What is the Dotted Line in Google Analytics?

Seeing a dotted line suddenly appear on a Google Analytics chart can be confusing. Your immediate thought might be that the data is incomplete, the tracking is broken, or that you're looking at a forecast of future traffic. The good news is that it’s none of those things. This article explains what that dotted line actually means, why Google uses it, and what you can do about it.

What the Dotted Line in Google Analytics Really Means

In short, a dotted line on a Google Analytics chart signals that the report you're viewing is based on sampled data. This happens in both Universal Analytics (the older version) and Google Analytics 4.

Instead of analyzing every single event or session for your report — which could be millions or even billions of data points — Google Analytics looks at a smaller, random subset of that data. It then extrapolates from that subset to estimate what the totals would be. It's a bit like trying to figure out what flavor a giant pot of soup is. You don't need to drink the entire pot, a single spoonful gives you a very reliable idea of what the whole thing tastes like.

Google does this for one primary reason: speed. Processing the entirety of your website's raw data for a complex, customized report takes a lot of computing power and time. By using a sample, GA can deliver your report in seconds instead of minutes. The dotted line is simply a visual cue letting you know that the numbers you're seeing are a high-quality estimate, not a precise count of every single action.

Why Does Google Analytics Sample Your Data Anyway?

As mentioned, it's all about providing a speedy user experience. But what triggers this sampling? It's not totally random. Sampling typically kicks in when your data requests hit certain thresholds. While Google doesn't share the exact formulas, here's the general idea.

It's Usually About Complexity

Standard, pre-built reports in GA4 (like the Traffic acquisition or Pages and screens reports) almost never use sampled data. They pull from pre-aggregated data tables that are designed for quick and accurate reporting.

Sampling usually appears when you start creating more complex queries in the Explore section of GA4. These are called "ad-hoc" queries because they are generated on the fly. You're more likely to see a dotted line if your exploration includes:

A very long date range: Analyzing 14 months of data is much more intensive than analyzing 7 days.
Multiple dimensions: Asking for traffic by Source, Medium, Campaign, and Country is more complex than just asking for traffic by Source.
Complex segments or filters: Drilling down to a very specific user group (e.g., mobile users from California who visited a specific landing page and also viewed your pricing page) requires a much more detailed query.
Huge amounts of data: For free GA4 properties, queries that involve more than 10 million events are subject to sampling. For paying Google Analytics 360 users, that threshold is much higher (1 billion events).

In essence, the more detailed and specific your question, the more work Google's servers have to do, making it more likely that it will switch to a sample to give you a quick answer.

How to Check for and Interpret Sampled Data in GA4

In older versions of Google Analytics, there was often a clear text-based warning at the top of the report. In GA4, it's a bit more subtle.

At the top of your report (usually near the title), you'll see a small shield icon.

A green shield with a checkmark means the report is based on 100% of your data. No sampling has occurred.
A yellow shield with an exclamation point means the report is based on a sample.

If you hover over the yellow shield, a box will appear that tells you what percentage of the available data was used to create the report. For example, it might say, "This exploration is based on 52.4% of available data."

So, Is a 52% Sample Good or Bad?

This is where context matters. If your report is based on millions of total events, a 52% sample is still drawing from an enormous pool of data and is likely to be very accurate for directional insights. However, if your data set is much smaller, a 52% sample of a small number could lead to a higher margin of error. The higher the percentage, the more closely the sampled data will reflect reality.

Should You Actually Be Worried About Dotted Lines?

The answer is a very common one in data analysis: it depends.

When Sampling is Probably Fine

For many day-to-day tasks, sampled data is perfectly acceptable and provides reliable insights. You probably don't need to worry if you are:

Looking for broad trends: Are our US sales generally going up or down this quarter? Did the last blog post cause a noticeable spike in traffic? Sampled data is great for spotting these directional patterns.
Comparing performance at a high level: Which marketing channel is bringing in more users overall — Organic Search or Paid Social? A sample is more than enough to give you a clear winner.
Working with high traffic sites: If you get millions of hits a month, even a 10% sample represents hundreds of thousands of sessions, which is usually a statistically significant data set.

When You Need Precision: Times to Be Cautious

In some situations, estimations are not good enough. You should be wary of sampled data and take steps to avoid it when you are:

Performing financial reporting: When reporting on e-commerce revenue or transaction data, you need exact numbers. Estimations could cause you to misreport performance to stakeholders.
Analyzing A/B test results: The winner of an A/B test can often be decided by a very slim margin. Sampled data could inaccurately suggest a winner or loser on a conversion test that is actually too close to call.
Making critical budget decisions: If you are planning to cut a specific campaign or ad group because it appears to have a poor ROI, you want that decision to be based on 100% of the data, not an estimate.
Working with low-conversion events: If you're analyzing an event that happens very rarely (like a "Request Enterprise Demo" form fill), sampling might miss these events entirely or misrepresent their frequency, leading to bad conclusions.

4 Ways to Deal With (or Avoid) Data Sampling in GA4

If you've decided that a sampled report isn't precise enough for your task, you have a few options to get a full, unsampled version.

1. Simplify Your Report (The Quickest Fix)

The easiest method is to reduce the complexity of your query. Try one of these adjustments:

Shorten the date range: Instead of looking at the last 90 days, try looking at the last 30 days. You can often run three separate 30-day unsampled reports to get the same insight as one 90-day sampled report.
Remove dimensions or segments: Do you really need to break down your landing page traffic by device, country, and acquisition source all at once? Removing one of those dimensions might be enough to get you back to an unsampled report.

2. Use the Data Quality Icon Request

If you’re still getting used to GA4, sometimes you’ll see an option to get a more detailed report if requested. Next to the sampling icon, you can click to get a higher-precision report which can process in just a few minutes, depending on the complexity of your query. Just an extra helpful tip!

3. Switch Your Reporting Identity

This is a more technical adjustment. In your GA4 property settings, you can choose how Google identifies users.

To access it, go to Admin > Data display > Reporting Identity. You’ll see a few options. The "Blended" identity combines several methods and may also include modeled data to fill in gaps. This complexity can sometimes make sampling more likely. Changing your setting to "Observed" relies only on collected data (like cookies and User-ID) and is less likely to trigger sampling. Be aware that this may change your reported user counts, as it won't include estimates for users who can't be observed directly.

4. Export to BigQuery (The Power User Move)

For complete control and guaranteed unsampled data, the best solution is to use the native GA4 integration with BigQuery. BigQuery is Google's data warehouse, and you can set up GA4 to send a copy of all your raw, un-sampled event data there every day.

This provides you with the ground truth for all A/B tests, financial reports, and granular analyses. The catch? You'll need to know some SQL (the language for querying databases) or use a tool that connects to BigQuery to analyze and visualize the data. This option offers the highest level of precision but also has the steepest learning curve.

Final Thoughts

That dotted line in Google Analytics isn't a problem to be feared, but rather a signal to be understood. It means Google is giving you a fast, reliable estimate by looking at a portion of your data. For spotting general trends, this is often all you need. But when precision matters — especially with financial data or A/B tests — it’s important to use strategies like simplifying your report or exporting your raw data to get the full picture.

Knowing that many teams don't have the time to wrestle with these workarounds is why we built our platform. We designed Graphed to remove common data headaches, including data sampling. You can connect your Google Analytics account in just a few clicks, and a copy of your full, raw data is piped into an accessible data warehouse for your use. From there, you can ask for the unsampled insights you need in simple, plain English and we build live, accurate dashboards that are always based on 100% of your data. Give Graphed a try to get clear, unsampled marketing reports done in seconds, not hours.