Graphed LogoGraphed
Graphed LogoGraphed
How It WorksAgentsIntegrationsPricing

How to Change Data Type in Looker Studio

Cody Schneider••9 min read

Getting your numbers to add up correctly in a Looker Studio report is often as simple as making sure Looker knows it’s looking at numbers in the first place. When a date is treated like text or revenue is mistaken for a simple string of characters, your entire analysis can fall apart. This guide will walk you through exactly how to change data types in Looker Studio, covering both permanent fixes at the data source level and quick changes for individual charts.

GraphedGraphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Watch Graphed demo video

What Are Data Types and Why Do They Matter?

In Looker Studio (and in data analysis in general), a "data type" is just a label that tells the software how to interpret a column of data. Think of it like a label on a container. If you have a box labeled “Toys,” you know you can play with what’s inside. If it’s labeled “Food,” you know you can eat it. You wouldn’t try to eat a toy, and you wouldn't try to add up a column of text.

Data types are what let Looker Studio know it can perform a SUM or AVERAGE on a Revenue column, correctly plot a time-series chart with a Date column, or display locations on a map using a City column. If the data type is wrong, your charts won't work, your calculations will fail, and you’ll get confusing error messages. Correctly setting data types is the foundation of any accurate and useful dashboard.

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Common Looker Studio Data Types

Looker Studio has a variety of data types, but you'll run into these most often:

  • Text (ABC): This is for strings of characters, like names, descriptions, categories, or IDs. You can count or group by text fields, but you can’t perform mathematical calculations on them. (e.g., 'Campaign Name', 'Product SKU', 'Country Name').
  • Number (123): This covers all numerical data. It can be a simple number, a currency, or a percentage. This is the type required for any metric you want to aggregate (sum, average, min, max, etc.). (e.g., 'Sessions', 'Cost', 'Conversion Rate').
  • Date & Time (Calendar Icon): This type is for dates, timestamps, or both. Setting a field to a Date type is essential for creating time-series charts, using date range controls, and calculating periods like 'year over year'. (e.g., '2024-10-26', 'Oct 26, 2024, 10:30 AM').
  • Boolean (T/F): This represents true or false values. It’s useful for filtering or creating flags in your data, like "Is a Subscriber?" (True/False).
  • Geo (Globe Icon): This type is specifically for geographical data like 'Continent', 'Country', 'City', or 'Latitude, Longitude'. Fields with a Geo type are the only ones you can use to build map visualizations.
  • URL (Link Icon): This type is for website links. When used in a table, a URL field can automatically create clickable links, which is great for reports that reference specific pages or assets.

Most of the time, Looker Studio makes a decent guess when you connect a data source like Google Analytics or Google Sheets. But it frequently gets things wrong, especially with formats it doesn't recognize.

When Looker Studio Guesses Wrong (And How to Spot It)

Incorrect data types are behind many common Looker Studio frustrations. Here are a few telltale signs that you need to make a correction:

  • You drag Revenue into a scorecard, but it won't let you sum it, and the icon next to it is "ABC" instead of "123". This often happens if your data includes currency symbols (e.g., "$1,234.56"), as Looker may interpret the "$" as a character and classify the whole field as Text.
  • Your date field (Date) doesn't work in a time-series chart or with a date range filter. This happens when your dates are in an ambiguous format (like 01-02-2024) and Looker reads them as text strings.
  • You try to display your sales by state on a map, but nothing shows up. Your 'State' column might be classified as standard Text instead of Geo > Region.
  • The Order ID metric in your table is showing a massive, meaningless number like "2,456,189,451". This is because Looker classified it as a Number and is trying to SUM all the order IDs together, when really it should be treated as a Text dimension.

Fortunately, fixing these issues is straightforward. There are two primary places you can change a data type: at the data source level or at the chart level.

Method 1: Change Data Types at the Data Source Level (The Best Practice)

Changing a data type at the data source level is the most robust and recommended method. When you make a change here, it applies to every chart and every report page that uses that data source. It fixes the problem at its root, ensuring consistency everywhere. Do this once, and you won't have to fix it again.

Here’s how to do it step-by-step:

  1. Open Your Data Source Editor: In your report, navigate to the menu at the top and click on Resource > Manage added data sources.
  2. Select and Edit: A list of all data sources connected to your report will appear. Find the one you want to modify and click the EDIT button to its right.
  3. Locate the Field: You will now see a list of every field (dimensions and metrics) from your data source. You can use the search bar at the top to find the exact field you need to change. Each field has its name, its data type (with an icon), and its default aggregation.
  4. Change the Data Type: Find the field you want to modify. Its current data type is a clickable dropdown. For example, if a Cost field is incorrectly set to Text, you'll see "Text" and an "ABC" icon. Click on it.
  5. Select the New Type from the Dropdown: A menu will appear with all available data types. For our Cost example, you’d navigate to Number > Currency > and then select your currency (e.g., USD - US Dollar). For a date recognized as Text, you’d click on it and select Date & Time > Date.
  6. Save Your Changes: Once you've made your selections, click the blue DONE button in the top-right corner of the data source editor screen. Looker Studio will take a moment to refresh and apply the changes across your entire report.

That's it! Your field will now be correctly interpreted everywhere.

GraphedGraphed

Build AI Agents for Marketing

Build virtual employees that run your go to market. Connect your data sources, deploy autonomous agents, and grow your company.

Watch Graphed demo video

Method 2: Change Data Types at the Chart Level (The Quick Fix)

Sometimes you only need a field to behave differently for a single, specific visualization. For example, your Year field might be stored as a Number (e.g., 2023, 2024), which is great for tables. But for one particular chart, you might want it to be treated as a text dimension to prevent it from being aggregated. In this case, altering the data type at the chart level is the perfect quick fix.

This method only affects the chart you're editing and won't change the underlying data source settings.

Here’s how it works:

  1. Select the Chart: Click on the specific table, scorecard, or graph you want to modify within your report.
  2. Open the Setup Panel: The chart's properties panel will appear on the right side of the screen. Make sure you are in the SETUP tab.
  3. Find the Field: In the SETUP tab, you'll see the dimensions and metrics currently being used in your chart. Locate the field you want to change.
  4. Click the Field's Icon: To the left of each field's name, you'll see an icon representing its current data type (e.g., a green 'ABC' chip for Text, a blue '123' chip for Number, or a small calendar). Click on this icon.
  5. Choose the New Type: A small pop-up menu will appear. Hover over the Type option and then select the new data type you want to apply for this specific chart.

The chart will instantly update to reflect the change. Remember, any other charts using that same field from the same data source will remain unaffected.

Troubleshooting with Formulas: The CAST() and PARSE_DATE() Functions

Sometimes, simply selecting a new data type isn't enough, especially if the underlying data is 'messy'. Looker Studio might throw an error because it can’t convert a text string like "$1,500" into a number directly. In these situations, you need to create a calculated field using formulas to clean the data first.

Fixing Numbers with Text: The CAST() and REPLACE() Combo

Let's say your Ad Spend data from a Google Sheet looks like this: "$100", "$250", "$50". Looker sees the "$" and calls it Text. To fix this, you combine two functions:

  • First, use REPLACE() to remove the dollar sign. REPLACE(Ad Spend, "$", "")
  • Then, wrap that in a CAST() function to officially convert the result into a number.

The final formula in your calculated field would be:

CAST(REPLACE(Ad Spend, "$", "") AS NUMBER)

Free PDF · the crash course

AI Agents for Marketing Crash Course

Learn how to deploy AI marketing agents across your go-to-market — the best tools, prompts, and workflows to turn your data into autonomous execution without writing code.

Fixing Unrecognized Dates: The PARSE_DATE() Function

This is one of the most useful functions in Looker Studio. It tells the software exactly how to read a date that's stored as text or a number. For example, if your date is formatted as a single number string like 20241026 (YYYYMMDD), Looker won’t know what to do with it.

You can create a new calculated field with the following formula to convert it:

PARSE_DATE("%Y%m%d", YourDateField)

The first part ("%Y%m%d") is the 'format code' that tells Looker to expect a 4-digit year (%Y), followed by a 2-digit month (%m), followed by a 2-digit day (%d). The second part (YourDateField) is the name of the field you're fixing.

Final Thoughts

Setting the correct data types is one of the most fundamental skills for building reliable and functional dashboards in Looker Studio. By making sure your numbers, dates, and text are properly identified - either at the data source level for a permanent fix or on a specific chart for a quick adjustment - you eliminate countless errors and unlock the tool's full analytical power.

While mastering details like data types gets you much closer to effective reporting, you're still left with the manual work of dragging, dropping, and configuring every chart. At Graphed, we've removed that friction by connecting directly to your data sources and letting you use plain English to build your reports. Just ask, "Create a dashboard showing my Shopify sales vs. Facebook Ad spend for this quarter," and we build it for you instantly. Instead of spending hours managing fields and chart settings, you can get answers and insights in seconds. Find out more at Graphed.

Related Articles