GA4 & Nearshoring: AI Attribution in 2026

Nearshoring to Latin America is accelerating, but it’s creating a huge attribution headache. Your supply chains are becoming ‘agentic’, meaning AI is making decisions about logistics and inventory that directly affect the customer’s path to purchase. How do you track the ROI of your marketing spend when an algorithm, not a person, influences a key part of the journey? You need to get serious about how you track and measure every touchpoint, from the first ad click to the final delivery confirmation that was routed by an AI.

Setting Up Advanced AI Attribution in Google Analytics 4 for Nearshoring Supply Chains

You can’t prove your marketing works if you can’t track the results, and that’s the bottom line. With nearshoring, a customer’s journey is messy, it might start with a Google search in the US, involve checking stock at a distribution hub in Mexico, and end with support from a call center in Colombia. Standard attribution falls apart here. GA4 is really the only tool for this job because its event-based model is built to track these kinds of fragmented actions, not just simple pageviews. This guide shows you how to set up GA4 to actually measure the influence of your agentic supply chains, especially those connected to Latin American markets, so you can see what’s truly driving sales.

Step 1: Configuring Data-Driven Attribution (DDA) in GA4

Older analytics tools that just gave 100% of the credit to the last click were useless for modern marketing. A customer might see a display ad, get a remarketing email, and then click a branded search ad to buy, the last-click model pretends the first two interactions never happened. GA4’s Data-Driven Attribution (DDA) fixes this by using machine learning to give partial credit to every single touchpoint that contributed to the sale. This is essential for nearshoring, where a customer in Texas might do initial research from a global campaign but only convert after seeing your site promise fast shipping from a new warehouse in Monterrey.

1.1 Accessing Attribution Settings

1. Log into your GA4 property.
2. Go to Admin (the gear icon, bottom left).
3. Under the “Property” column, find and click on Attribution settings.

1.2 Selecting the Data-Driven Model

In the Attribution settings, you’ll see “Reporting attribution model” and “Conversion window.”

1. Change the “Reporting attribution model” dropdown to Data-driven. This is the most critical setting for getting attribution right because it stops using arbitrary rules (like first or last click) and instead uses your actual user data to figure out what’s working.
2. Adjust your “Conversion window.” For complex nearshoring, especially in B2B, the sales cycle can stretch for months. Extend the “Acquisition conversion window” to 90 days and the “Other event conversion window” to 30 days. This makes sure you don’t ignore the early-funnel ads that a B2B buyer clicked two months before finally making a purchase order.

Pro Tip: DDA isn’t a “set it and forget it” feature. You have to check your path reports. If you see a lot of conversions where the first touchpoint is falling just outside your 90-day window, you need to extend it further. Sticking with the default 30-day window will cause you to systematically undervalue your awareness campaigns in any business with a long sales cycle.

Expected Outcome: Your GA4 reports will finally start showing how different channels work together to create a conversion. Instead of seeing “Direct” get all the credit, you’ll see the fractional credit given to the paid social ad, the email newsletter, and the organic search click that led to that final visit. You’ll get a much better read on campaigns targeting Latin American customers, whose buying habits might not match what you see in the US or Europe.

Step 2: Implementing Server-Side Tagging via Google Tag Manager

Client-side tracking is dying. Between Safari’s ITP, Firefox’s ETP, and Chrome’s upcoming phase-out of third-party cookies, relying on the user’s browser to send data is a recipe for massive data loss. Server-side tagging (SST) via Google Tag Manager (GTM) fixes this. It takes the data from your website and sends it to GA4 from your own server, making the connection more durable, secure, and faster. If you’re running nearshoring operations, you absolutely need to maintain data integrity across borders, and that makes SST table stakes.

2.1 Setting Up a GTM Server Container

1. Inside GTM, create a new Container, but choose Server for the platform.
2. You need to provision a tagging server. Most businesses can just use the automatic provisioning option inside GTM, which connects to a Google Cloud Run instance. It’s much easier than setting it up manually.
3. After setup, you get a unique “Container Config URL.” You must create a subdomain on your own site (like gtm.yourdomain.com) and point it to this URL.

2.2 Configuring Your GA4 Client and Tag in the Server Container

1. In your GTM server container, navigate to Clients.
2. Click New and select GA4. This client is what “listens” for the data coming from your regular, client-side GTM container.
3. Now go to Tags.
4. Click New and pick Google Analytics: GA4.
5. Enter your GA4 “Measurement ID” (the G-XXXXXXXXXX one).
6. Set the trigger to fire on the events you need, usually starting with All Pages.

Editorial Aside: I see too many marketers put off server-side tagging because it seems complicated. It’s not optional anymore. The accuracy you get back, especially when trying to attribute conversions in a complex nearshoring setup, pays for the initial setup hassle ten times over. You get cleaner data today which makes your attribution models actually reflect reality.

Common Mistake: Forgetting to set up a custom subdomain for your tagging server. If you just use the default appspot.com URL that Google gives you, browsers will still see it as a third-party request and may block it, which defeats the whole purpose. Use a first-party subdomain, always.

Expected Outcome: Your GA4 data collection will become far more accurate because it’s less affected by ad blockers and browser privacy settings. That means your attribution models are working with a more complete picture of the customer journey, including all the subtle interactions influenced by your agentic systems in nearshoring locations.

Step 3: Using BigQuery Export for Granular Attribution Analysis

The GA4 interface is fine for high-level reports, but the real insights for something as complex as agentic supply chain attribution are buried in the raw data. GA4’s native export to Google BigQuery is where the magic happens. You get access to the unsampled, event-level data for every single user, which lets you build custom attribution models and run path analyses that are impossible in the standard UI.

3.1 Linking GA4 to BigQuery

1. In GA4, go to Admin.
2. In the “Property” column, find and click BigQuery Linking.
3. Click Link, choose your Google Cloud project, and pick a dataset location. Make sure you select the daily export option to keep your data fresh.

3.2 Querying Conversion Paths in BigQuery

Data will start showing up in BigQuery within 24 hours. Then you can start writing SQL to dig in. This is a basic query to pull the event history for all users who made a purchase in the last week:

SELECT user_pseudo_id, (SELECT value.string_value FROM UNNEST(event_params) WHERE key = 'page_location') AS page_location, event_name, event_timestamp
FROM `your-project-id.analytics_XXXXXX.events_*`
WHERE _TABLE_SUFFIX BETWEEN FORMAT_DATE('%Y%m%d', DATE_SUB(CURRENT_DATE(), INTERVAL 7 DAY)) AND FORMAT_DATE('%Y%m%d', CURRENT_DATE()) AND user_pseudo_id IN ( SELECT user_pseudo_id FROM `your-project-id.analytics_XXXXXX.events_*` WHERE _TABLE_SUFFIX BETWEEN FORMAT_DATE('%Y%m%d', DATE_SUB(CURRENT_DATE(), INTERVAL 7 DAY)) AND FORMAT_DATE('%Y%m%d', CURRENT_DATE()) AND event_name = 'purchase' )
ORDER BY user_pseudo_id, event_timestamp

From there, you can write more advanced SQL to do things the GA4 interface can’t, like building a custom time-decay model that gives more credit to touchpoints closer to the conversion, or isolating paths for users who interacted with specific content about your Latin American supply chain.

Pro Tip: Look for patterns linking marketing to logistics. Are people who visit the “local distribution” page converting at a higher rate? Did they see a specific ad campaign first? BigQuery lets you ask and answer these hyper-specific questions, giving you hard data to prove the impact of your nearshoring setup.

Expected Outcome: You’ll have the power to run your own custom attribution analyses. You can move beyond GA4’s built-in models and create one that perfectly reflects your business, identifying the exact supply chain touchpoints that are actually influencing customer decisions.

Step 4: Segmenting Audiences for Nearshoring Insights

To figure out if your nearshoring strategy is actually working, you have to isolate and compare different groups of users. GA4’s audience builder is perfect for this. You can create segments of users based on their location or their specific interactions with your site, letting you see how your nearshoring operations affect behavior.

4.1 Creating Geographic Segments

1. In GA4, go to Audiences in the left menu.
2. Click New audience and then Create a custom audience.
3. Set up a condition like this: User segment > Geography > Country/Region, and then select the countries you’re focused on, like “Mexico” or “Colombia.”
4. Give it a clear name, like “Users in Mexico,” and save.

4.2 Creating Behavioral Segments for Agentic Influence

If you have specific digital touchpoints related to your supply chain (like an AI chatbot for shipping questions or a page detailing regional stock), you can segment users who interact with them.

1. Follow the same steps as above to create a new custom audience.
2. This time, set the condition to be an Event. Choose the specific event name, like chatbot_interaction or view_local_fulfillment_page.
3. You can get even more specific by combining this with a geographic condition to create an audience like “Latin American users who interacted with AI chatbot.”

Common Mistake: Making your segments too generic. A segment for “South America” is way too broad to be useful. You need to get granular, segment by specific countries (or even cities, if your data is good enough) to get actionable insights about how your nearshoring efforts are performing in different markets.

Expected Outcome: You’ll be able to pull up attribution and conversion path reports for just the users affected by your nearshoring strategy. This lets you make specific marketing adjustments, like increasing ad spend in a region where local fulfillment is driving high conversion rates, and get a much clearer picture of the ROI from those operations.

Step 5: Auditing Event Parameters for Complete Data Collection

Your attribution model is only as good as the data you feed it. For something as complex as an agentic supply chain, the richness and consistency of your event data are everything. If you aren’t tracking a parameter like `supply_chain_region` every time a user checks inventory, then your DDA model is flying blind and can’t assign credit properly. This is why you have to audit your event setup constantly.

5.1 Reviewing Custom Definitions

1. In GA4, head to Admin.
2. Under the “Property” column, click Custom definitions.
3. Look through your list of custom dimensions and metrics. Did you remember to register `supply_chain_region` as a custom dimension? If you don’t register parameters here, you can’t use them in your reports.

5.2 Debugging with DebugView

GA4’s DebugView (in the Admin panel) is your best friend for checking your tracking in real time. It’s an essential tool for validating your setup.

1. Go to your website and perform an action that should trigger one of your nearshoring-related events (e.g., interact with the chatbot about shipping from a Latin American warehouse).
2. Watch the event stream in DebugView. Click on the event and check its parameters. Is the supply_chain_region parameter there? Is the value correct? If not, you’ve found a tracking bug that needs to be fixed.

Pro Tip: This requires talking to your developers. Your agentic systems (like an AI logistics platform) must be configured to pass data back to your analytics. If an AI makes a choice that affects a customer, that decision needs to be fired as a GA4 event with all the relevant context. Marketing can’t do this alone.

Expected Outcome: You’ll have confidence that your GA4 data is clean and complete. You’ll know that all the critical information needed to attribute value across your entire nearshoring supply chain is actually being collected, which allows the DDA model to do its job and give you insights you can trust.

Getting attribution right for nearshoring and agentic supply chains is a continuous job, not a one-off project. But by properly setting up GA4’s Data-Driven Attribution, moving to server-side tagging, digging into the raw data with BigQuery, building smart audience segments, and auditing your event tracking, you can get the insights you need. This is how you prove the value of your marketing spend and truly understand the impact of your nearshoring investments. It’s also how CMOs can start to build trust in AI shopping and other automated parts of the customer experience.