First-party data
How Malaysian Ads Ask You to Get in Touch (2026)
Which response mechanism Malaysian Facebook and Instagram ads actually ask for, measured phrase by phrase across 222,921 running ads and cut by vertical, with the full dataset as a free CSV.
Updated September 2026 · AdPlay.ai Team
Malaysian ads ask you to open WhatsApp. Across 222,921 Malaysian Facebook and Instagram ads running as at their last check on 25 August 2026, 13.6% carry the word WhatsApp in their body copy, with a further 3.5% using the Malay spelling. Every other mechanism is an order of magnitude smaller: 1.9% invite a visit, 1.2% ask for a call in Malay and 0.8% ask for a direct message. It splits hard by vertical, from 26% of services ads down to 4.5% in app / software, which is the difference between a category that books work and a category that takes a checkout.
Every guide about Malaysian advertising will tell you WhatsApp matters here. None of them counts it. This page reads the actual body copy of Malaysian ads and measures which response the advertiser asks the reader for, one exact phrase at a time, in English and in Malay. The method is deliberately narrow and the page is explicit about what it therefore cannot see.
WhatsApp is not one option among several
13.6% of running Malaysian ads carry the word WhatsApp somewhere in their body copy, 30,239 ads. Add 3.5% for the Malay spelling and 3.1% for the phrase asking you to WhatsApp them directly, and the mechanism is in a different bracket from everything else on the list. Everything else is small. 1.9% invite you to visit, 1.3% say walk in, 1.1% name a branch. 1.2% ask you to make contact in Malay and 0.6% say call us in English. Direct messages, the mechanism most international ad advice assumes, sit at 0.8% for the English phrase. Those figures cannot be added together, and the reason is worth stating rather than hiding in a note. One ad can carry a WhatsApp number and a walk-in address in the same caption, so the probes overlap by design. There is also no way to compute the union honestly here: the search cannot express an OR across phrases and ad body copy is not a filterable field, so a combined figure would have to be estimated. This page publishes what it measured rather than what it could estimate.
| Phrase | Mechanism | Language | Ads | Share of running ads |
|---|---|---|---|---|
| English | 30,239 | 13.6% | ||
| wasap | Malay | 7,776 | 3.5% | |
| whatsapp kami | English | 6,828 | 3.1% | |
| visit our | Walk in | English | 4,316 | 1.9% |
| klik link | Click or register | Malay | 4,064 | 1.8% |
| walk in | Walk in | English | 2,870 | 1.3% |
| hubungi kami | Phone call | Malay | 2,766 | 1.2% |
| cawangan | Walk in | Malay | 2,512 | 1.1% |
| datang ke | Walk in | Malay | 2,394 | 1.1% |
| dm us | Direct message | English | 1,884 | 0.8% |
| register now | Click or register | English | 1,777 | 0.8% |
| message us | Direct message | English | 1,732 | 0.8% |
| pm kami | Direct message | Malay | 1,502 | 0.7% |
| daftar sekarang | Click or register | Malay | 1,476 | 0.7% |
| call us | Phone call | English | 1,344 | 0.6% |
| call now | Phone call | English | 1,093 | 0.5% |
| click the link | Click or register | English | 985 | 0.4% |
| telefon | Phone call | Malay | 753 | 0.3% |
| inbox kami | Direct message | Malay | 57 | 0% |
The split is between booking work and taking a checkout
The vertical cut is where this becomes useful. Services ads ask for WhatsApp 26% of the time. Home / Garden follows at 20% and Transport at 19.6%. At the other end, App / Software sits at 4.5%. The line is not language or budget, it is whether the sale has a conversation in it. A category that books work, quotes a job, or takes an appointment puts a chat thread in the ad because the next step genuinely is a conversation. A category that ships a product from a catalogue does not, because the next step is a cart. So the practical reading depends on which side you are on. If you sell a service and your ads do not carry a chat route, you are asking a Malaysian buyer to take a step almost nobody in your category asks them to take. If you sell a product, the low numbers in your vertical are not a gap in the market, they are the market telling you the checkout is the right destination.
| Vertical | dm us | call us | walk in | click the link | |
|---|---|---|---|---|---|
| Services | 26% | 2.1% | 1.4% | 1.8% | 0.6% |
| Home / Garden | 20% | 1.2% | 0.9% | 1.2% | 0.5% |
| Transport | 19.6% | 0.6% | 0.8% | 1.9% | 0.5% |
| Healthcare | 16.1% | 0.9% | 1% | 1.6% | 0.4% |
| Real Estate | 15% | 0.8% | 1.2% | 2.4% | 1.5% |
| Electronics / Tech | 13.9% | 0.5% | 0.5% | 1.4% | 0.3% |
| Travel | 13.8% | 0.7% | 0.9% | 1.1% | 0.3% |
| Education | 13.3% | 0.7% | 0.6% | 0.8% | 0.8% |
| Beauty / Personal Care | 11.5% | 1% | 0.4% | 0.8% | 0.3% |
| Retail / Marketplace | 11.5% | 0.5% | 0.4% | 1.5% | 0.3% |
| Food / Beverage | 11.2% | 0.5% | 0.4% | 1.3% | 0.3% |
| Kids | 10.2% | 0.5% | 0.3% | 1.6% | 0.4% |
| Consumer Goods | 7.6% | 0.4% | 0.2% | 0.9% | 0.2% |
| Fitness | 6.9% | 2% | 0.3% | 1.7% | 0.9% |
| Finance | 6% | 0.1% | 0.1% | 0.7% | 0.5% |
| Clothing / Apparel | 4.9% | 0.4% | 0% | 0.7% | 0.1% |
| App / Software | 4.5% | 0.2% | 0.1% | 0.2% | 0.6% |
What this method cannot see
The single largest blind spot is the button. Meta's call-to-action button has its own field, an ad can carry a WhatsApp button with no mention of WhatsApp in its text, and that field is not one this dataset can filter on. So an ad whose entire contact instruction is a button is invisible here, and every figure on this page is a floor rather than a rate. Three narrower limits. Ads with no body copy at all stay in the denominator, because excluding them would require filtering on the same field the probes search. Phrases are matched inside body copy only: searching every field would nearly double the WhatsApp number by catching landing-page URLs, which is a different question with a link in it rather than a better answer to this one. And a phrase is not a prominence, so an ad mentioning WhatsApp in its ninth line counts the same as one leading with it. One check worth reporting because it could have gone the other way. Every headline probe was run twice, once matching the whole phrase and once under the loose strategy that drops query words until something matches. The two agree exactly on all five, so these are phrase counts rather than word counts. The comparison is in the CSV.
Where this page stops and the offer study begins
Some phrases sit on a boundary and it is better to draw it in public than to argue about it later. Shop now appears in 6.3% of running Malaysian ads and beli sekarang in 1.8%, and neither is counted as a contact mechanism here. The distinction is what the phrase demands. Shop now is an offer instruction: it tells you the thing is for sale and where. WhatsApp us is a contact instruction: it tells you which app to open and that a human is on the other end. They answer different questions for an advertiser deciding what to put in a caption, so they belong on different pages. Both boundary phrases are published in the CSV with their counts and a note saying which study owns them, so anyone who disagrees with where we drew the line can redraw it from the same file rather than from scratch.
Reuse this data
Every probe is one row in a CSV published under CC BY 4.0, free to reuse including commercially with attribution to AdPlay.ai. Each row carries the exact query string that produced it, the surface it searched, the denominator, and the note that applies to it, so the whole thing is reproducible against the same archive rather than something you have to take on trust.
By the numbers
Frequently asked questions
What percentage of Malaysian Facebook ads use WhatsApp?
13.6% of Malaysian ads running as at their last check carry the word WhatsApp in their body copy, with another 3.5% using the Malay spelling wasap. Both are floors, because an ad can carry a WhatsApp button with no mention of it in the text and the button field cannot be counted here.
Which Malaysian industries use WhatsApp in ads the most?
Services lead at 26%, followed by Home / Garden at 20% and Transport at 19.6%. App / Software sit lowest at 4.5%. The pattern follows whether the sale involves a conversation rather than anything about budget or language.
Should my ad ask people to WhatsApp or to click through?
This data cannot tell you which performs better, because the archive holds no spend, clicks or results. What it tells you is what your category does: if you sell a service where the next step is a quote or a booking, a chat route is the norm, and if you sell a product from a catalogue it is not.
Why do the percentages not add up?
Because one ad can ask for several things. A clinic ad can carry a WhatsApp number, a phone number and a branch address in the same caption, so the probes deliberately overlap and are never summed. There is also no honest way to compute the combined figure, since the search cannot express an OR across phrases and ad copy is not a filterable field.
Do Malaysian ads ask people to call?
Rarely, in writing. 1.2% use the Malay phrase for get in touch, 0.6% say call us and 0.5% say call now. Telephone contact has largely moved into the chat thread, which is what the WhatsApp figures describe.
Are ads in Malay different from ads in English here?
The Malay probes are smaller across the board, but that reflects how many ads are written in each language rather than a difference in behaviour. Both spellings of WhatsApp appear, at 13.6% and 3.5%, and each language pair is published separately in the CSV so you can read them side by side rather than through a merged figure.
How were these numbers produced?
Each is an exact-phrase search inside ad body copy only, against Malaysian ads running as at their last check, counted through the search index rather than estimated. The CSV carries the query string for every row, so any figure can be reproduced against the same archive.
Can I reuse this dataset?
Yes, under CC BY 4.0 with attribution to AdPlay.ai. Cite the snapshot date with any figure you take, since the archive changes as ads are collected and re-checked.
The data behind this page
Every figure on this page is a row in that file. Measured as at 2026-08-25. Published under Creative Commons Attribution 4.0, so reuse it with a credit to AdPlay.ai.
Sources
Keep exploring
Turn ad research into winning ads
See what 16,000 Malaysian brands advertise, then generate on-brand creative, all in one tool.
7-day free trial · No credit card required