iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
The 140% figure is a claim, not a benchmark you can plan around. A July 25, 2026 article in The Tech Edvocate says an embedding-focused optimization strategy produced an average 140% increase in AI citations in client tests linked to Brainlabs clients. The article does not report the sample size, test period, platforms, baseline, control group, or calculation method, so no one outside those tests can check the number. The useful takeaway is narrower: you need a defined, repeatable way to measure your own brand’s AI visibility before you can tell whether any change helps.
What the 140% claim does and does not establish
The figure appears in a secondary article, “Boost AI Search Visibility 140%: The Metric Your Brand Needs” in The Tech Edvocate, published July 25, 2026. Read it as follows:
- It is attributed, not verified. The article credits the increase to unnamed client tests. No primary study or Brainlabs report was located for this claim.
- It cannot be reproduced from the article. Without a baseline, a comparison group, or a dated observation window, the 140% cannot be separated from ordinary fluctuation in AI answers.
- It does not isolate a cause. The article recommends several content practices, but it does not show how much each one contributed to the reported increase.
- It does not transfer to your brand. A lift measured in one set of client tests says little about a different industry, site, or set of queries.
Define “AI visibility” before you measure it
The article uses “AI citations” to mean an AI-generated answer that uses or refers to a brand’s content, and it notes that this may not produce a link or a site visit. It does not define a standard metric. “AI visibility” can therefore mean several different outcomes, and they should be tracked separately because each one answers a different question.
| Outcome | What it tells you | What it does not tell you | How to record it |
|---|---|---|---|
| Brand mention | The AI answer names your brand or product in its text. | Whether the answer linked to your site or drove a visit. | Count mentions per query, per platform, per run, using the rule you wrote beforehand. |
| Linked citation | The answer includes a link or source pointing to a page on your domain. | Whether the user saw the link or clicked it. | Count links to your domain in the answer or its source panel. |
| Answer inclusion | Your content appears to have informed the answer, whether or not it is named. | Which passage was used, unless the platform shows sources. | Judge by matching the answer’s claims to your page. Record the judgment rule. |
| Impressions | How often your content surfaced in an AI interface. | Whether it was used in the answer. Availability varies by platform, and the article does not address it. | Use only where a platform reports it. Note the reporting definition. |
| Referral traffic | Visits from AI platforms recorded in your analytics. | Answers that were seen but not clicked, which is the most common case where a link appears. | Segment referral sources in your analytics tool and note the referrer names you find. |
Report these as separate numbers. A rise in brand mentions with no change in linked citations is a different result from a rise in referral traffic, and only the first two can be read directly from the answers themselves.
#1 Best Overall
How to measure your brand’s AI search visibility
The question “How can I measure my brand’s AI search visibility?” is answered by a fixed test you can rerun. The steps below are editorial guidance for making a claim testable. They are not a method taken from the article.
- Choose the platforms. List each AI product you will check and record its name, the date you test, and any version or model label the interface shows. Results from one platform do not transfer to another.
- Build a fixed query set. Start with a small list of questions that real buyers ask about your category, including brand-specific and comparison questions. Save the exact wording. Do not edit the list between measurement periods.
- Set the observation schedule. Pick fixed dates and times for each run. AI answers can vary from one run to the next, so run each query more than once per period and record every run rather than the best one.
- Write the counting rules first. Decide in advance what counts as a mention (for example, your brand name appearing in the answer text), what counts as a linked citation (a link to your domain), and how you will judge answer inclusion. Apply the same rules to every run.
- Record a baseline. Measure at least one full cycle before you change any content. The baseline is what every later figure is compared against.
- Compare like with like. Repeat the same queries on the same platforms on the same schedule. If you change more than one page at a time, you cannot tell which change, if any, produced the difference.
- Cross-check with analytics. Compare the citation counts with referral traffic from AI sources. A rise in mentions that does not appear in referral data is a finding to note, not a failure of the test.
Strategies the source recommends, and how to test them
The Tech Edvocate article recommends four practices. The article presents them as recommendations, not as the proven cause of the reported 140% change. Each can be tested on your own pages using the measurement plan above.
Anticipate related subquestions
Map the follow-up questions a buyer asks after the first one, such as pricing, compatibility, alternatives, and setup steps. Answer the most common ones on the same page, or link to dedicated pages that answer them directly. To test this, add the subquestions to your query set as separate rows, then compare coverage before and after the change.
Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minutePC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Write clear, complete answers
Open each section with a direct answer in one or two sentences, then add the context, conditions, and exceptions. AI systems can only use what the page states plainly, so vague introductions and unstated assumptions make it harder for a system to lift an accurate passage. To test this, rewrite one page, keep its URL unchanged, and track its linked citations against your baseline for the same queries.
Use precise terminology and structured data
Use one consistent name for each product, feature, and category across the site, and define any term your audience may not know. Add structured data, such as schema.org markup, only where it matches the visible page content. Markup that describes content a visitor cannot see is a quality risk. Validate the markup with a structured data validator before publishing, then test the page as you would any other change.
Refresh important pages on a schedule
Set a review date for pages that carry prices, specifications, statistics, or dates. Update the facts, remove outdated figures, and record the revision date on the page. Because the article gives no evidence on how often refreshes matter, treat the schedule as a maintenance practice and measure its effect the same way as the other changes.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.When results do not move
A flat or negative result is common in a first test. Check these points before concluding that a change did not work:
Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Clear out junk files and repair common Windows errors3Scan for outdated or missing drivers - takes under a minute- Too many changes at once. If several pages or tactics changed in the same period, the result cannot be attributed to any one of them.
- Query drift. Confirm that the wording of the queries, the platforms, and the schedule were identical to the baseline.
- Changed counting rules. A new reviewer or a loosened definition of “mention” can move the numbers without any change in visibility.
- Window too short. One or two runs can reflect normal variation. Extend the observation period before drawing conclusions.
- Pages not accessible. Check that the updated pages load for crawlers and are not blocked by robots directives or a login wall.
- Platform change. Note the date of any visible platform update during the test, since it can change how answers are built or sourced.
How to report a visibility result
If you publish a percentage from your own tests, include the information needed for a reader to judge it:
- The platforms, query set, and number of runs
- The observation dates and the baseline period
- The counting rules for mentions, linked citations, and answer inclusion
- Which changes were made, and whether they were made together
- Whether referral traffic moved in the same direction as citations
Without these details, a figure such as “140%” describes an outcome you cannot reproduce, which is the same limitation that applies to the figure in the source article.
Quick Recap
“
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
