# How to Implement Cohort Analysis for Campaign Performance
In the fast-paced Australian digital landscape, many small business owners make the mistake of looking only at immediate conversions. Cohort analysis allows you to look deeper, grouping your customers by acquisition date or campaign to see how they behave over weeks or months. This is the key to understanding true Customer Lifetime Value (CLV) and whether your marketing spend is actually building a loyal base or just buying one-time clicks.
Why Cohort Analysis Matters
Standard reporting tells you how many people clicked an ad yesterday. Cohort analysis tells you how many of the people who clicked that ad in January are still buying from you in June. For Brisbane businesses running seasonal promotions or subscription models, this data is the difference between scaling a profitable campaign and wasting budget on low-quality leads.---
Prerequisites
Before you begin, ensure you have the following:- Google Analytics 4 (GA4) installed: This is the industry standard for Australian businesses.
- Historical Data: At least 30–60 days of data is required to see meaningful trends.
- Conversion Tracking: You must have events (like 'purchase' or 'generate_lead') set up and firing correctly.
- Clear Campaign Tagging: Ensure your links use UTM parameters (Source/Medium/Campaign) so GA4 can group users correctly.
---
Step 1: Access the Explorations Suite
Log into your Google Analytics 4 property. On the left-hand sidebar, click on the Explore icon (it looks like a graph with a magnifying glass). This is where the advanced analysis happens, beyond the standard surface-level reports. What you should see: A screen titled "Explorations" with a template gallery at the top.Step 2: Select the Cohort Exploration Template
In the template gallery, click on the box labelled Cohort exploration. This pre-configures the workspace with the dimensions and metrics needed for cohort analysis, saving you the hassle of building it from scratch. What you should see: A table with "Cohort" rows and "Week" columns, likely populated with some default data.Step 3: Define Your 'Cohort Inclusion' Criteria
In the Variables column (the far-left pane), look for Cohort Inclusion. This defines what makes a user part of a group.- For most Australian SMEs, you should set this to First touch (acquisition date).
- This means users are grouped by the date they first interacted with your website.
Step 4: Set Your 'Return Criteria'
Directly below inclusion, you’ll find Return Criteria. This is the action you want the user to take again later to be considered "retained."- If you run an e-commerce store, set this to purchase.
- If you are a service-based business (like a plumber or accountant), you might set this to session_start or a specific lead form submission.
Step 5: Configure the Granularity (Time Intervals)
In the Tab Settings column (the middle pane), find Granularity. You can choose between Daily, Weekly, or Monthly.- Pro Tip: For most small businesses, Weekly is the "sweet spot." Daily is often too noisy, and Monthly takes too long to show actionable trends.
Step 6: Break Down by Campaign
This is where the "Campaign Performance" part comes in. We don't just want to see all users; we want to see users from specific ads.- In the Variables column, click the + icon next to Dimensions.
- Search for "Session campaign" or "First user campaign."
- Click Import.
- Drag this new dimension into the Breakdown section in the Tab Settings pane.
Step 7: Choose Your Calculation Type
In the Tab Settings, look for Cohort Calculation. You have three choices:- Standard: Counts users who return in a specific period.
- Cumulative: Adds the metric (like revenue) over time.
Step 8: Adjust the Metric Type
By default, GA4 often shows "Active Users." To see campaign ROI, change the Value in Tab Settings to Purchase Revenue or Transactions. What you should see: The cells in your table will change from user counts to dollar amounts, showing how much revenue a specific cohort has generated over their lifetime.Step 9: Use Segments for Deeper Insight (Optional)
If you want to compare how Brisbane-based users perform against the rest of Australia, you can create a Segment. Click the + in the Segments box, select User Segment, and filter by "City" or "Region."Step 10: Analyse the Heatmap
Look at the colours in your table. Darker blue cells indicate higher retention or revenue.- Vertical Analysis: Are your newer cohorts (bottom rows) performing better than your older ones? This shows if your marketing is improving.
- Horizontal Analysis: How quickly does revenue drop off? If you lose 90% of users by Week 2, you may need a better email follow-up sequence.
---
Pro Tips for Australian Business Owners
- Account for Seasonality: Remember that Australian retail has unique peaks. A cohort acquired during the EOFY (End of Financial Year) sales will naturally behave differently than one acquired in quiet February. Don't judge your February campaigns by June standards.
- Exclude Internal Traffic: Ensure your own office IP address is excluded in GA4 settings, otherwise, your staff's frequent visits will skew your retention data to look unnaturally high.
- Combine with UTMs: Be extremely disciplined with your UTM tagging. If you don't name your campaigns consistently (e.g., using both "Summer_Sale" and "summer-sale"), GA4 will treat them as two different cohorts.
Common Mistakes to Avoid
- Too Small Sample Sizes: If a campaign only brought in 10 people, a cohort analysis won't be statistically significant. Aim for cohorts of at least 50-100 users for reliable insights.
- Ignoring the 'Lag': Revenue often takes time to settle. Don't panic if the most recent week in your cohort looks empty; some customers take a few days to make a decision.
- Mixing Acquisition Sources: Don't compare organic search cohorts directly against paid search cohorts without noting the cost difference. Paid cohorts should ideally have a higher "Cumulative Value" to justify the spend.
Troubleshooting
- "No data available": This usually means your date range is too short or your filters are too restrictive. Try expanding the date range in the top left of the Exploration suite.
- "The numbers don't match my standard reports": Exploration reports use a different processing engine than standard GA4 reports and may involve data sampling. They are meant for identifying trends, not for 100% accounting accuracy.
- "Campaign names are showing as (not set)": This happens when a user's session didn't have a clear source. Check your ad links to ensure UTM parameters are correctly appended.
---
Next Steps
Once you’ve mastered cohort analysis, you can start optimizing your budget by shifting funds away from campaigns with high initial conversions but low long-term retention.Need help setting up advanced tracking or interpreting your GA4 data? The team at Local Marketing Group can help you turn these insights into a high-growth strategy. Contact us today to book a strategy session.