A/B Test Mail Merge Subject Lines (Google Sheets)

Learn how to A/B test mail merge subject lines from Google Sheets in Gmail — set up variants, split your list, measure opens, and roll out the winner to lift open rates.

How to A/B Test Mail Merge Subject Lines from Google Sheets (Gmail)

Your subject line decides whether a personalized email gets opened at all. If you run mail merge from Google Sheets, you can A/B test subject lines directly from your spreadsheet: send two or more variants to a slice of your list, measure which gets more opens, then roll the winner out to everyone else. This guide shows the workflow with Mail Merge for Gmail and Google Sheets.

Test subject lines and lift your open rate — start free →

Gmail and Google Sheets are trademarks of Google LLC, used here descriptively; DocGPT.ai Mail Merge is independent and not affiliated with or endorsed by Google. Details are general and current as of June 2026.

Why subject-line testing moves open rates

Two emails with identical bodies can get very different open rates depending only on the subject line. Testing replaces guesswork with evidence: instead of betting your whole list on one line, you let a small sample tell you which phrasing earns more opens, then send the winner to the rest. Over a few campaigns, small lifts in open rate compound into meaningfully more replies and conversions.

Setting up A/B variants in a Google Sheet column

Keep your variants in the spreadsheet so everything lives in one place:

Add a column such as Subject variant to your recipient Sheet.
Write your two (or more) subject lines you want to test — for example, Variant A: “A quick idea for {{Company}}” and Variant B: “{{First name}}, worth 2 minutes?”.
Optionally add a Group column (A or B) so you can see which recipient received which line.
Keep the email body identical across variants so the subject line is the only thing you are testing.

Splitting your list into test groups from the Sheet

Take a representative sample of your list for the test — a slice large enough to show a difference but small enough to leave most recipients for the winning line. Assign roughly half of the sample to Variant A and half to Variant B (you can alternate rows, or use a simple formula to label groups). Keep the groups comparable so the only meaningful difference between them is the subject line.

Running the mail merge with variant subjects

Send Variant A to its group and Variant B to its group, with the same personalized body merged from your Sheet. A worked example: from a 1,000-recipient list, test on 200 — 100 get Variant A, 100 get Variant B. You hold the remaining 800 until you know which subject won. Send both test groups at a similar time of day so timing does not skew the result.

Measuring opens per variant with tracking

Turn on open tracking so each variant records how many recipients opened. Compare the open rate (opens divided by delivered) for Variant A versus Variant B, not just raw counts, since group sizes may differ slightly. Give the test a little time to mature — opens trickle in over hours — before you read the result.

Picking the winner & rolling it out to the rest of the list

Once one variant clearly leads on open rate, use that subject line for the remaining recipients — in the example, the held-back 800. Selecting the winner is a manual decision you make from the tracked results; record which line won so you can reuse the insight in future campaigns. If the two variants are close, treat the result as inconclusive and test a bolder difference next time.

Tips: sample size, one variable at a time, send-time

Use a large enough sample. Tiny tests produce noisy results; a few opens’ difference on 20 recipients means little.
Change one thing at a time. Test the subject line alone — keep the body, sender, and send-time consistent.
Control send-time. Send both variants in the same window so time of day does not contaminate the comparison.
Re-test periodically. Audiences and inboxes change, so yesterday’s winner is not guaranteed forever.

Compliance & deliverability cautions

Open-rate testing must still respect the rules. Send only to permissioned, relevant contacts; keep both subject lines honest — no clickbait or misleading “Re:” tricks that erode trust and trip spam filters. Include a clear sender identity and an opt-out path where appropriate, respect Gmail and Google Workspace daily sending limits, and avoid spammy phrasing. Good deliverability is what makes open-rate testing meaningful in the first place.

Run your first subject-line test from Google Sheets — start free →

A/B testing subject lines FAQ

How do I A/B test subject lines in a Gmail mail merge?

Put your subject-line variants in a Google Sheet column, send each variant to a comparable slice of your list with open tracking on, compare the open rates, then send the winning subject line to the remaining recipients.

How big should my test sample be?

Large enough that a difference in opens is meaningful — a small handful of recipients produces noisy, unreliable results. A common approach is to test on a portion of the list and reserve the majority for the winning variant.

How do I measure which subject line won?

Use open tracking and compare the open rate (opens divided by delivered) for each variant rather than raw counts, since group sizes can differ. Give opens time to accumulate before deciding.

Should I test more than the subject line at once?

No. Change one variable at a time. If you also change the body, sender, or send-time, you will not know which change drove the difference in opens.

How do I keep A/B testing compliant?

Send only to permissioned contacts, keep both subject lines honest (no misleading “Re:” or clickbait), include a clear sender identity and opt-out where appropriate, respect Gmail/Workspace sending limits, and avoid spammy phrasing.

Start testing from your Sheet

You do not need a separate email platform to improve open rates. Keep your variants in Google Sheets, test on a sample, and let Mail Merge for Gmail and Sheets send the winner to the rest of your list.

Try Mail Merge for Gmail and Sheets →