Blogging Titan publishes three kinds of page: original studies, tool pricing and comparisons, and how-to guides. This page explains how each one is researched and checked, so you can judge our work by how it was made. It also lists the mistakes our own checks have caught.
- 42 original studies
- Prices read from vendors’ own pages, dated
- Every correction logged
Original studies
A study starts with a question we cannot answer from existing sources, for example “how many of the pages Google’s AI Overviews cite rank in its top 10?” Before collecting anything we fix four things in writing: the list of queries or pages, the search settings, what counts as a hit, and the number we will report.
Where the data comes from
| Source | How we collect it | Example |
|---|---|---|
| Google results and AI Overviews | Searches run in a real browser with fixed location and language settings. We store the top 10 organic results and every AI Overview citation, with the date and the exact query used. | Rank-Citation Disconnect |
| ChatGPT and Perplexity answers | The same fixed question list put to each engine, with every cited URL recorded. | AI Citation League Table |
| Page archives | Earlier versions of a page from the Internet Archive’s Wayback Machine, compared line by line with the live page. | The Fake Refresh |
| Our own server | Every request from a named AI crawler logged on bloggingtitan.com. | AI crawler log |
| Vendor pricing pages | Each plan’s monthly, annual, introductory and renewal price, read from the vendor’s own site with the currency checked. | True annual cost |
How we keep the counts honest
Collection and judgement are kept apart. Whoever extracts the data records exactly what the page says, word for word, and makes no calls about whether it is right. Comparison happens afterwards, in one place, against rules written down beforehand.
When a page gives two figures and we have to choose one, we take the figure that is less helpful to our own conclusion and say so in the method section.
After a draft is written, a separate check recomputes every number from the raw data with one instruction: find errors. That check has changed published headlines (see the log below). Each study also ends with a limits section saying what the sample cannot tell you, and it reports results that went against our hypothesis.
Prices, reviews and comparisons
Every price on our pricing pages, comparisons and studies comes from the vendor’s own pricing page, never from another review, and carries the date we read it. Our pricing pages keep a dated history of each plan, so you can see when a price changed. We re-read them regularly and record the date of each check.
Many vendors show prices in your local currency. We read pages from the UK, so we check the currency symbol on every figure before it goes into a table. A missed pound sign once put Elementor’s prices on the site as dollars; that correction is in the log below.
Where a review says we used a product, the page should show it: our own screenshots, settings or results. Where we have not used a product, the page says it is a price and feature comparison built from the vendor’s published information. This is the same standard as our Proof-of-Use Test, and we hold our own reviews to it. Some reviews from 2023 predate this rule and are being brought up to it as they are refreshed.
Affiliate links are marked and disclosed on the page. A vendor cannot pay to change a verdict, a ranking or a price. See our affiliate disclosure.
How-to guides
The rule for guides is that every factual claim links to the study or source behind it, and every statistic names its source and the date of the figure. We hold to it because our own page-one census found that seven in ten statistics on ranking blogging pages cite nothing. Where we have our own data on a topic, the guide uses it. Older guides that fall short of this rule, mostly from 2023, are being sourced or merged as they come up for review.
Corrections log
These errors were found by our own checks or reported to us, and fixed on the article itself.
| Page | What was wrong | What changed |
|---|---|---|
| Teaser Rate Test | On two pages that quoted their entry plan at two prices, we had picked the price that matched the vendor without disclosing the other. | Both re-scored against us. The headline rate of wrong entry prices moved from 23% to 29%. |
| Ghost Author Test | A miscount in how the 15 unverifiable pages split by type, and two descriptions the data did not support. | Split corrected to 9, 2, 3 and 1; both descriptions rewritten. |
| Paraphrase re-roll | Six errors, including a source count that double-counted, a survivor list that left out YouTube, and a survival metric reported without its ceiling. | All six fixed. Two other posts that described the result as “86% of sources change” were corrected to the pairwise figure of 57%. |
| Divi alternatives and the Elegant Themes review | Elementor prices read in pounds were published as dollars. | Replaced with US dollar prices on 3 October 2026, with a dated note in the text and a redrawn chart. |
Spotted something wrong? Email contact@bloggingtitan.com with the page and the problem. Confirmed errors go on this list.
Related: All original research · Editorial policy · About Blogging Titan