The Open Index · Methodology
How a number earns its place in the Open Index.
The rules we follow, the grades we give, how we handle studies that disagree, and what we refuse to publish. If we break a rule, tell us.
6 steps
from source to published number
Webtec Research
Written byKris GjorgjioskiKeyword research and industry benchmarks, Webtec Research
Checked byHristijan NajcheskiHead of SEO and AEO (technical and AI SEO), data verification
Updated October 15, 2026 · Edition 1Refreshed quarterly · next January 2027
Built from research on
Google
OpenAI
Perplexity
Gemini
Anthropic
Cloudflare
Semrush
THE PIPELINE
Six steps between a source and a chart.
- 1
Find
A researcher finds the primary source. Secondary articles and aggregator blogs are leads, never sources.
- 2
Read
We read every figure on the source’s own page and record publisher, title, date, period, sample and method.
- 3
Grade
Grade A: independent, official or peer-reviewed. Grade B: vendor or agency study with a disclosed method.
- 4
Compare
Where studies overlap, we put them side by side and explain the gap instead of picking a winner.
- 5
Check
Hristijan N. traces each number back to its source before publication.
- 6
Publish and log
Pages go live with an update date and a correction log. The whole Index is refreshed quarterly, and changes are dated and explained.
THE GRADES
Who is telling us, and what do they have to gain?
Independent, official or peer-reviewed
Research organizations with no product to sell, public datasets, platform documentation and platform operational data, and academic papers. We still read the method and note limits.
Vendor or agency study with a disclosed method
Software vendors and agencies publish useful large datasets, and they have an interest in the topic. We label them, state the sample and period, and never present them as independent.
WE PUBLISH
- Figures read on the primary source
- The date, period, sample and method beside every number
- Studies that disagree, side by side, with the reason
- Gaps, shown as gaps
- A grade and a “read with care” note for every source
WE LEAVE OUT
- Numbers we saw only in secondary articles
- Single-site results presented as industry benchmarks
- Multipliers (for example “AI traffic converts 4x to 23x better”) we cannot trace to a dataset
- Industry percentages attributed to a study we could not find on the publisher’s own pages
- Guesses that fill a gap in the data
WHEN STUDIES DISAGREE
Four reasons honest studies give different numbers.
Different samples
A panel of 900 people, a keyword database and a network of customer sites measure different things.
Different periods
AI search changed fast in 2025. A March and a December figure describe different products.
Different definitions
A “click”, a “citation” or an “industry” can be defined several ways.
Different methods
Parsing, filtering and forecasting choices change results, and authors sometimes change them between editions.
Our rule: compare trends and rankings inside a study, and compare direction across studies. Never mix absolute numbers from different studies in one calculation.
REFRESH SCHEDULE
Refreshed every quarter, on a fixed calendar.
AI search moves fast, so a benchmark that is a year old is a liability. Each edition re-checks every source.
- Edition 2JanuaryPlanned for January 2027
- Edition 3AprilPlanned for April 2027
- Edition 4JulyPlanned for July 2027
- Edition 5OctoberPlanned for October 2027
Re-check
Every source is opened again. Figures that changed are updated, with the old value in the log.
Add and retire
New studies are graded and added. Outdated or superseded ones are retired and noted.
Interpret again
Our interpretation blocks are re-read against the new numbers and rewritten if the picture changed.
WEBTEC RESEARCH AND OUR OWN WORK
First-party findings come later, and only when they are safe.
We also learn from managing AEO and SEO programmes. We publish such findings only when they are anonymised, aggregated, approved by leadership and checked by Hristijan. No such data is in this edition: everything on these pages comes from published, cited sources.
CORRECTIONS
Found a mistake? We fix it in public.
Email hello@trywebtec.com with the page, the number and your source. Corrections are dated and explained in the log.
2026-10-15: Edition 1 published with 19 sources.
Next scheduled refresh: January 2027.
Preferred sources on Google
Make Webtec a preferred source on Google.
Pick us once, and Google shows you more of our research and guides in Top Stories when we publish. It takes one click and you can change it at any time.
A LIVING REFERENCE
Refreshed every quarter.
We re-check every source in January, April, July and October, add new studies and retire old ones. This is Edition 1 (October 15, 2026); the next is due January 2027. Changes are logged on the methodology page.
