendor's cloud, or can the platform run inside your own environment when residency or security rules require it? Security and compliance. SOC 2 Type II, ISO 27001, GDPR alignment, encryption, role-based access control, and audit logging - confirmed, not just claimed. Governance and auditability. Can you show where a piece of data came from and who accessed it? Pricing predictability. Does cost scale with data volume and frequency, or is it predictable as you grow? AI readiness. Can the platform deliver governed data into the systems AI depends on, and expose data to agents through a controlled interface? Dataddo is built to answer each of these. See the Dataddo platform . ### What makes Dataddo a good fit for large enterprises? Dataddo is an enterprise data movement platform designed for organizations that treat data pipelines as managed infrastructure rather than self-serve tooling. Four things make it suited to enterprise scale: One layer for every source and destination. SaaS APIs, databases with change data capture, files, and object storage move through a single platform, so teams stop stitching together tools that each break differently. It runs in your environment. A single cloud control plane orchestrates data planes that execute where your data lives - in your cloud, on-premises, or both - so sensitive data does not have to transit a vendor's cloud. Dataddo absorbs the maintenance. API changes, schema drift, and connector upkeep are Dataddo's responsibility, backed by SLAs and a dedicated solutions architect for enterprise accounts. It passes the security review. SOC 2 Type II, ISO 27001, GDPR alignment, encryption with bring-your-own-key support, role-based access control, and audit logging. Explore the Dataddo platform . ### Should we build our own data pipelines instead of buying a tool? Writing a script for one integration is easy. The cost shows up later, and it is rarely one-off. Each source you connect by hand becomes something your team has to maintain forever: APIs change without notice, schemas drift, authentication tokens expire, and rate limits shift. With a handful of sources this is manageable; across dozens it becomes a recurring engineering job that competes with the work you actually hired those engineers to do. Buying a managed platform changes who owns that maintenance. With Dataddo, API changes, schema drift, and connector upkeep are handled on Dataddo's side - your team configures pipelines and keeps control, without owning the break-fix work. Custom connectors for sources that do not exist yet are built by Dataddo, typically within about four weeks. The honest question is not whether the integration is unique enough to justify building — it is whether you have a person whose job it is to maintain it when things break. If the answer is no, you are not building a solution; you are creating an on-call incident waiting for an owner. Read more about the Dataddo platform . ### Who maintains my data pipelines when a source API changes? Dataddo does. This is the core of how Dataddo differs from self-serve tooling: Dataddo owns the connector contract end to end, so when a source API changes or a schema drifts, fixing it is Dataddo's responsibility, not your team's. In most cases API changes are handled before they affect live pipelines; schema changes are detected and absorbed automatically so pipelines do not break when sources change. Concretely, that means your team is not on call for someone else's API. You get: Proactive monitoring of connections, with notifications when something needs attention Automatic handling of source schema changes SLA-backed incident response, with a dedicated solutions architect for enterprise accounts Custom connector builds for sources that do not exist yet, typically within about four weeks This is why customers describe Dataddo as a pipeline team they do not have to hire - the operational burden moves to Dataddo while control stays with them. ## FAQ: Sovereignty ### Where is my data processed - does it leave my environment? It depends on how you deploy. For fully managed cloud-to-cloud pipelines, data moves directly between your sources and destinations. For sensitive workloads, the data plane can execute inside your own environment, so your data stays within your perimeter and does not pass through Dataddo's cloud. Learn more about control plane and data planes . ### Where can Dataddo be deployed? Anywhere your data needs it to run. Dataddo deploys in any public cloud (AWS, Azure, Google Cloud), in European sovereign and regional clouds (such as STACKIT, OVHcloud, Scaleway, Hetzner, and Exoscale), on-premises, or in a hybrid setup with multi-region agents. It is portable across Kubernetes, OpenShift, and VMware Tanzu. Both the data plane and the control plane can run in the EU region or provider you choose. See the versatile platform architecture . ### Does Dataddo support EU data residency and GDPR requirements? Yes. Dataddo is GDPR-aligned, and because the data plane runs in the region you choose, data can stay within your jurisdiction to meet EU data residency requirements. The control plane can also be EU-hosted, so both planes can run under EU law. Learn more about control plane and data planes . ### Does Dataddo meet EU data sovereignty requirements? Yes. Dataddo is built by an EU-domiciled company (HQ: Prague, Czech Republic) with no foreign parent, and it is structured so your data stays under EU control end to end. The data plane runs in your own environment, so data never transits Dataddo's cloud. The control plane can also be EU-hosted, keeping both planes under EU jurisdiction. As an EU legal entity with no non-EU parent, Dataddo cannot be compelled to grant foreign access to your data. This maps directly to SOV-3, the data and AI criterion in the EU Cloud Sovereignty Framework : pipelines must be developed, hosted, and governed under EU control. Dataddo meets all three. It carries no SEAL (Sovereignty Effectiveness Assurance Level) rating of its own, because it is not a cloud provider — instead it preserves the SEAL level of whatever cloud you run it on. This is where US-based integration tools fall short: they route your data through their own US-controlled clouds, capping your effective sovereignty at the vendor's SEAL level no matter how sovereign your destination is. Because Dataddo keeps the data plane in your environment, it removes that ceiling — and the same architecture positions you for the Cloud and AI Development Act (CADA) , proposed June 2026, which will require data pipelines to be developed, hosted, and governed under EU control. Read more about built for high-security environments . ### Is Dataddo suitable for regulated industries? Yes. Dataddo is used by enterprises in banking, insurance, healthcare, and the public sector, where the in-environment data plane combined with SOC 2 Type II, ISO 27001, and GDPR alignment meets strict regulatory requirements. Read more about built for high-security environments . ## FAQ: Pricing ### How is Dataddo priced? Dataddo's standard pricing is based on the number of data flows. A data flow is the connection between a data source (or sources) and a destination - for example, sending data from Facebook Ads to Google BigQuery counts as one flow. This keeps budgeting predictable: costs do not escalate with data volume, extraction frequency, or number of data sources. Headless and embedded deployments are priced individually using integration units, which account for rows and actions per month, per connector. For current tiers and enterprise pricing, see our pricing page . ## FAQ: Powering Analytics ### What are the benefits of integrating data with Dataddo? - No more manual CSV uploads. Automate all your data connections. - No more switching between platforms to see data. By syncing data from all your platforms to an analytics tool, you can monitor important cross-platform metrics from a single place. - Analytics-ready data. Dataddo automatically unifies the format of data from your various apps, so that it's ready to analyze by the time it gets to your dashboard. - Long-term scalability. Dataddo is a comprehensive, any-to-any data integration tool designed to meet the needs of any professional or organization that works with data—from solo marketers to global enterprises. This means you can start by using it to send data from online services to analytics tools and then, when your organization is ready, use it to centralize data in a warehouse, replicate data between databases, or send data from warehouses into business apps like CRMs and marketing automation platforms. A central screen for managing all data connections, multiple users per account, and multi-tenant deployment makes it easy for various departments and teams to adopt Dataddo for their own use cases. - Predictable pricing. Always know what you're paying and never get a bad surprise at the end of the month. Since Dataddo's pricing is based on number of data connections (i.e., flows) costs do not escalate with data volume, extraction frequency, or number of data sources. - Proactive pipeline monitoring and maintenance. Our engineers proactively monitor connections and manage all API changes behind the curtain. This means you don't have to worry about your connections breaking in the middle of the night. - Inbuilt data quality tools. Dataddo automatically unifies the formats of data it sends to dashboarding tools, so your data will always be ready to analyze. We also enable rule-based monitoring and data quality checks, to help you prevent inaccuracies and errors in any data you transfer with Dataddo. - Connects any A to any B. Never worry that you might be "stuck" with a tool that can't connect to one of your services. In addition to offering a massive connector portfolio of apps and databases, we build new connectors for clients all the time. ### How does Dataddo compare to Supermetrics? See our Supermetrics comparison page for the full breakdown. ## FAQ: Reverse Etl ### What are the benefits of reverse ETL? Generally speaking, reverse ETL gives business teams like sales and marketing insights from information that only organizations know about themselves. To illustrate, let's look at CRMs. CRMs collect a lot of customer data. Payment amounts, support tickets, acquisition information—the data's there. But it's not all there. Companies have their own specific way of calculating certain metrics, in particular metrics based on first-party data, and it's difficult for CRMs to display these without heavy—and costly—customizations. Imagine your company sells software as a service, and that customers with three or more user accounts tend to have a lower risk of churn. There is no way your CRM could know this straight out of the box. But, if you have the right customer data in your data warehouse, your engineers can run computations there, then send the risk scores back into your CRM via a reverse ETL tool, making them visible to your sales and support teams alongside all other customer data. With little to no effort, these teams will then be able to identify who is at risk of churn. This is why reverse ETL is often referred to as the "last mile" of the modern data stack—it enables organizations to display any information in any business app. ## FAQ: Database Replication ### What are the benefits of database replication? Generally speaking, database replication helps keep data accessible across locations and platforms for the whole organization. More specifically, it is used for: Analytics. Replicating data from a production database to a data warehouse, i.e. a safe sandbox for analytics teams. Data migration. Simplifies the data migration processes during system upgrades or transitions to new database platforms. Disaster recovery. Redundancy and backup solutions ensure data availability and reliability in case of system failures or disasters. Data consistency. Keeps data consistent across databases and other systems, preventing discrepancies and ensuring uniformity. Performance optimization. Distributes data load across multiple databases, improving read performance and reducing latency. Improving scalability. Distributing data across multiple servers or cloud environments enables systems to scale better. Automated real-time data sync & updates. Keeps data automatically updated and synchronized in real-time and between systems, reducing manual efforts and errors. ## FAQ: Data Security Soc2 ### What is the Dataddo SOC 2 report? Dataddo has obtained an SOC 2 Type II report for its data integration platform, which outlines the security controls implemented for the platform and evaluates their appropriateness and effectiveness in meeting the AICPA Trust Service Criteria. The report serves as an independent assessment of Dataddo's ability to manage data with regard to security, availability, and confidentiality. ### Which Dataddo services are covered by the SOC 2 Type II report? The scope of the SOC 2 Type II report includes all products and features within the Dataddo platform. ### What regions are covered by the Dataddo SOC 2 Type II report? The report covers all regions in which Dataddo is available for use. ### Who performs the independent third-party audit of Dataddo for SOC reports? BDO Czech Republic performs the Dataddo SOC 2 audits. ### How often are Dataddo SOC 2 audits performed? Annually. ### Is an NDA required to receive Dataddo SOC reports? Yes, an NDA is required to review the Dataddo SOC 2 Type II report. Please contact us to begin the process. ## FAQ: Security ### What security certifications and compliance does Dataddo hold? Dataddo is SOC 2 Type II certified, ISO 27001 certified, GDPR-aligned, and PCI DSS compliant. Read more about built for high-security environments . ### How does Dataddo encrypt data and control access? Dataddo uses end-to-end encryption with bring-your-own-key support (AWS KMS, Azure Key Vault, HSM), network isolation, and PII detection, masking, and tokenization at ingestion. Access is governed by SSO (SAML 2.0, OIDC), role-based access control, and immutable audit logging. Read more about built for high-security environments . ### Does Dataddo provide data lineage and audit logging? Yes. Dataddo gives you end-to-end traceability for the data it moves. Every flow records its source, its destination, and its run history, so you can show where a piece of data came from and how it reached its destination. Access is recorded through immutable audit logging - which identity or agent accessed which data, and when - and detailed flow logs and monitoring track each pipeline run. For data exposed to AI agents, access is governed per token and tied to specific SQL transformations of a data flow, so the audit trail captures exactly what each agent was permitted to see. These controls fall within the scope of Dataddo's SOC 2 Type II report. Read more about built for high-security environments . ### What governance controls apply when adding a new data source? Adding a source in Dataddo happens within the same governance layer that covers every pipeline, so a new connection does not open a new gap. The controls that apply: Access control. Single sign-on (SAML 2.0, OIDC) and role-based access control determine who can create and manage the connection. Scoped credentials. Tokens can be tied to specific SQL transformations of a data flow, giving effectively row- and column-level control over what each consumer or agent can read. Sensitive data handling. Personally identifiable information can be detected, masked, tokenized, or excluded at ingestion, before it reaches any destination. Audit trail. The new flow is recorded with immutable audit logging and run-level flow logs from its first execution. Data quality checks. The rule-based Data Quality Firewall can be applied per column as data starts flowing. Read more about built for high-security environments . ## FAQ: Headless Data Products ### Who should use Dataddo's Headless Data Integration for data products? Any organization building a data product whose main focus is to generate insights from data; for example, CDPs or data analytics platforms—these need data from various sources to work properly, but their main functionality is analytics. By connecting to the unified Dataddo API, such organizations can put all of Dataddo's integration functionality under the hood of their product, and focus instead on developing the product's insight-generation functionality. Doing this will: Shorten time to market Improve market adaptability Save money and engineering resources ## FAQ: Headless Custom Workloads ### Who should use Dataddo's Headless Data Integration for custom workloads? Any organization that wants to go beyond our user interface to achieve more control over data integrations — for example, to automate repetitive jobs like historical data loads, or to implement client-specific configurations. Doing this via the Dataddo API allows you to: Automate any complex or repetitive integration workload Implement custom authentication and authorization flows Subscribe and manage Dataddo via AWS, Azure, or GCP marketplaces ## Pricing Pricing details are maintained on one page and are not duplicated here, so that this reference cannot go stale against it. A free trial is available without a credit card, and enterprise deployments (cloud, hybrid, or on-prem) are quoted individually. For current plans, inclusions, and the metering model, see https://www.dataddo.com/pricing - that page is authoritative. ## Supported Connectors (sample) Sources: Salesforce, HubSpot, Pipedrive, Google Analytics 4, Facebook Ads, Instagram Ads, LinkedIn Ads, Google Ads, Microsoft Ads, Shopify, WooCommerce, Stripe, Recurly, Zendesk, Freshdesk, Jira, GitHub, Snowflake, BigQuery, PostgreSQL, MySQL, SQL Server, Oracle, MariaDB, MongoDB, Informix, SAP S/4HANA (via OData), and 400+ more. Destinations: Snowflake, Google BigQuery, Amazon Redshift, Databricks, Microsoft Fabric, Azure SQL Database, Azure Synapse, PostgreSQL, MySQL, SQL Server, Amazon S3, Azure Blob Storage, Google Sheets, Looker Studio, Power BI, Tableau, and business applications including HubSpot and Klaviyo. First-class support for open table formats: Apache Iceberg, Delta Lake, Apache Hudi. Full list: https://www.dataddo.com/connectors - authoritative catalog at https://docs.dataddo.com/docs/available-connectors ## Security & Compliance - SOC 2 Type II certified - ISO 27001 certified - GDPR aligned - PCI DSS compliant - EU data residency available; the data plane can run entirely inside the customer's own environment - Network isolation - sensitive data can be processed in a fully isolated Data Plane in the customer's private cloud or on-premises environment, never traversing public infrastructure - End-to-end encryption in transit and at rest; bring your own keys via AWS KMS, Azure Key Vault, or HSM - PII detection, masking, and tokenization at ingestion - SSO (SAML 2.0, OIDC), role-based access control (RBAC), immutable audit logging, source-to-destination traceability - Recognized by Gartner: Gartner Cool Vendor; Honorable Mention in the 2024 Gartner Magic Quadrant for Data Integration Tools - Rated on G2 (4.7), Capterra (4.4), GetApp (4.4), Gartner Peer Insights Reference: /platform/security, /platform/data-security, /platform/data-security/soc2, https://docs.dataddo.com/docs/security ## Case Studies Some studies are anonymized at the customer's or Marketing's request; where a company is not named, it should not be identified. - Global CPG (anonymized, beauty and consumer goods): Dataddo deployed inside the customer's own GCP environment to automate pipelines across 65,000+ brand social accounts, with zero manual authorization work - Wärtsilä (industrial manufacturing): UI and API used together to consolidate marketing data - Sensire (healthcare): migration of a proprietary on-premise data infrastructure to the cloud, saving up to 3 man-days monthly - Livesport (real-time sports data): high-volume, low-latency data movement, saving 3-4 full-time engineering equivalents - Publicis Groupe Brasil (advertising): headless REST API embedded into an internal data product, saving 3 FTEs and 2+ months of feature development - Ringside.ai (marketing analytics): ELT and reverse ETL underpinning an attribution platform, saving 2-3 developer equivalents - KAARISMA: manual data management eliminated in support of an AI strategy, saving 1 FTE and 90% of time spent on data - Boldr: 177 pipelines optimized for reliable internal reporting - FoodChéri (food delivery): MongoDB replicated to BigQuery for faster insights with less engineering - ID&T Group (entertainment): data from 1M+ festival fans and dozens of social accounts, saving 2.5 man-days per week - WWL (lighting): sales, marketing, and customer data unified from 18 sources across 24 eshops - Uber Eats: 50+ social and marketing accounts unified, 50+ man-hours per week saved - Search Engine Journal (media): customized advertiser dashboards without manual extraction - Sat-7 (broadcasting): hundreds of social media accounts consolidated into one dashboarding app - Ogilvy Spain (advertising): multi-account social media reporting for agency clients Full list: https://www.dataddo.com/case-studies