I ran my tool on the people who rated it gold
Azim Marketing Agent — ten AI agents analysed the KANZ platform in Saudi Arabia and produced a thirteen-section report in four minutes. Here is the whole thing, including what it got wrong.
The idea
Why this site
PlainlyThe easy move is to show a report for a site I picked because it would come out flattering. I did the opposite: a site that knows me, knows the tool, and can judge whether every line in it is accurate.
KANZ is a Saudi AI recruitment platform, and the organiser of the KANZ AI Hackathon — where this very agent received a Gold rating, certified by the Lebanese American University.
So what you read below is not a pitch whose angle I chose. It is what came out of the tool when I pasted one URL and pressed one button, on a site in an industry that is not mine, a market that is not mine, and a country that is not mine.
Background
The KANZ AI Hackathon
This tool was not built for this page. It was built to compete in the KANZ AI Hackathon, and that is where its rating came from — a judging panel that did not know who built it.
Why this matters hereA hackathon at that scale means the rating came out of a comparison against thousands of entries, not one person’s opinion. That is the difference between an attendance certificate and a judged result.
What the judges scored
Azim Marketing Agent received a Gold rating at the KANZ AI Hackathon, certified by the Lebanese American University under reference KANZ-ADV-6995, and was listed as a featured project in the event’s AI showcase.
Innovation
Ten agents running in sequence along one pipeline, each building on the output before it — not separate calls stitched together at the end.
Technical execution
A working tool on a live bilingual site, with a report system, usage limits and provider failover — not a demo prototype.
Impact
What takes a marketing consultant weeks of manual research comes out in four minutes, free, for anyone who owns a website.
An event of unusual scale
During the same event I took part in setting a Guinness World Records title, in its official wording: “Most participants in an AI video lesson” — with 14,075 participants, on 15 July 2026.
To be clearThe record belongs to KANZ and the Saudi Ministry of Human Resources and Social Development, not to me. I was one of 14,075 attendees. The value here is the scale of the event inside which my project earned a Gold rating — not the attendance certificate itself.
Two separate numbers, never to be conflated: 4,535 entries submitted to the hackathon, and 14,075 attendees at the Guinness-recorded lesson.
How it works
What happens after the button
The visitor pastes a URL. Ten agents then run in sequence, each one building on what came before — not ten separate questions answered in isolation.
Crawl and read
Opens the site, reads its copy and structure, and works out the business, its audience and its geography.
Research and context
Finds the real competitors in that same market, and gathers the context that feeds every step after it.
Measure and diagnose
Six areas scored separately, each with a written reason rather than a number left hanging.
Build the plan
A ten-channel strategy, ad campaigns with their copy, a thirty-day content plan, and a three-month roadmap.
Agent one
How it read the business
This is the step everything else rests on. Get it wrong here and the entire report points in the wrong direction.
Why this mattersThe agent was given nothing about KANZ. It read “Riyadh, Jeddah and Dammam”, “Vision 2030” and “five days” off the site itself — and those specifics are what separate a real analysis from a filled-in template.
Diagnosis
Marketing health — six areas
Overall 68 out of 100. Each area is scored on its own, because a site can be strong on brand and weak on trust — and those need entirely different treatment.
SEO
The site captures high-intent local keywords well, but lacks long-tail content covering Saudi labour law compliance and the technical side of integration.
Content
The messaging is sharp and aligns well with Vision 2030, though it needs more case studies to prove the five-day turnaround to sceptical HR managers.
Brand
The positioning as a premium AI-driven leader is clear, and separates the brand effectively from competitors that still lean on manual work.
Trust
The 200+ companies metric is strong, but the absence of visible, detailed integration documentation creates friction for enterprise decision-makers.
Conversion
The five-day promise is a compelling hook, but the path from landing page to demo could be improved with interactive ROI tooling.
Ads
Search campaigns are well targeted at the Riyadh and Jeddah hubs, but the creative needs to address the “AI replacing humans” fear more directly to lift click-through.
Competitive landscape
Three real competitors
This is the hardest test for the tool. An invented competitor is spotted in a second; the right ones mean the agent understood the market, not just the company name.
| Competitor | Strength | Weakness | Your opening |
|---|---|---|---|
| Bayzat | Strong regional brand presence and comprehensive HR and payroll integration. | No specialised, end-to-end AI recruitment automation. | Lead with the five-day metric against their manual-heavy recruitment modules. |
| LinkedIn Talent Solutions | An unmatched database of professional profiles and global market dominance. | High cost, and no localised automation for Saudi hiring workflows. | Position as the lower-cost local alternative that automates the screening they charge for. |
| Oracle Taleo | Enterprise-grade reliability and deep integration with existing ERP systems. | Complex implementation and slow adoption because of a legacy interface. | Emphasise ease of use and speed of deployment against their heavy setup. |
Priorities
Four openings, ranked by return against effort
The order is the point: highest impact and lowest difficulty first. A list that does not rank itself leaves the reader exactly where they started.
Vision 2030 compliance integration
Build automated Saudisation compliance checks into the recruitment workflow. It answers a live regulatory pain point for local HR managers and ties the brand directly to national digital transformation mandates.
Five-day turnaround case study series
Publish detailed, data-backed case studies from companies already on the platform. It supplies social proof and turns the five-day claim from a promise into a number that can be checked.
Localised SEO content hub
A content strategy targeting high-intent terms such as “AI recruitment automation Saudi Arabia” and “automated hiring platform Riyadh”, to capture decision-makers who are already searching.
Integration API documentation
A dedicated technical portal covering integration with the common regional HR systems. It removes the barrier for enterprise buyers who fear data silos and long implementations.
Search terms
Chosen by buying intent, not search volume
A term with a million searches and no buying intent brings traffic, not customers. The agent ranks by intent first, and writes down why each term was picked.
Limits
What the agent doesn’t do — and one real mistake
A tool that overstates itself is worse than no tool. These are the actual limits, written down before you find them yourself.
One mistake in this very report
In the content plan, day thirteen, the agent proposed a piece on “HR tech trends shaping the Saudi market in 2024” — and the current year is 2026. Language models inherit their training year, and do not know today’s date unless it is passed to them.
I have left that error visible rather than quietly editing it out, because removing it would make everything above it unverifiable.
And the limits that are known
- The projected figures are estimates, not measurements. The report’s before-and-after numbers come from comparison with similar businesses, not from measuring your site — which is why they are omitted from this page entirely.
- The analysis reads what is published publicly. It cannot see your ad accounts or your analytics, so its read on advertising is inference from what is visible.
- The report is a starting point, not a verdict. It compresses weeks of manual research; it does not replace human judgement that knows your circumstances.
- Running it again may give a different result. The analysis is a snapshot of its moment, and the site itself changes.
Ashraf Azim
Digital marketing consultant in Dubai who builds his own AI tools rather than renting them. Five free tools running live on this site in two languages, one of them registered as intellectual property with the UAE Ministry of Economy.
Run it on your own site
The same tool, the same four minutes, the same thirteen-section report. Your URL is the only thing it needs.
