New customer versus returning customer conversion rates at checkout

New customer versus returning customer conversion rates at checkout is a metric that separates stores that simply attract traffic from those that actually monetize it.

Why the split matters for revenue optimisation

When you look at the checkout funnel, you are really looking at two parallel streams. A first‑time visitor arrives with no purchase history, no saved payment details and often a higher perceived risk. A returning shopper, on the other hand, has already demonstrated trust, may have items saved in a wishlist, and usually expects a faster, frictionless experience. Studies from the Baymard Institute show that the average cart abandonment rate for new visitors is around 69 %, while returning visitors abandon at roughly 56 %.

The gap translates directly into lost revenue. If you process 10 000 sessions a month, with 70 % new and 30 % returning, and the average order value (AOV) is R2 500, a 13 % difference in conversion (30 % vs 43 %) means you are leaving roughly R1 800 000 on the table each month. Understanding the distinct behaviours of each group lets you apply targeted fixes instead of generic tweaks that only move the needle a little.

The data‑driven method to isolate the two rates

Below is a step‑by‑step framework that can be implemented in any analytics platform (Google Analytics, Adobe Analytics, Matomo, etc.). The method is built around three pillars: segmentation, measurement, and hypothesis testing.

1. Create the segmentation

• In your analytics tool, set a custom dimension called Customer Type with values “New” and “Returning”. Most platforms already have a built‑in flag for first‑time users; map that to “New”.
• Apply the dimension to the checkout success event (e.g., purchase or order_complete).
• Validate the segment by pulling a simple report: Sessions → Customer Type → Conversion Rate.

2. Measure the baseline

Run the report for the last 30 days. Record the following numbers for each segment:

• Sessions
• Checkouts started
• Checkouts completed (conversion)
• Average order value

For example, a typical boutique fashion store might see:

This baseline gives you a clear picture of the current split and the revenue impact of each group.

3. Form hypotheses based on friction points

Look at the checkout funnel for each segment separately. Common friction points include:

• Mandatory account creation for new users.
• Missing guest checkout option.
• Long address entry forms that returning users have saved.
• Insufficient trust signals (security badges, clear return policy) for first‑time buyers.

Write each hypothesis in a testable format, e.g., “If we enable guest checkout for new users, their conversion will increase by at least 5 %.”

A hypothetical worked example: turning a 13% gap into a 5% lift

Consider a South African home-goods retailer with a 13% gap between new and returning conversion rates.

Step 1 – Identify the biggest drop‑off

The funnel analysis showed that 45 % of new users abandoned on the “Create password” screen, while returning users never saw that screen. The checkout completion rate for new users after entering shipping details was only 28 % versus 40 % for returning users.

Step 2 – Prioritise the fix

Based on impact and effort, the team chose to add a “Continue as guest” button that bypasses password creation. The change required a small UI tweak and an update to the order‑processing API to accept orders without a linked user account.

Step 3 – Run the A/B test

• Variant A (control): existing flow, mandatory account creation.
• Variant B (test): guest checkout option visible on the first screen.
• Sample size: 5 000 new sessions per variant (calculated using a 95 % confidence level and 1 % margin of error).

Step 4 – Analyse results

After 14 days, the test produced the following numbers:

The lift of 22 % on new‑customer conversion raised the overall checkout conversion from 30 % to 33 % (a 10 % relative increase). With the same traffic volume, the retailer added roughly R1 200 000 in monthly revenue (30 % increase on new‑customer sales).

Step 5 – Roll out and monitor

After confirming statistical significance, the guest checkout was deployed site‑wide. Ongoing monitoring showed the new‑customer conversion stabilising at 34 % while returning‑customer conversion remained at 43 %.

How to apply the method to your own store

Follow these concrete actions to start improving New customer versus returning customer conversion rates at checkout in your own environment.

Audit your current checkout flow

1. Map every step a user takes from cart to order confirmation.
2. Tag each step with an event that records the Customer Type dimension.
3. Export the data for the last 60 days into a spreadsheet.

Identify the top three friction points for new customers

Use a simple drop‑off calculation: (sessions that reach step X - sessions that reach step X+1) ÷ sessions that reach step X. Focus on the steps where the new‑customer drop‑off exceeds the returning‑customer drop‑off by more than 10 percentage points.

Implement quick wins

• Guest checkout: If you do not already have it, add a one‑click guest option.
• Auto‑fill address: Use a third‑party address lookup service (e.g., Google Places API) to reduce typing.
• Trust badges: Place SSL and payment‑provider logos above the “Place order” button for first‑time visitors.
• Clear return policy: Show a concise return statement on the payment page.

Run systematic A/B tests

For each hypothesis, set up a test with a minimum sample size of 4 000 new sessions per variant. Use a reliable testing platform (Google Optimize, VWO, Optimizely) and run the test for at least one full business cycle to capture any weekday/weekend variation.

Iterate based on data

When a test reaches statistical significance, roll out the winning variant. Then revisit the funnel and repeat the process. Over time you will shrink the gap between new and returning conversion rates, often to under 5 percentage points.

Common mistakes and how to avoid them

Even experienced operators can fall into traps that dilute the impact of optimisation work.

1. Mixing metrics

Do not compare “add‑to‑cart” rates with checkout conversion. The focus must stay on the final purchase event for each customer type.

2. Ignoring sample‑size requirements

A test with only 500 new sessions will produce noisy results. Use a calculator such as the one on Moz to determine the proper size.

3. Changing multiple elements at once

When you test a new payment gateway and a redesigned form in the same experiment, you cannot know which change drove the lift. Keep each test isolated.

4. Forgetting to update the segmentation after a change

If you introduce a “saved cards” feature, you must adjust the “Returning” definition to include users who have a saved card even if they have not purchased before. Otherwise the data will be skewed.

5. Assuming returning customers need no optimisation

Returning shoppers still abandon at a significant rate. After you have lifted the new‑customer side, repeat the same funnel analysis for returning users and look for opportunities such as loyalty‑program reminders or one‑click re‑order.

Quick checklist for the next 30 days

• Set up the Customer Type dimension and verify data collection.
• Run the baseline report for the past 30 days and note the conversion gap.
• Identify the top three new‑customer drop‑off steps with a >10 % differential.
• Implement at least one quick win (guest checkout, trust badge, address auto‑fill).
• Design and launch an A/B test for the biggest friction point, targeting 4 000 new sessions per variant.
• Analyse results, roll out the winner, and document the revenue impact.
• Schedule a second round of tests focused on returning‑customer friction points.
• Use the free store scanner to spot hidden leaks: run a free scan of your store.