Question 1

What can I write into Microsoft Ads from ClickHouse?

Accepted Answer

Custom audiences, where each row the model returns becomes a member, and offline conversion events for conversions that completed away from the click. Audiences are where most teams start. Conversions become useful when the outcome you care about is deep in the product or in billing, which is exactly what a ClickHouse model is usually built from.

Question 2

Can a suppression list contract on its own?

Accepted Answer

Yes, using mirror mode. Add mode only ever adds and remove mode only removes, but mirror reconciles the audience against your model on every run. Microsoft supports it where several other platforms do not. One thing to watch on this particular source: the incremental read cannot see rows that disappear from ClickHouse, so if membership depends on rows being deleted rather than marked, mirror will not have the information it needs.

Question 3

Which identifiers can I match on?

Accepted Answer

Email only for audiences here, which is worth knowing while modelling. Meta and TikTok also accept phone numbers and Reddit takes a mobile advertising ID, but the Microsoft path is email based. That caps your audience at the people you hold a good address for, so if match rate matters it is usually worth improving address quality in the model rather than expecting the platform to compensate.

Question 4

How often should this run?

Accepted Answer

Daily is usually right, for two reasons. Audience membership rarely shifts meaningfully within a day, and every run is a genuine query against your cluster. If you do want it tighter, put a tracking column on the model so runs read a bounded window rather than rescanning, and keep the selected columns to the identifiers you actually map. That keeps the cost proportionate to what the platform can act on.

Question 5

What goes wrong with hand managed audiences?

Accepted Answer

They decay without anyone noticing. A list uploaded in spring still excludes people who lapsed and became worth targeting again, and still includes people who bought months ago, while the campaign runs on regardless. Tying the audience to a model means it is rebuilt on every sync from the same definitions your reporting uses, so the drift never accumulates in the first place.

Sync ClickHouse to Microsoft Ads (Bing Ads)

Sync ClickHouse to Microsoft Ads (Bing Ads) with Polytomic

Mirror keeps it honest

Email based matching

Conversions too

Or sync Microsoft Ads (Bing Ads) to ClickHouse

Trusted by businesses worldwide

Step-by-step integration guide

Connect ClickHouse in Polytomic

Connect Microsoft Ads (Bing Ads) in Polytomic

Create a sync between ClickHouse and Microsoft Ads (Bing Ads)

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 Microsoft Ads (Bing Ads)?