Dealer Group AI Shortlists in Malaysia: Beyond Paid Ads

Your group runs a dozen or more showrooms across the Klang Valley, Penang and Johor Bahru, carries several marques, and wins the auction for most brand and "car dealer near me" keywords. Meta lead forms keep the sales advisers busy. Yet when a fleet manager or a family asks ChatGPT which authorised dealer to visit for a test drive in Shah Alam, the answer names two rivals and a used-car platform. That is the gap in dealer group AI shortlists, and it is not one your media agency can bid its way out of.

This article explains why paid search leaders go missing from AI answers, the prompts Malaysian car buyers actually type, a five-step audit you can run this week, and the fixes that make a multi-outlet group easier for ChatGPT, Gemini and Perplexity to find, describe correctly and recommend.

Why dealer groups that win on ads miss AI shortlists

Dealer groups miss AI shortlists because the assistants that build them do not read your ad auctions. They read pages, profiles and articles they can crawl, and they name the dealers those sources describe most clearly.

Two inputs, neither of them paid

An AI assistant answers from two places. One is what the model learned in training, which is fixed at a point in time and often vague about which outlet sells which marque. The other is live retrieval: the assistant searches the web, reads current pages and cites some of them.

The engines document their retrieval crawlers. OpenAI says its OAI-SearchBot is used to surface websites in ChatGPT's search features, and that sites which opt out will not be shown in ChatGPT search answers. Perplexity says PerplexityBot is designed to surface and link websites in its search results. Google says that to appear as a supporting link in AI Overviews or AI Mode, a page must be indexed and eligible to show with a snippet, with no additional technical requirements. None of these routes involve your Google Ads or Meta account.

Why landing pages built for leads do not get cited

Most dealer group sites are built for conversion. Model pages carry a hero banner, a price "from" figure, a lead form and a WhatsApp button. Outlet details sit in a store locator widget loaded by script, and service menus are PDFs. That works for paid traffic but leaves very little text for an assistant to quote.

Google's AI features guidance lists "making sure that important content is available in textual form" as a basic practice. A buyer's question needs a group name, a marque, an outlet, a town, opening hours and proof of authorised status. If those sit together in a sentence on paultan.org, a Carlist.my dealer profile or a forum thread, that page is the easier citation.

Why a group is harder to describe than a single dealer

Groups trade under several names: the listed or holding company, each franchise subsidiary, the showroom brand and the marque's own dealer locator. Outlets open, move and change franchises. The market itself is busy: The Star reported that Malaysia's total industry volume reached 820,752 units in 2025, with battery electric vehicle sales pulled forward before incentives on fully imported BEVs expired at the end of that year. New EV brands mean new franchises, new outlets and more chances for AI tools to confuse which of your companies does what.

What buyers in Malaysia actually ask AI

Malaysian car buyers ask AI about dealers by marque, location, service and trust, in English, Bahasa Malaysia and Mandarin. Fleet and family buyers ask different questions, so test both.

Shortlisting a dealer

• "Best authorised Honda dealer in the Klang Valley for a test drive this weekend"
• "Pengedar Proton sah di Johor Bahru yang ada servis dan alat ganti" (Authorised Proton dealer in JB with service and spare parts)
• "Where can I test drive a BYD or Chery EV in Penang?"

Comparing outlets and offers

• "Which dealer group in Malaysia handles corporate fleet purchases for 50 cars?"
• "吉隆坡哪家丰田经销商的售后服务比较好？" (Which Toyota dealer in KL has better after-sales service?)
• "Is [outlet name] an authorised dealer or a recond importer?"

Checking after-sales and trust

• "Authorised service centre for Mazda near Shah Alam open on Sunday"
• "Complaints about [group name] service centre"

Notice that only two of these prompts would be answered by a model page. The rest need outlet, service and reputation facts, which is exactly what conversion-built sites leave out.

How to check where you stand

You can build a baseline in a week with a spreadsheet, a fixed prompt list and two colleagues from sales and aftersales. Our measurement method uses the same approach with larger panels.

Step 1: Check crawler access on every domain

Groups often run a corporate site, a site per marque and campaign microsites built by different vendors. Check robots.txt on each for OAI-SearchBot, PerplexityBot and Googlebot, then ask IT whether the CDN or bot-protection service blocks or challenges them. A blocked marque site is a common reason one franchise is visible and another is not.

Step 2: Write prompts per marque and region

Write 5 to 10 prompts per marque, covering each region where you have outlets, both buyer types (family and fleet) and each language your sales team hears. Include prompts that name your group directly, so you can check accuracy as well as presence.

Step 3: Run each prompt several times

Run each prompt three times in ChatGPT, Gemini and Perplexity, logged out or in a clean account. Answers vary between runs and users, so one screenshot proves little. Note whether the assistant searched the web, because that tells you live pages are in play.

Step 4: Log five things per answer

For each run, record: whether your group or outlet is named, whether the marque and authorised status are right, whether the address and hours are right, which competitors appear, and which sources are cited. The cited-sources column shows you which motoring sites, directories and review pages the assistants trust for your marques.

Step 5: Compare with your own outlet pages

Pick the prompts where you were missing or described wrongly. Open your page for that outlet and ask whether a stranger could answer the prompt from the visible text alone, without the locator widget, a PDF or a phone call. Where the answer is no, you have your first fix list.

How to fix it

Make your own pages and profiles the clearest record of which outlet sells and services which marque, then make sure trusted third parties say the same thing. The steps below are written for multi-brand groups.

Publish one fact page per outlet and marque

Each outlet needs a permanent HTML page with a crawlable URL. In the opening lines, state the outlet name, the operating company's legal name, the group it belongs to, the marques it is authorised for, whether it is a 3S or 4S centre (sales, service, spare parts and body and paint), the full address, opening hours for sales and service, and the official phone or WhatsApp number. Keep the store locator, but do not let it be the only place these facts exist.

A short fact table helps buyers and AI tools alike:

Fact	What to state
Outlet and operator	Outlet name, legal entity, parent group
Authorised marques	Each marque, and whether sales, service or both
Facilities	3S or 4S, EV charging, body and paint, insurance claims
Location	Street address, area, state, nearest landmark or highway exit
Hours	Sales hours and service hours, including weekends
Official channels	Outlet page URL, official WhatsApp, marque locator link

Set up Business Profiles the way Google allows for dealers

Google's car dealership guidance says every dealership can have a profile for the dealership and further profiles for departments such as sales, service and parts, named with the dealership name plus the department. New-car dealerships can also have a profile for each brand of new cars sold; used-car dealerships cannot. Use the categories Google suggests (Car dealer, Car service, Auto parts business). Clean, separate profiles give AI tools a clearer signal of which outlet handles which marque than one profile listing six brands.

Add LocalBusiness markup per outlet

Google's local business structured data documentation says to mark up each location as a LocalBusiness, using the most specific subtype available, and to describe departments with nested department items where their details differ. For an outlet, that means an AutoDealer with name, address, phone, opening hours and a department for the service centre if its hours differ. Markup must match the visible page. It does not guarantee a mention, but it states the same facts in a form machines read reliably. Our technical AI visibility work usually starts with this and the crawler checks.

Align names across the group, marques and media

Choose one written form for the group name, each franchise company and each outlet, and use it on the corporate site, marque dealer locators, Business Profiles, Bursa announcements if you are listed, LinkedIn, Carlist.my and Mudah dealer pages, and press releases. Where one company operates several marques, say so in a sentence on each outlet page. This is the single most common reason AI tools describe a group as three unrelated dealers, or merge two separate groups into one.

Write the answers fleet and family buyers want

Add short FAQs to outlet and marque pages in the buyers' own words: "Is [outlet] an authorised [marque] service centre?", "Do you handle corporate fleet orders?", "Is the service centre open on Sunday?", "Can I trade in a non-[marque] car?". Keep answers factual and checkable. Leave out promotional claims you could not defend; assistants rarely repeat claims they cannot check against another source. Publish Bahasa Malaysia and Chinese versions where your buyers use them, keeping names identical.

Earn coverage that pairs the group with each marque

Use the cited sources from Step 4 as your list. In Malaysia that often means motoring media (paultan.org, CarBase.my, WapCar), marketplace dealer profiles, business press for listed groups and Google reviews. Launch events, new outlet openings and EV centre announcements are natural stories: make sure each release names the group, the operating company, the marque and the outlet location in the first paragraph. Our authority work focuses on this kind of coverage.

What to measure

Track mention rate, accuracy, citation share and competitor presence, by marque and region, every month from the same prompt set. Read them as trends, since individual answers change between runs and users.

• Mention rate: share of prompt runs where the group or one of its outlets is named, split by marque, region, language and buyer type.
• Accuracy: share of mentions with the right marque, authorised status, address and hours. Wrong service hours or a closed outlet cost real visits; fix those first.
• Citation share: how often your own pages are among the cited sources, against marque locators, marketplaces and media.
• Competitor presence: which groups appear where you do not, and which sources back them.

When you report to the board, state the method: prompts, runs per prompt, engines, dates. Put these numbers next to paid search share of voice, not instead of it. Expect uneven movement: a new outlet page may start being cited for one marque within weeks and not for another.

Common mistakes

• Treating ad share of voice as visibility. Strong paid search says nothing about whether assistants name you.
• Blocking AI crawlers by accident on marque sites or microsites through CDN or WAF rules.
• Hiding outlet facts in a locator widget or PDF service menu that crawlers cannot read as text.
• One Business Profile listing every marque instead of the brand and department profiles Google allows for new-car dealers.
• Leaving old outlet pages live after a relocation or a lost franchise, so AI tools keep sending buyers to the wrong address.
• Judging the problem from one screenshot forwarded by the managing director instead of a repeatable prompt panel.

How Shubi helps

Shubi is an AI visibility company based in Malaysia. For dealer groups we run prompt panels per marque, region and language, check crawler access across every domain, and trace which sources the assistants trust, then help fix outlet facts, naming and coverage. Our AI Visibility Intelligence service is the usual starting point, and we do not promise rankings or guaranteed mentions. Message Shubi on WhatsApp