Question 1

What can a ClickHouse model write into Salesforce?

Accepted Answer

Standard objects such as accounts, contacts, leads, opportunities and cases, plus your custom objects, subject to what the connected user can edit. In practice this direction is mostly used to put computed values onto records that already exist, an engagement score on an account or a usage figure on a contact, rather than to create records wholesale, since Salesforce is usually where records are born.

Question 2

How does Polytomic connect to ClickHouse?

Accepted Answer

Over the native TCP protocol using the official ClickHouse Go driver, not over HTTP. There is no HTTP mode, so the port matters: normally 9440 with TLS or 9000 without, and pointing at the HTTP port simply will not connect. This is the single most common setup mistake on this connector. ClickHouse Cloud is the same story, its native endpoint with your ClickHouse credentials rather than a bespoke authentication path.

Question 3

How does Polytomic decide which record to update?

Accepted Answer

You map one column from your model to an identity field on the Salesforce object, commonly an email for contacts or an external ID on accounts. Polytomic looks the record up on that pair, treating a match as an existing record to update and anything unmatched as new. Exact equality is the default. Create, update, and update or create modes all need that identity mapping in place before they will run.

Question 4

Will this consume our Salesforce API quota?

Accepted Answer

It uses some, but the write path is designed to stay within your allowance rather than exhaust it. The more useful thing to plan is volume: pushing a score onto every account every hour is a lot of API calls for data that changes daily. Setting a tracking column on the ClickHouse model so only genuinely changed rows are sent is the practical way to keep both the API usage and the query cost proportionate.

Question 5

Why send behavioural data to the CRM?

Accepted Answer

Because reps work in Salesforce and will not go elsewhere to check a number. Product usage lives in whatever holds your events, which for high volume products is often ClickHouse, and it stays invisible to the sales team until something moves it. Reverse ETL is that something, and it keeps the CRM field tied to a model rather than to a spreadsheet somebody updates when they remember.

Sync ClickHouse to Salesforce

Sync ClickHouse to Salesforce with Polytomic

Standard and custom objects

Native protocol reads

No storage on the read side

Or sync Salesforce to ClickHouse

Trusted by businesses worldwide

Step-by-step integration guide

Connect ClickHouse in Polytomic

Connect Salesforce in Polytomic

Create a sync between ClickHouse and Salesforce

Map fields and enable the sync

Frequently asked questions

The sync and ETL platform that does it all

Powerful data sync

Advanced analytics

Enterprise security

Easy setup

Ready to connect ClickHouse with Salesforce?