Hightouch started as reverse ETL: a way to sync data from a company's warehouse into the business tools that need it. It now describes itself as an agentic marketing platform built on a composable CDP, and in April 2026 it raised a $150 million Series D at a $2.75 billion valuation.

This deep dive starts from Hightouch's integrations as the CdpHub directory records them: 346 documented integrations from its catalog and its documentation, each with its direction and a link to Hightouch's page. It then turns to the products Hightouch has built around them, using Hightouch's own documentation.

Hightouch by the numbers

346
documented integrations with 296 tools
300
destination listings to 281 tools, 68 of them in advertising
46
inbound listings, 27 of them warehouse, database, lake, file and spreadsheet sources
$150M
Series D in April 2026, at a $2.75 billion valuation

The catalog: the warehouse in, everything out

95 percent of Hightouch's listings are warehouse integrations as How we count defines them: reverse ETL and data sharing, and reads from or writes to warehouses, databases, lakes and files. The other 15 listings are stream sources and a OneTrust consent plugin for Hightouch Events, message sends for its AI Decisioning agents, and hooks into dbt, Fivetran, Datadog, PagerDuty and Slack. Of these, only the OneTrust plugin runs in the browser.

46 ways in. Hightouch reads 27 warehouse, database, lake, file and spreadsheet sources: Snowflake, BigQuery, Databricks, Redshift, Azure Synapse, Microsoft Fabric, ClickHouse, Teradata Vantage, Trino, Amazon Athena and Greenplum; operational databases such as PostgreSQL, MySQL, SQL Server and Oracle; Iceberg tables on Snowflake and BigQuery; the streaming database Materialize; and files and spreadsheets in Amazon S3, Google Cloud Storage, Azure Blob Storage, SFTP, Google Sheets and Excel. Models can also come from dbt, Looker and Sigma. Hightouch Events reads event streams from Kafka, Confluent Cloud, Redpanda, Kinesis, SQS and Pub/Sub, and a data extraction feature writes six ad platforms' daily audience-size estimates back to the warehouse. OneTrust consent data comes in through OneTrust's Snowflake Native App or the Events plugin, and Fivetran and dbt Cloud jobs can start syncs.

300 destination listings to 281 tools. Advertising is the largest group (71 listings to 68 tools), followed by Data Infrastructure (46 listings to 42 tools), Email & Marketing Automation (44 to 38) and CRM & Sales (32 to 30). Sixteen tools have more than one destination listing, among them Meta (custom audiences and the Conversions API), Braze (users, cohorts and AI Decisioning) and Snowflake (tables, Iceberg tables, data shares and a clean room).

The largest warehouse-native catalog in the directory. Hightouch's 346 listings compare with 195 for Fivetran Activations (formerly Census) and 132 for DinMo, and 331 of them are warehouse integrations, against 194 and 131. More than 30 of its tools appear on no other platform, among them sales tools, ad platforms such as Expedia Group Advertising and Epsilon Retail Media, and back-office software such as QuickBooks, Sage Intacct and SAP IBP.

Apps come in through Connections or a loader, not the catalog. Apart from the six ad platforms' audience-size reports and OneTrust consent data, no CRM, commerce or other operational app appears as a Hightouch source in the directory: Salesforce, HubSpot, Shopify and the like are destinations only. Hightouch documents a Connections feature that imports data from apps such as Salesforce, Klaviyo and Braze into the warehouse, but it doesn't publish a list of supported apps, so check the ones you need or plan for a data loader.

Where Hightouch's catalog meets the others

  • Fivetran Activations is the closest match: the two share 143 tools, 43 percent of all the tools the two document, the largest overlap and the most similar pair of catalogs in the directory.
  • Packaged CDPs overlap less: Hightouch shares 116 tools with Tealium, 104 with Segment and 85 with mParticle, and 110 with RudderStack, the hybrid.
  • Its destinations are mainstream. 83 percent of the destination tools Hightouch documents are documented by at least one other platform, and it documents 24 of the 25 tools in the CDP Market Landscape 2026 report's most-connected chart as destinations, all but Google Tag Manager.
  • Other platforms connect to it too. Six of the directory's platforms document an integration with Hightouch: Braze (reverse ETL, cohort import and the Personalization API), Bloomreach, CleverTap, Insider One, Optimizely and mParticle.

What Hightouch has built on top of reverse ETL

The integration catalog is only half of the product now. According to Hightouch's documentation:

  • Customer Studio (Business tier) lets marketers build audiences from warehouse data without SQL, alone or with an AI agent's help, sync them to destinations, and run multi-step journeys. Data teams define the schema it works from.
  • Identity Resolution builds unified profiles in the warehouse, with deterministic matching on identifiers and AI-based probabilistic matching on top.
  • Hightouch Events collects behavior with Hightouch's own browser, mobile and server SDKs or reads it from event streams, checks it against data contracts, and writes it to your warehouse or streams it to destinations.
  • Real-Time Personalization evaluates audiences on live events combined with warehouse data, typically in less than a second, and sends membership changes through realtime syncs. A separate Personalization API serves warehouse-backed customer data from a managed cache.
  • AI Decisioning runs agents that learn which message, channel and timing work best for each person, within the goals and guardrails you set, and sends the messages through Braze, Iterable or Salesforce Marketing Cloud, or through a custom channel to other destinations.
  • Match Booster, a Business-tier add-on, adds identifiers from Hightouch's licensed third-party data, such as alternate emails, phone numbers or device IDs, to the audiences and conversions it sends to ad platforms, to raise match rates. It currently covers the United States, with the EU in early access.

In 2026 Hightouch has added AI products for content, ads and lifecycle marketing (Content Assembly, Ad Studio and Lifecycle Studio). In January 2026 it announced that Gartner had named it a Leader in the 2025 Magic Quadrant for Customer Data Platforms, its first inclusion.

Where Hightouch fits, and where it doesn't

It fits a warehouse-first data team. If customer data is already modeled and governed in the warehouse, Hightouch extends that work into activation without a second copy of the data: by default, its sources are queried in place.

Real time is documented, with limits. Realtime audiences evaluate live events as they arrive, but when an audience also uses warehouse conditions, Hightouch checks those against a cache that refreshes on a schedule you set, and realtime audiences don't currently support traits, experiments or journeys. Test the use cases that matter to you before counting on in-session decisions.

Identity is a product, but it lives in your warehouse. Hightouch's Identity Resolution builds profiles from the data in your warehouse, which keeps it transparent and under your governance. It also means the results are only as good as the data your team has loaded and modeled there.

Marketers still need a data team behind them. Customer Studio is built for marketers, but the schema, models and traits come from the data team, and Customer Studio requires the Business tier.

It isn't a tag manager. Hightouch Events covers first-party collection, but the catalog documents no vendor tags. A team that deploys and governs dozens of vendor pixels still needs a tag manager next to it.

How it compares

Among the composable platforms in the directory, Hightouch has the largest catalog. Beyond activation, it documents event collection, real-time personalization, identity resolution and AI decisioning. Fivetran Activations, which was Census until it joined Fivetran, sits next to Fivetran's own data loaders, and DinMo loads some app and ad data itself.

Against packaged CDPs such as Segment, Tealium and mParticle, Hightouch competes on architecture rather than catalog size. All three carry the client-side layer. Segment (555 listings with 459 tools) and Tealium (886 with 636) also document more, while mParticle's catalog (326 with 205) is smaller than Hightouch's. With Hightouch's own event collection and real-time audiences, the gap between the camps has narrowed. What still divides them is tag management and where the profile lives: in the vendor's store or in yours.

The bottom line

Hightouch is no longer just a reverse ETL tool. Its catalog is the largest warehouse-native one in the directory, mainstream where it matters and with a long tail of sales, advertising and back-office tools, and it has built events, identity, real-time personalization and AI decisioning on top. It fits best in organizations that have already made the warehouse their customer data foundation and have a data team to maintain it. Teams that need a tag manager, or a profile store they don't have to build, should compare it with a packaged CDP on their own stack.

How we count

  • Integration counts are every integration with a third-party tool that Hightouch documents on a public page, in its integrations catalog and its documentation, as of October 2026, from the CdpHub directory. Hightouch's own SDKs, APIs and features (such as the Hightouch Events SDKs, the Personalization API and its toolkit for building custom destinations) are not counted, and neither are Connections imports, because Hightouch names supported apps only as examples.
  • Run types: client-side means a tag, pixel, SDK or plugin in the site or app carries the data; server-side means the platform's servers exchange data by API, webhook, stream or file; warehouse means the tool is a warehouse, lake, bucket or database, or the integration is reverse ETL or data sharing.
  • Overlap counts compare distinct tools as the directory groups them. Per-platform shares are rounded down, as in the directory's tables, and the similarity of two catalogs is rounded to the nearest point. Counts change as catalogs change; the live directory is the reference.

Sources