SEO Title Testing Log Template

Title Testing Log (fill-in)

When to use: One row per title change on one URL, with query mix, displayed title, before/after numbers, owner and verdict.

SEO title testing log

One row per title change on one URL. Fill the row on the day you ship the change, then come back when the result window closes.

Per-row detail (fill for every change)

• Query mix before: top 5 queries and their share of the page's impressions: [query - impressions]
• Branded vs non-branded: [impressions from each, before and after]
• Filters used: [page URL, country, device, search type Web: text-based]
• Recrawl date: [date Google first showed the new title, or date of URL Inspection recrawl]
• Title Google displayed: [your title / rewritten to [text]] for [top query]
• Before: clicks [n], impressions [n], CTR [n], average position [n]
• After: clicks [n], impressions [n], CTR [n], average position [n]
• Other changes in the window: [content edits, internal links, Google updates, season, campaigns]
• Verdict: [keep / revert / extend window / inconclusive], because [reason]

Next action

• Test owner: [name] - action: [ship the new title and fill the row] by [date]
• Analyst: [name] - action: pull the before and after numbers from Search Console on [window end date + 3 days]
• Page owner: [name] - action: approve the verdict and [keep, revert or roll out] by [date]

Completion check

• Only the title changed on this URL during the result window, or every other change is logged.
• Before and after windows have the same length and start on the same weekday.
• The days between the change and Google showing the new title are left out.
• You checked which title Google actually displayed for the top queries.
• The query mix is compared, not only the page totals.
• The verdict names a reason and a next step, even when it is inconclusive.

Candidate and Hypothesis Picker

When to use: Choose URLs worth testing and write a hypothesis tied to the queries the page already gets.

Pick the URL and write the hypothesis

A title test is worth logging when the page already gets impressions, but searchers skip it. Start from data, not from a title you dislike.

Good candidates

• Steady impressions over the last 3 months, so there is something to compare.
• CTR lower than other pages of yours at a similar average position.
• Stable ranking for the main queries. Check the current positions in the Rank Tracker before you touch the title, and re-run it during the window.
• The top queries use words the current title does not.
• No other change planned for the page in the next 6 weeks.

Hypothesis builder

Because the top queries are [queries] and the current title says [old angle], changing it to [new angle] should raise CTR for [query group] without losing [query group to protect].

Weak vs better hypotheses

Preview how each version is cut off on desktop and mobile in the SERP Simulator, then score the wording in the Headline Analyzer.

Testing many similar pages at once with a control group? Use the SEO Split Testing Template instead. This log is for one URL at a time.

Comparison Window Planner

When to use: Set fair before and after windows in the Search Console Performance report, with a seasonality check.

Set the comparison windows in Search Console

A single-URL title test compares a period before the change with a period after it. The windows decide whether the read is fair.

Window rules

1. Use the same length for both windows, such as 28 days each.
2. Start both on the same weekday. Search Console dates are in Pacific Time.
3. Skip the gap between shipping the title and Google showing it. Google says reprocessing can take a few days to a few weeks.
4. Leave out the newest days. Search Console marks recent data as preliminary, and it can still change.
5. Avoid windows that span a holiday, sale or seasonal peak on one side only.

Steps in the Performance report

1. Open Performance, Search results, and set Search type to Web: text-based.
2. Add a Page filter for the exact URL. Use a custom regex with ^ and $ if other URLs contain the same text.
3. Open the date filter, choose Compare, and set the before and after windows.
4. Read the Queries tab. The Difference column shows the change per query.
5. Repeat with a Device filter if mobile and desktop titles may be cut off differently.

Seasonality check

• Compare the same after window with the same dates last year: [impressions then / now].
• Check a similar page you did not change over the same windows: [control URL and its CTR change].
• If the control moved the same way, the title is probably not the cause.

Search Console only allows one comparison at a time. Log the numbers in the sheet before you change the filter.

Want a second opinion on a test plan? Book a free consultation with Anatolii Ulitovskyi.

Query Mix and Metric Caveats Check

When to use: Read impressions, CTR and position the way Google counts them, then apply clear verdict rules.

Read the metrics the way Google counts them

A good title can look bad in a careless read. These checks come from how Search Console defines its metrics.

Metric caveats

• Impressions: counted when a link to your site appears in the results the user saw. More impressions can mean new, weaker queries.
• CTR: clicks divided by impressions. New low-intent queries can pull CTR down even if the title works better.
• Average position: the topmost position of your link, recorded only when there is an impression. A position change moves CTR on its own.
• Page vs property: the chart counts by property, page rows count by page. Use page-level numbers for this log.
• Anonymized queries: some queries are hidden for privacy. Query rows will not add up to the page total.
• Canonical URL: data is assigned to the canonical Google picks. Log that URL.

Query mix check

Verdict rules

• Keep: CTR rose for the main query group at a similar position, and clicks did not fall.
• Revert: CTR fell for the main group at a similar position, or the page lost queries it used to win.
• Extend: too few impressions in the window to say. Add another full window.
• Inconclusive: position, season or query mix moved too much, or Google did not show the new title.

The branded queries filter in Search Console helps split brand searches from the rest, where it is available for your property.

Title Rewrite Check

When to use: Confirm Google displayed your new title for the top queries before you judge the result.

Did Google show your new title?

Google builds title links automatically. It often uses your title element, but it can also use the main heading, og:title, anchor text or other text on the page.

Check before you read the numbers

1. Search the top 3 queries from the log in a private window, on mobile and desktop.
2. Write down the title link Google shows for each: [query - displayed title].
3. If the displayed title is not your new title, mark the row as rewritten.
4. A rewritten row cannot prove or disprove the hypothesis. Fix the cause, then start a new window.

Common causes Google lists for rewrites

• Half-empty title, such as only the site name.
• Obsolete title, such as an old year that conflicts with the page.
• Inaccurate title that does not match what the page is about.
• Near-identical titles across many pages of the site.
• No clear main heading, or several headings with the same weight.
• Site name repeated, or a title in a different language from the page.

Quick fixes to log

• Match the H1 and the new title in meaning.
• Keep the brand short, at the start or end, after a separator.
• Remove repeated keywords. Google calls keyword stuffing out as a problem.

Rebuild the tags with the Meta Tag Generator and check the H1 and title together across the site with the Website Audit.

Need title formulas to test? Start from the Meta Title and Description Template.

Spreadsheet Header (CSV)

When to use: A paste-ready header row and allowed values to run the log in Google Sheets or Excel.

Paste-ready spreadsheet header

Paste this row into cell A1 of Google Sheets or Excel, then split text to columns by comma. One row per title change.

test_id,url,canonical_url,old_title,new_title,hypothesis,change_date,recrawl_date,before_start,before_end,after_start,after_end,filters,top_queries_before,branded_share_before,branded_share_after,clicks_before,clicks_after,impressions_before,impressions_after,ctr_before,ctr_after,position_before,position_after,displayed_title,title_rewritten,other_changes,season_note,control_url,verdict,owner,next_action,status

Allowed values

• title_rewritten: no, partly, yes
• verdict: keep, revert, extend, inconclusive
• status: planned, live, waiting, read, closed

Sheet tips

• Compute CTR in the sheet as clicks divided by impressions, so rounding stays the same.
• Highlight rows where position moved by more than one place. Read those with care.
• Highlight rows where title_rewritten is yes. They need a fix before any verdict.
• Filter by url to see every title a page has had, oldest first.

Worked Example (fictional coffee gear shop)

When to use: Two filled rows for an invented shop: one kept change and one title Google rewrote.

Worked example: Brindlecove espresso guides

Example only. Brindlecove is an invented coffee gear shop. Its URLs, titles, dates and numbers are made up to show a filled-in log.

Row 1 read: descaling guide

• Recrawl: Google showed the new title from 11 Sep. Days 2 to 14 Sep were skipped.
• Displayed title: the new title, for all 3 top queries checked.
• Before: 312 clicks, 14,200 impressions, average position 4.1.
• After: 405 clicks, 13,900 impressions, average position 4.0.
• Query mix: the same 5 top queries, with a similar share of impressions.
• Control: the cleaning tablets guide, unchanged, had flat CTR over the same windows.
• Verdict: keep. Clicks rose at the same position and query mix. Re-read after 8 weeks.

Row 2 read: grinders category

• Displayed title: Google showed "Coffee Grinders", taken from the page H1, not the new title.
• Verdict: inconclusive. The H1 was changed to "Burr Coffee Grinders for Espresso" on 20 Oct and a new window started.

Why this works

• Each row names one change, so the numbers point to one cause.
• The rewrite check stopped the team from judging a title searchers never saw.
• The control page shows the lift was not a season or update effect.