What is a marketing data warehouse?
A marketing data warehouse is a database that collects data from your ad platforms, analytics and sales systems in one place and in one agreed format, so campaigns can be judged against cost and margin instead of the revenue each platform reports for itself. For an online store it is the same build as an ecommerce data warehouse, and the only difference is which sources come first. Ours runs on Google BigQuery, in your own account.
Can you build a data warehouse for a company outside the EU that sells in Europe?
Yes, we build data warehouses for companies outside the EU that sell in Europe. We work from Kraków, in the EU time zone, and in English. The warehouse sits in your own Google Cloud account, so you hold the agreement with Google and choose where the data is stored: BigQuery offers an EU multi-region and individual European regions, and we settle the location with you in the source audit. In the same audit we agree how to treat currencies, VAT and returns for each country, because a profit figure for a store selling in several markets depends on those definitions.
Why does a store need a data warehouse if it has GA4?
GA4 describes user behaviour on the site and knows neither product cost, margin nor returns, and it keeps raw data for a limited time. A warehouse brings data from ads, the store, the warehouse and company systems into one place and lets you calculate profit rather than sessions. They are two different layers: GA4 is a source, the warehouse is where the sources meet.
More: free GA4, GTM and ad pixel scanner
How much does a data warehouse implementation cost?
Implementation starts from €1,230 net as a one-off, and maintenance is a monthly subscription quoted with it. With more source systems, or data that first needs to be put in order, the figure grows and we give it after the audit. The source audit itself is a separate one-off order and ends with a data map and a cost estimate. On top of that comes the cloud cost, paid directly to the provider and dependent on usage.
How long does a data warehouse implementation take?
Six to ten weeks from the end of the audit to the first working report, and most of that time goes not on building but on agreeing definitions and cleaning up data on the source side. Expansion to further channels and views then goes in stages and does not block you from using what is already in place.
Does the data stay with us?
Yes, the data stays with you. The warehouse sits in your Google BigQuery account, under your agreement with Google, and JustIdea has access granted by you. When the work ends you keep the database, the models, the reports and the full history. That is an important difference from tools where the data sits with the vendor and disappears along with the subscription.
Which systems can the warehouse connect to?
On the marketing side Google Ads, Meta, Microsoft Ads, Allegro, GA4 and Search Console, and on the sales side Shopify, PrestaShop, WooCommerce, Shoper, IdoSell, Magento and BaseLinker, because these already have ready-made connections. We bring in the ERP and the warehouse management system through a data import that we agree with you. The difficulty is not the number of integrations but whether a given system exposes data at all, and whether an order can be identified in it unambiguously. That is what we check in the audit.
How is this different from a Looker Studio report?
Looker Studio is a visualisation layer: it shows whatever it is given. Connected straight to the ad platforms, it repeats their limitations, including the same sale being counted by several channels. The warehouse is the layer underneath, and it resolves such conflicts before a number reaches the chart. We usually build the reports themselves in Looker Studio, but on the warehouse, not on the platforms.