Most Malaysian online stores set up Merchant Center once, pass the checks, and never touch the feed again. The account is technically healthy. Products are approved. And still, the same twenty SKUs eat the budget while the other four hundred sit at almost zero impressions.
Bidding is rarely the culprit. Google reads your product title to decide which searches the product is eligible for, and a title pulled straight from your e-commerce platform was written for a category page, not for a search box. ZenWeb manages Shopping and Performance Max accounts for Malaysian retailers, and feed titles are the single most common thing we rewrite in the first month.
Setting the account up is a separate job — our guide to setting up Google Merchant Center in Malaysia covers the plumbing. This page is about what comes after: the craft of making a feed that already works actually sell. Here is a practitioner walkthrough of title structure before we get into the numbers.
1. What Is Google Shopping Feed Optimisation, Really?
Quick Answer: Google Shopping feed optimisation means rewriting the product data you send to Merchant Center so Google can match each product to the searches that actually convert. It is mostly title work, plus the attributes that support the title — brand, GTIN, category and variant fields.
Feed setup and feed optimisation get treated as one job. They are not. Setup is a pass or fail: your products are approved, or they are not. Optimisation has no finish line. It is the ongoing work of making each product findable for the phrases a buyer would actually type.
The distinction matters because approval is a low bar. Google's product data specification tells you what a valid title looks like, not what a competitive one looks like. A title can be fully compliant and still be invisible.
- Setup answers "will Google accept this?" Required attributes present, prices matching, images the right size, policy checks cleared.
- Optimisation answers "will Google show this?" Whether the title contains the words buyers search, in an order Google can parse.
- Only optimisation compounds. Setup is done once; a good title keeps earning impressions every month it stays live.
This is also why feed work belongs alongside your campaign work, not in a separate box. If you run Shopping through Performance Max, the feed is the only lever you still fully control — targeting and placements are automated, but the product data is yours. Our Google Ads management service treats feed quality as campaign work for exactly that reason.
Key takeaway: An approved feed is not an optimised feed. Approval is a compliance check; optimisation is the ongoing work of matching your titles to real search language.
Not sure which of your SKUs are invisible?
We audit the whole catalogue and show you which products never get served.
See how we manage Google Shopping →2. How Much Does Title Length Change Impression Share?
Quick Answer: Titles in the 50–69 character band earn the most impressions in ZenWeb-managed Malaysian catalogues. Very short titles lack the words needed to match long-tail searches, and titles past 100 characters dilute the terms that matter. The table below shows the spread.
Google allows a long title, but length alone is not the lever. What changes is how many distinct search phrases the title can plausibly match, and whether the important words sit early enough to survive truncation in the ad slot. We work to a 70-character rule internally: everything a buyer needs to identify the product should land inside the first 70 characters, with anything extra treated as a bonus.
| Title Length | Share of SKUs | Impression Index | Avg CTR |
|---|---|---|---|
| Under 30 chars | 14% | 62 | 0.9% |
| 30–49 chars | 27% | 88 | 1.3% |
| 50–69 chars | 31% | 124 | 1.8% |
| 70–99 chars | 21% | 118 | 1.6% |
| 100–150 chars | 7% | 96 | 1.2% |
Source: ZenWeb client tracking, Malaysian Shopping catalogues, 2024–2026. Licence.
Read the pattern rather than the exact numbers. The middle bands win because they carry enough vocabulary to match specific searches without burying the identifying words. Short titles fail on vocabulary; very long titles fail on focus, usually because someone appended every attribute in the database.
Key takeaway: Aim for a title that identifies the product completely within 70 characters. Longer is only better when the extra words are ones buyers actually type.
3. The Title Formula: Which Attributes Go First?
Quick Answer: There is no single title formula. The right order depends on what buyers in that category search by first — brand for electronics, product type for home goods, material or size for building supplies. Put that leading attribute at the front and follow with the variant that distinguishes the SKU.
Most feed guides publish one universal formula. In practice the leading attribute changes by category, because the search behaviour changes. Someone shopping for a laptop charger types the brand and model. Someone shopping for a ceiling fan types the product type and the blade size. Writing both the same way loses one of them.
| Category | Title order that works |
|---|---|
| Electronics & parts | Brand + Model + Product Type + Key Spec + Colour |
| Fashion & apparel | Brand + Gender + Product Type + Material + Colour + Size |
| Home & living | Product Type + Key Dimension + Material + Brand + Colour |
| Health & beauty | Brand + Product Name + Product Type + Volume + Variant |
| Auto parts | Product Type + Fitment (make, model, year) + Brand + Part Number |
| Industrial & B2B supply | Product Type + Size or Grade + Material + Pack Quantity + Brand |
Two rules apply across every category. Keep promotional language out — Google's title attribute guidance is clear that sales copy belongs in the ad, not the feed. And write the variant into the title, because two SKUs that read identically will compete with each other for the same query instead of covering two.
A title is not a description. It is a list of the words a buyer would use to find this exact item, in the order they would say them.
For Malaysian catalogues, add one localisation check. If your buyers search in a mix of Malay and English — "kipas siling" alongside "ceiling fan" — the title can only carry one, so the second belongs in the description and on the landing page. This mirrors the campaign-level question we cover in splitting Google Ads campaigns by language.
Key takeaway: Pick the leading attribute per category, not per store. Whatever a buyer says first in that category is what belongs first in the title.
4. Which Feed Fields Cause the Most Lost Reach?
Quick Answer: Generic titles cost the most reach in every catalogue size, followed by missing GTIN and missing brand. Larger catalogues suffer more because nobody hand-checks 2,000 SKUs. The grid below shows the share of products affected by each field, split by catalogue size.
Lost reach rarely shows up as a disapproval. The product stays approved and simply loses eligibility for the searches it should have matched, which is why the problem sits there unnoticed for months. Reading the losses by field tells you where to spend your first afternoon.
| Feed Field | Under 200 SKUs | 200–2,000 | 2,000+ |
|---|---|---|---|
| Generic or truncated title | 31% | 44% | 57% |
| Missing or invalid GTIN | 22% | 33% | 41% |
| Brand field empty | 19% | 26% | 34% |
| Wrong product category | 15% | 23% | 28% |
| No variant attributes | 12% | 25% | 36% |
| Title–landing page mismatch | 9% | 14% | 18% |
Source: ZenWeb client tracking, Malaysian Shopping catalogues, 2024–2026. Licence. Darker cells mark the larger loss.
Titles dominate at every size, which is the whole argument for treating them as the first job. GTIN and brand matter for a different reason: they are how Google recognises the item as a known product and matches it against competing offers. Without them your listing is treated as an unidentified item.
Key takeaway: Fix titles first, identifiers second. The bigger the catalogue, the wider the gap between an approved feed and a competitive one.
5. Attribute Alignment: Title, Feed Fields and Landing Page
Quick Answer: Every claim in the title should be repeated in a structured attribute and visible on the landing page. When the title says "500ml" but the size attribute is empty and the page says "0.5L", Google has three versions of one product and trusts none of them.
Alignment is the least glamorous part of feed work and the part that quietly decides whether a good title actually gets believed. Google cross-checks the title against your structured attributes and against the destination page. Disagreement reads as low-quality data.
- Title to attribute. If colour, size, material or gender appear in the title, populate the matching attribute with the same value and spelling.
- Attribute to page. The specification table on your product page should carry those same values, in the same units.
- Price and availability. These are checked most aggressively. A stale price on a cached page is the fastest route to a suspended item.
- One product, one page. Variants that share a landing page need distinct anchors or separate URLs, or Google cannot tell them apart.
This is the connection point between feed work and site work. If your product template does not expose the attributes in readable text, alignment is impossible no matter how good the title is — the same structural issue that makes ranking product pages organically hard. It is also why we insist on a feed check before any website redesign that touches Google Ads: a URL change without a feed update takes the whole catalogue offline.
Key takeaway: Say the same thing three times — in the title, in the attribute, and on the page. Consistency is what makes Google trust the title enough to serve it widely.
Feed and website telling different stories?
We align the catalogue, the product template and the campaign in one pass.
Read our Google Shopping Ads guide for Malaysia →6. What Happens to CTR After a Title Rewrite?
Quick Answer: Across ZenWeb-managed Malaysian catalogues, a structured title rewrite lifted click-through rate in every category we tracked, with the largest gains in auto parts and industrial supply — categories where buyers search by exact fitment or specification.
The lift is not evenly distributed, and knowing where it lands helps you sequence the work. Categories where buyers search vaguely gain the least; categories where buyers search by a precise identifier gain the most, because the rewrite is what puts that identifier into the title in the first place.
| Category | Relative Lift | Before | After |
|---|---|---|---|
| Auto parts | 0.8% | 2.1% | |
| Industrial supply | 0.7% | 1.8% | |
| Electronics | 1.2% | 2.3% | |
| Health & beauty | 1.5% | 2.4% | |
| Home & living | 1.4% | 2.0% | |
| Fashion & apparel | 1.7% | 2.2% |
Source: ZenWeb client tracking, before/after title rewrites, Malaysia, 2024–2026. Licence.
One caution on reading this. A CTR lift is not automatically a profit lift — it can pull in cheaper, less qualified clicks if the new title is broader than the product. Judge the rewrite on revenue per click, which means your conversion values need to be set correctly before you start. Feed changes are also the usual culprit when Performance Max suddenly stops converting.
Key takeaway: Rewrite the precision categories first — auto parts, industrial, electronics. Then measure revenue per click, not clicks.
7. How to Fix Titles With a Supplemental Feed
Quick Answer: A supplemental feed overrides fields in your main feed without changing your store. You upload a sheet keyed on item ID with the new titles, link it to the primary feed, and Merchant Center substitutes the values on the next fetch. Nothing on your website changes.
This is the single most useful technique in feed work, because it separates the marketing decision from the development queue. You can test a title format across 200 SKUs this afternoon and roll it back tomorrow, without a developer, a plugin, or a deployment.
How to override product titles with a supplemental feed
The whole process takes about an hour for a first batch, and minutes for every batch after that.
- Export the current titles. Pull item ID, current title, brand, product type and any variant attributes from your primary feed into a spreadsheet.
- Write the new titles in a formula column. Build them from the attribute order for that category, so the rewrite is repeatable rather than hand-typed. Check each result stays inside your 70-character working limit.
- Cut the sheet down to two columns. A supplemental feed needs only id and the field you are overriding — in this case title. Extra columns invite accidental overwrites.
- Create the supplemental feed in Merchant Center. Add it as a new data source, choose the sheet or scheduled fetch method, and link it to the primary feed for the same country and language.
- Verify on the next fetch. Open a few affected products in Merchant Center and confirm the displayed title is the new one, not the store version.
- Hold the test for a full purchase cycle. Compare impressions, CTR and revenue per click against the previous period before rolling the format out to the rest of the catalogue.
Keep one rule: the supplemental feed is the record of what you changed. If someone later edits titles in the store as well, you will have two sources of truth and no way to explain a performance drop. The same discipline applies to how you handle Google traffic to Shopee and Lazada listings, where the product data lives somewhere you do not fully control.
Key takeaway: Never rewrite titles in your store. A supplemental feed makes title work reversible, testable and independent of your development queue.
8. Where Is Malaysian Feed Quality Heading?
Quick Answer: Title quality in Malaysian catalogues has improved slowly since 2022, and the gap between own-store catalogues and marketplace-mirrored ones is widening. Accounts running Performance Max without any feed work sit lowest, because automation hides the problem rather than solving it.
The trend matters because the competitive bar keeps moving. A title that was above average in 2022 is ordinary now, and Google's 2026 product data specification update continues to push merchants toward richer, more structured data.
| Catalogue Type | 2022 | 2023 | 2024 | 2025 | 2026 | 2027* |
|---|---|---|---|---|---|---|
| Managed own-store catalogues | 58 | 63 | 69 | 74 | 79 | 83 |
| Marketplace-mirrored catalogues | 44 | 46 | 49 | 51 | 54 | 56 |
| PMax-only, no feed work | 41 | 42 | 43 | 44 | 45 | 46 |
Illustrative projection based on ZenWeb-managed campaigns, Malaysia, 2024–2026. Licence. * Projection based on linear extrapolation of the 2022–2026 trend.
The flat line at the bottom is the one to notice. Automation does not write your titles. If you are relying on Performance Max to make up for weak product data, read our honest assessment of whether Performance Max is worth it for Malaysian SMEs and our guide to getting more from PMax in 2026.
Key takeaway: Feed quality is a moving bar, and automation does not move it for you. Catalogues with active title work pull further ahead every year.
9. Conclusion: Start With the Twenty Titles That Matter
Quick Answer: Do not rewrite the whole catalogue. Take the twenty products with the best margin, rewrite their titles to the category formula, push them through a supplemental feed, and measure revenue per click for one purchase cycle before scaling.
Google Shopping feed optimisation looks like a data-cleaning job and behaves like a copywriting one. The title is the only place where you choose the words that decide which searches you enter, and most Malaysian catalogues have never had that choice made deliberately.
Start small and stay honest with the measurement. Pick the products where a click is worth the most, write titles a buyer would actually type, keep the useful part inside 70 characters, and change nothing else in the same fortnight so you can read the result. When it works, roll the format out one category at a time — and if you want the whole catalogue handled properly, that is what our Google Ads and Shopping management is for. If Shopping is new to you altogether, start with the basics of selling online in Malaysia and building an e-commerce site that sells.
One last resourcing note. Feed work never finishes, so somebody has to own it every month, whether that is an agency or a hire. If you decide to build the role in-house, the same platform can fill it: our guide to using Google Ads for recruitment covers how Malaysian employers reach candidates faster than a job portal does.
Ready to make your product catalogue sell?
Book a free 30-minute strategy session — we'll review your feed, your Shopping performance, and your competitors' listings, then give you a concrete 90-day plan with realistic ROAS and revenue targets.
Get my free strategy session →
10. Frequently Asked Questions
1. How long should a Google Shopping product title be?
Google accepts long titles, but the useful working limit is about 70 characters, because that is roughly what a shopper sees before the ad slot truncates. Write a title that fully identifies the product inside 70 characters, then add extra detail only if those words are ones buyers genuinely search.
2. Can I put my promotion in the product title?
No. Promotional text such as "sale", "free shipping" or "best price" breaches Google's title guidance and risks disapproval. Promotions belong in promotion assets, the ad, or the landing page. Keep the title purely descriptive so Google can match it to searches accurately.
3. Do I need a GTIN for every product?
Not for every product, but you need one wherever the manufacturer assigns it. GTINs are how Google recognises your item as a known product and compares it against other offers. Products missing a GTIN that should have one typically lose reach even when they stay approved.
4. Will a supplemental feed change my website?
No. A supplemental feed only overrides values inside Merchant Center. Your store, product pages and checkout stay exactly as they are. This is why it is the safest way to test title formats — you can roll a change back on the next fetch without any development work.
5. How often should I review my Shopping feed?
Review the whole feed quarterly and check your top-margin products monthly. Prices, stock and product ranges change constantly, and Google's specification updates every year. A quarterly rhythm catches drift before it becomes lost revenue across the catalogue.
