Versus Article Experiment Plan Template

Experiment Plan Sheet (fill-in)

When to use: One sheet per X vs Y test: fill it in before you open either tool, then keep it open while you test.

Versus test plan

Write the plan first. If a field changes during the test, note the change and the reason, don't overwrite it.

Run log (one row per task per tool)

Next action

• Owner: [tester] - action: [run all tasks in both tools] by [date]
• Reviewer: [editor] - action: check the run log and evidence before the article is drafted

Completion check

• Both tools got the same input pack, on the same dates.
• The plan match is written down, with the reason.
• Every task has a row for both tools, including failures.
• Every result has a dated screenshot or export.
• Prices have a capture date and currency.
• Caveats and disclosures are listed before the verdict is written.

Matched Plans and Cost Capture

When to use: Pick plans that are a fair match and record the real cost of running your tasks, so the price section holds up.

How to match plans

• By price: pick the plan from each tool closest to the same monthly spend.
• By limits: or pick the cheapest plan in each tool that can run every task in your list.
• Choose one rule, write it in the plan, and use it for both tools.
• Use the same billing period for both: monthly vs monthly, or annual vs annual.
• Test on the paid plan when you can. If you use a trial, check it has the same limits as the plan you name.

Cost capture sheet

Rules

• Screenshot each pricing page on the day you test. Prices and plan names change.
• Quote list prices, not a private discount you were offered.
• If a task needed an add-on, count it in the cost for that tool.
• Say in the article when you last checked prices.

Task List and Same Inputs

When to use: Write the tasks and build one input pack, so both tools face exactly the same work.

Build the input pack

• Seed keywords: [the same 20 to 50 keywords], saved as one file.
• Test domain and competitors: [your site + 2 competitors].
• Location, language and device settings: [US, English, desktop].
• Files to import: [CSV, sitemap, brief], with the same file name and version.
• Account settings: defaults, unless a task says otherwise. Note any change.

Need a seed list? Pull one from Keyword Research and freeze it as the input file before testing starts.

Task list

Rules for fair tasks

• Pick tasks from the reader's job, not from either tool's feature list.
• Run the tasks in the same order in both tools, or swap the order on purpose and note it.
• Run both tools on the same day, so data is from the same point in time.
• If a tool can't do a task, record "not possible" with a screenshot. That is a result.

Outcome Capture and Evidence Log

When to use: Record what happened in a way an editor, or a sceptical reader, can check later.

What to capture per task

• Result: pass, partial or fail, and one line on what you saw.
• Effort: minutes and number of steps or screens.
• Output: the export, report or screen the reader would use.
• Numbers the tool shows: copy them with the date, they change by the day.
• Surprises: errors, limits hit, confusing labels.

Screenshot rules

• File name: [TestID_TaskID_Tool_YYYY-MM-DD].png, for example VS-007_T2_X_2026-10-28.png.
• Show the date or the browser address bar where you can.
• Capture failures too, not only good screens.
• Keep the raw file. Crop or blur private data only in the copy used in the article.
• Record the app version or release note date if the tool shows one.

Google's guidance on reviews asks for evidence such as visuals, audio or links of your own experience, and for quantitative measurements. This log is where that evidence comes from.

To compare against the pages already ranking for the versus keyword, review the results with the SERP Analyzer and note what evidence they show.

Caveats, Bias and Disclosure

When to use: List what could tilt the result before you write the verdict, and say it plainly in the article.

Caveat checklist

Disclosure

• The FTC says affiliate relationships must be disclosed clearly and conspicuously, close to the recommendation.
• Per FTC guidance, a single homepage disclosure is not sufficient, and one at the very top or bottom of a page may be missed.
• Free access given by a vendor in exchange for a review is also a connection to disclose.

What not to claim

• Results beyond what you tested: "best for everyone" from 30 keywords.
• Accuracy rankings without a stated method.
• Speed or uptime as fact from one run.

Results Record and Article Handoff

When to use: Turn the run log into a short results record the writer can use, then move to the article outline.

Results summary

Verdict notes for the writer

• Pick X if: [reader situation backed by tasks]
• Pick Y if: [reader situation backed by tasks]
• Cost at the matched plans: [X total] vs [Y total], captured [date]
• Caveats to print: [top 2 or 3]
• Retest date: [date, or next major release]

Next step: write the article

Take this record into the Comparison and Versus Article Template for the writing structure: verdict, criteria, deep-dives and use cases.

Draft headline options with the Comparison Article Title Generator, and plan sections with the Content Brief tool.

Worked Example (fictional rank trackers)

When to use: A filled-in plan for an invented publisher testing two invented rank trackers for a versus article.

Worked example: Tallyrank vs Mossgauge

Example only. The publisher, tools, people, prices and results are invented to show what a filled-in plan looks like.

Results record (fictional)

• T1: both pass. Tallyrank 9 min, Mossgauge 14 min.
• T3: Mossgauge needed the add-on, so its real cost was $34/month.
• T4: Mossgauge export had no device column. Marked partial, screenshot kept.
• T5: both answered correctly, Mossgauge in 2 hours, Tallyrank next day.

Why this plan is fair

• Plans matched by one written rule, and the add-on was counted in cost.
• The tester's familiarity with Tallyrank is printed as a caveat.
• Failures have screenshots, not just wins.

Want to check your own tracking setup while you test? Track the same keywords in the Rank Tracker as a third reference point.