Case study, Dubai
Emtech is now a cited source in five AI engines
Technical repair, AI crawler access, answer ready service pages, trust design and a lead path, in that order.
One client, one market, one set of queries. Not a promise for your site.
This page tells the story in the order it happened: the problem, the work, how we checked it, the result as it stands, the limits and what we would do differently. Where a detail is not something we can document, we leave it out instead of guessing.
The short version is that the result did not come from a trick. It came from removing the things that stopped engines reading the site, then making the pages clear enough to be worth citing. Where this page describes method in general terms, it says so.
Starting point
What was holding growth back
Crawlers blocked
Security and crawl settings kept search and AI bots out.
Unclean foundation
Sitemap, schema, canonicals and caching needed work.
Weak buyer fit
Structure and trust signals did not suit enterprise buyers.
Not answer ready
Service pages lacked clear answers and FAQs.
Unclear entity
Engines lacked context on what the company does.
No lead path
Calls to action and forms needed repair.
The problem
A considered purchase, and a site that engines could not read
Emtech sells IT annual maintenance contracts to businesses in the UAE. That is a considered purchase. The buyer is often an IT manager, an operations lead or a founder who needs equipment looked after, wants to compare providers and asks questions before making contact. Those questions now go to AI engines as often as to a search box, and the answer the engine gives shapes the shortlist.
The first problem was access. Security settings and crawl rules were keeping search and AI bots out of the site. A page that a crawler cannot fetch cannot be indexed, summarised or cited, however good the writing is. This is a common hidden fault, and it is easy to miss because the site looks fine to a human visitor.
The second problem was the foundation underneath the pages. The sitemap, structured data, canonical tags and caching all needed work. None of these is glamorous, but together they decide whether an engine can understand which URL is the right one, what it contains and how fresh it is.
The third problem was fit. The structure and trust signals of the old site did not suit enterprise buyers. The service pages lacked clear answers and FAQs, so even a crawler that got in found little it could lift into an answer. The company entity was unclear, and calls to action and forms needed repair, so a visitor who did arrive had no reliable way to ask for help.
- Access: bots could not read the pages
- Foundation: sitemap, schema, canonicals, caching
- Content: no direct answers or FAQs
- Conversion: weak calls to action and forms
Access
Bots could not read the pages
- Security and crawl settings blocked bots
- Invisible to a human visitor
Foundation
Pages hard to interpret
- Sitemap and schema needed work
- Canonicals and caching needed work
Content
Nothing to lift into an answer
- No direct answers
- No FAQs
Conversion
No reliable way to ask
- Calls to action needed repair
- Forms needed repair
The fix
One system, in order
The order matters. Each layer depends on the one below it.
Answer sections, entity signals, schema.
Headings, process, FAQs, comparison context.
Clear hierarchy, service flow, mobile readability.
Crawl access, sitemap, schema, canonicals, caching.
Doing layer four before layer one is a common way these projects waste money.
The work, layer one
Technical repair and AI crawler access
Make the site readable before making it better.
This is how we approach crawler access on any site. We check what each crawler can actually fetch. That means reading the crawl rules and the security layer, because a firewall or bot protection setting can block a crawler even when robots.txt looks permissive. Search crawlers and AI crawlers are separate user agents with separate jobs. OpenAI documents separate crawlers for search, for model training and for actions a user asks for in ChatGPT. Perplexity and Anthropic likewise publish separate bots for search and for user requests. Which of these to allow is a business choice, so we set it deliberately and write it down.
Next we clean the foundation. A sitemap should list the URLs you want indexed and no others. Canonical tags should point each page at one preferred address. Structured data should describe what is visible on the page and nothing more. Caching should serve fast pages without serving stale or broken ones to crawlers.
We keep to a safe order: fix the thing that blocks reading first, confirm the fix with a fresh fetch, then move on. Changing many things at once makes it impossible to tell which change helped, and it makes a rollback harder if something goes wrong.
- Crawl and security settings reviewed per bot
- Sitemap, canonicals and schema cleaned
- Caching checked for crawlers
- Each fix confirmed by a fresh fetch
Test what each crawler can reach.
Open access, repair sitemap and tags.
Fetch again and confirm.
Write down what changed and why.
The work, layers two and three
Redesign for trust and a structure engines can follow
Enterprise buyers read differently.
The redesign was built around how an IT buyer evaluates a provider. That means a clear hierarchy so the service and the company are obvious in seconds, a service flow that explains what is covered and what happens after contact, and mobile readable layouts, because a decision maker often opens the first page on a phone between meetings.
Behind the visual change sat a content structure. Each service page got headings that match real questions, a clear description of the process, FAQs and the comparison context a buyer needs to tell one provider from another. This helps people first. It also gives an engine labelled sections it can lift cleanly, which is the practical meaning of answer ready.
We are careful not to write for engines at the cost of readers. A page that reads like a keyword list gets neither cited nor trusted. Our test is simple: could an IT manager skim the page and learn what is covered, what it involves and how to start, without needing to call?
- Hierarchy and service flow for enterprise buyers
- Headings that match real buyer questions
- FAQs and comparison context on service pages
- Mobile readable layouts
The work, layer four
Answer engine readiness and a clearer entity
Say who you are, what you do and for whom, in the same words everywhere.
Engines build an internal picture of a company from many places. If the website, directory listings and mentions describe it differently, the picture is vague and the engine hedges or leaves the company out. So we work on entity clarity: consistent naming, a plain description of what the company does and the same core facts wherever they appear.
On the service pages we add answer sections: a short, direct response to the question a buyer asks, placed under the heading that asks it, followed by the detail. We align structured data with what the page visibly says. Google states that no special schema or extra requirements apply to AI Overviews and AI Mode beyond normal SEO, with pages indexable and eligible to show a snippet. So we treat schema as honest description, not as a lever.
This layer cannot work alone. It sits on top of access, a clean foundation and pages worth reading. That is why it comes last.
- Consistent entity description and naming
- Direct answer sections under question headings
- Structured data matching visible content
- No claims the page cannot support
YouWhich companies offer IT AMC services in the UAE?
Buyers comparing IT AMC providers usually look for scope, response terms and a clear process. Emtech describes all three on its service pages.
Lead path
Turning visits into enquiries
Visibility without a way to ask is wasted.
- 1
Repair the call to action
Each page gets one clear next step, placed where a buyer decides to act.
- 2
Fix the forms
Forms are repaired so an enquiry reaches the team reliably, and tested end to end.
- 3
Keep the human handover
Questions the bot cannot answer go to a person, so a buyer never meets a dead end.
Proof
AI answers from live searches
Captured on the day. AI answers change, so we repeat checks.
Measurement
How we measured it
The measure here is direct observation. We ran buyer style queries for IT AMC services in the UAE in each engine and captured the answers: Google AI Mode, ChatGPT, Gemini, Microsoft Copilot and Perplexity. The six screenshots above are those captures. Emtech appears as a cited source or a named provider in each of the five engines. The Copilot comparison table is a separate capture where the engine lays providers side by side.
That method has a clear weakness, and we state it. A screenshot proves what one engine said to one query at one moment. AI answers vary by user, place and time, and the same query can produce a different list tomorrow. So we do not claim a fixed rank. We claim a pattern: across five engines, for the queries we tested, the company is now treated as a source. To keep the claim honest, we repeat the checks and look at whether the pattern holds.
We also separate the types of evidence. A named mention in an answer is not the same as a link in the source list, and neither is the same as a visit. If you commission this kind of work, ask which one a report is counting.
What a report is counting
Result
Where Emtech stands now
As documented in the captures above.
One client, one market, one set of queries. Not a forecast for your site, and not a guarantee of any future citation.
Limits
What this study does not show
It is not a ranking guarantee
Citation in an AI answer is not a fixed position. Answers shift, and Emtech may be listed in one run and absent in another.
It does not isolate one cause
Several things changed together: access, structure, content and design. We cannot say how much each contributed, and we will not pretend to.
It covers one query theme
The evidence is for IT AMC services in the UAE. It does not tell you how the company performs on unrelated queries.
It is not your market
A different trade, a more crowded field or a different region will behave differently.
What this study shows and does not show
- ✓A pattern across five engines for tested queries
- ✓Direct observation, captured on the day
- ✓Evidence for IT AMC services in the UAE
- A fixed ranking or permanent position
- Proof of which single change caused it
- A forecast for your site
Hindsight
What we would do differently
Capture a fuller baseline first
We would record the same five engine checks before any work begins, so the before and after sit side by side.
Log checks on a schedule
A fixed monthly set of dated captures makes the pattern easier to defend than a handful of screenshots.
Change in smaller batches
Where possible we would ship changes in smaller groups, to learn which ones matter most.
Before and after
What changed on the site
| Area | Before | After |
|---|---|---|
| Crawler access | Search and AI bots kept out | Access set deliberately per bot |
| Foundation | Sitemap, schema, canonicals, caching needed work | Cleaned and verified |
| Service pages | No clear answers or FAQs | Direct answers, process, FAQs, comparison context |
| Buyer fit | Structure unsuited to enterprise buyers | Hierarchy and trust signals built for them |
| Entity | Unclear to engines | Consistent description of who and what |
| Lead path | Calls to action and forms needed repair | Working path from page to enquiry |
FAQ
Emtech case study questions
What does IT AMC mean?
IT AMC is an IT annual maintenance contract, where a provider looks after a business equipment and systems for a fixed term. Emtech offers these in the UAE.
Is Emtech guaranteed to appear in AI answers?
No. Answers vary by user, place and time. The study shows a pattern across five engines for the queries we tested, not a permanent position.
Was it only a redesign?
No. Technical repair and crawler access came first. The redesign, content structure and answer readiness built on that base, and a lead path finished the job.
Why were the crawlers blocked?
Security and crawl settings kept search and AI bots out. This is common, and it is invisible to a human visitor, which is why we test per crawler.
Could my site get the same outcome?
We cannot promise that. If your site has similar blocks and gaps, the same order of work applies. An audit tells you what is actually in your way.
Keep reading
Guides on this topic
Why your business is not showing in ChatGPT, and how to diagnose it
If ChatGPT never names your business, one of seven causes is usually behind it. Use this diagnostic to find which one, in the order that saves the most time.
Read moreAI search traffic in Google Analytics: how to track ChatGPT and AI referrals in GA4
GA4 can show some visits from ChatGPT and other AI tools, but not all. This guide explains how channels and referrers work, how to set up a report, and where the data stops.
Read moreAI visibility metrics: citation share, share of voice and mention rate defined
Mention rate, citation rate, citation share and AI share of voice are often used loosely. This guide defines each one, shows how to compute it from a fixed prompt set, and explains the limits.
Read moreSimilar audit
Want your site search ready and AI visible?
Tell us about your site.
- Technical and AI access
- Content and entity review
- Lead path review
WhatsApp: +91 63777 67206
Email: hello@thegrowingai.com