Product update: Google Analytics and Google Ads connectors
Datachef now connects to Google Analytics 4 and Google Ads, so ad spend, site behavior and revenue can finally sit in the same tables.
Joe Fusaro
Founder, Datachef
Table of Contents
- What's new?
- What do you get from Google Analytics?
- What do you get from Google Ads?
- Why does it matter to have both?
- How do you set them up?
What's new?
Datachef now has connectors for Google Analytics 4 and Google Ads (yes, the artist formerly known as AdWords).
They sit alongside Salesforce, Marketo, Postgres and MySQL. Once connected, your GA4 events and your Google Ads campaigns are tables in the same place as your CRM, marketing automation and product data, so you can query, model and visualize them together.
What do you get from Google Analytics?
Event-level data, not reports.
GA4's reporting API only hands back aggregates. They're capped, sampled over long date ranges, and lump rare values into "(other)". So Datachef reads GA4's BigQuery export instead, which is the raw event stream: every page view, form submit and custom event, with its parameters.
What lands in Datachef:
- Events - one row per event, with device, geo and traffic source flattened into their own columns.
- Event parameters and items - as separate tables you can join back to events, instead of one big nested blob.
- Intraday and user tables - if you've turned those exports on in GA4.
GA4 keeps adding late events to a day for up to 72 hours, so an incremental sync waits for a day to settle before reading it, and never counts the same event twice. A full refresh (which is how the intraday table syncs) doesn't wait, because it replaces the whole table.
What do you get from Google Ads?
Your account structure plus daily performance:
- Reference tables - campaigns, ad groups, ads and keywords (positive and negative), as they're configured today.
- Daily performance - impressions, clicks, cost, conversions and conversion value by day, at the account, campaign, ad group, ad and keyword level.
Google keeps revising recent days (invalid clicks get removed, conversions get credited back to the day of the click), so you can choose between a full refresh that always matches what Google says now, or an incremental sync that waits a few days for the numbers to settle.
Why does it matter to have both?
Because neither one tells you whether an ad made you money.
Google Ads knows what you spent and what got clicked. GA4 knows what people did on your site after the click. Your CRM and marketing platforms know how prospects researched, engaged and ultimately purchased. Each system holds one piece to the puzzle, but getting them all together is where real value can be unlocked.
A few examples:
- Cost per opportunity by campaign - Google Ads spend, joined to GA4 sessions by campaign, joined to leads and opportunities in Salesforce or HubSpot.
- Keywords that bring in real buyers - Google Ads keyword performance next to which of those visitors went on to sign up in your product database.
- Landing pages that convert - GA4 page views and form submits by campaign, next to which leads Marketo scored as qualified.
That's click-to-cash reporting. The hard part was never the SQL, it was getting all of this into one place. If you want to see how to split the revenue across campaigns, I walked through it in How to create a multi-touch attribution model.
Datachef is a full-stack data platform (ETL, storage, modeling, visualization, reverse ETL and AI), so once the data is in, you can build the model, chart it, and ask the agent about it in natural language without bolting on more tools.
How do you set them up?
- Google Analytics 4 - link your GA4 property to BigQuery (Admin, Product links, BigQuery links), then give Datachef a service account that can read the export. The export isn't backfilled, so link it before you need the history. Step-by-step in the GA4 docs.
- Google Ads - you'll need your Customer ID and an OAuth client from your own Google Cloud project with Google Ads API access. The connection page walks you through it, and there's a setup guide you can pass to whoever owns your Google Cloud project.
Both run on a schedule after that, and you choose which tables and columns to sync.
Related reading
- Recipe: Click-to-cash reporting
- How to create a multi-touch attribution model
- What is a marketing data platform?
Want to see it on your own data? Watch a demo, or try Datachef free for 30 days.