From F&B to fintech, clinics to law firms, startups to enterprise. If your customers search on Google, we make sure they find you first, not your competitors.
One specialist team, focused only on the organic rankings that put you in front of ready-to-buy Singapore customers.
A clear, sequenced path from audit to rankings. You always know what we’re doing and why it matters for your leads.
Quick answer: The useful thing about ecommerce seo tips is not the tips, it is the order. Under 200 products, category differentiation and trust signals rank first. Between 200 and 2,000, template patterns dominate. Past 2,000, index and facet control outrank everything else, and past 20,000, crawl shaping leads.
Every list of ecommerce seo tips has the same defect. It is a list. Twenty-five items arrive in no particular order, each one defensible, and the reader picks the three that sound most interesting rather than the three that would actually move their store. On a catalogue that choice is expensive, because the same fix can be the single highest-return hour of your quarter or a complete waste of an afternoon depending on nothing more than how many products you sell. Controlling faceted navigation on a 90 product store is theatre. Ignoring it on a 9,000 product store is negligence. This post ranks the fixes instead of listing them, band by band, and states the reasoning behind each ordering so you can re-rank when your catalogue grows. If you want the commercial framing that sits above this, our e-commerce SEO page covers it. What follows is the priority queue.
Three factors decide where a fix belongs, and catalogue size moves all three at once.
Reach per hour. A template edit reaches every page built from that template. A page edit reaches one page. On a 120 product store the two are almost the same thing, so it barely matters which you do first. On a 6,000 product store a title pattern change touches 6,000 pages in an afternoon and hand-editing is arithmetic that never closes. Reach per hour is the factor that rises fastest with size, which is why template work climbs the ranking as the catalogue grows.
Risk of silent loss. Some problems cost you money without announcing themselves. A category that drops out of the index after a theme update does not throw an error, it simply stops earning, and on a store with five revenue categories somebody notices in a fortnight. On a store with 400 categories nobody notices at all. Risk of silent loss is what pushes monitoring and release discipline up the list at scale, and it is the factor almost every published tip list ignores entirely.
Dependency. A handful of fixes produce the input for others. Knowing which URLs earn revenue determines which categories you rebuild. Deciding your index policy determines which pages are worth writing. Doing dependent work out of order does not just waste hours, it means redoing the work when the upstream decision lands.
The contrarian point worth stating early: most tip lists in this market are ordered by how easy the item is to explain, not by what it returns. That is why content volume sits near the top of almost every one of them and index control sits near the bottom, when on any catalogue above about 2,000 products the true ranking is the reverse. We have taken over stores where a year of publishing had been faithfully delivered while two revenue categories sat deindexed the entire time.
Small Singapore catalogues are the common case, not the exception, and most advice written for them is borrowed from stores forty times the size.
1. Differentiate the category pages above the grid. Two or three short paragraphs stating what the category covers, who it suits, and how to choose between the options. Above the listing, not buried under it. With a small catalogue your categories are your ranking pages and there are few enough to write properly.
2. Put genuine specifics on your top 30 products. What it fits, what it does not fit, what the local warranty actually covers, why you would pick it over the next one up. At this size you can write the meaningful half of the catalogue by hand, and nobody with 8,000 products can.
3. Make the displayed price and the declared price identical, in SGD. Price, availability, currency and GST treatment need to match between what a shopper sees and what your markup states. This is the cheapest win on any store and the most commonly broken after a promotion app is installed.
4. Merge categories holding fewer than about eight products. A category with six items on a 150 product store is a filter pretending to be a page. Merging it upward concentrates demand rather than splitting it.
5. Fix the title pattern once. Product name, the attribute people actually search, then brand. One setting, whole catalogue, ten minutes.
6. Publish four buying guides that link into categories. Sizing, compatibility, care, what to buy first. Each one links into at least two categories and each of those categories links back.
7. Set the out-of-stock and discontinued rules. Temporarily unavailable keeps its page. Discontinued with a successor redirects to the successor. Nothing gets deleted by whoever happens to be on duty.
8. Add the local trust layer. Delivery windows, local returns address, GST registration, real reviews, and a company registration footprint. On small catalogues this affects conversion and click-through far more than any ranking factor.
9. Stop internal search result pages being indexed. One setting on most platforms, and it prevents a slow accumulation of near-empty URLs.
10. Crawl 200 URLs after every theme or app update. Twenty minutes, and it is the only thing standing between a quiet plugin change and a lost quarter.
Notice what is not on this list: crawl budget, log files and facet architecture. At this size a search engine can see your entire store in minutes and selling you a crawl efficiency project is selling you a solution to a problem you do not have. Stores of this size usually get more from the fundamentals covered in our small business SEO work than from anything catalogue-specific.
This is the band with the widest quality variance in Singapore, and the band where the ranking flips from page work to pattern work.
1. Set the template title, heading and meta pattern. You now have enough pages that one pattern decision outweighs a month of individual edits. Choose the pattern by looking at which qualifiers real shoppers type, not at your internal merchandising names.
2. Write the facet and parameter rule. Decide which filter combinations earn a real page, which are canonicalised to a parent and which are never linked at all. At this size the rule is a short document, not a project, and writing it now prevents the sprawl that makes the next band expensive.
3. Rebuild the category layer in revenue order. Not alphabetically, not by whichever one somebody complained about. Revenue order, with a fixed routine per category so the work is repeatable.
4. Validate structured data field by field against the rendered page. Not the source, the rendered output. Apps inject markup after load and the mismatch only appears there.
5. Build the internal link modules. Related products, parent to child, guide to category, category to guide. At this size a module beats a human deciding link by link.
6. Decide the variant strategy. Colours, sizes and pack quantities either get their own indexable pages because people search for them that way, or they are consolidated behind one canonical. Both are defensible. Having no decision is not.
7. Make pagination discoverable without JavaScript. If page two of a 40 page listing only exists after a script runs, most of your catalogue is functionally invisible.
8. Fix the category template speed, not the homepage. Your category template is the page that earns. Image dimensions, lazy loading below the fold and a properly sized hero are worth more there than anywhere else on the site.
9. Make breadcrumbs real and marked up. They carry hierarchy, they appear in results, and on a catalogue of this size they are a meaningful share of internal linking.
10. Clean up the redirects from discontinued lines. A year of deletions leaves a trail, and chains of three and four hops are normal by now.
In our experience this band is where the most money is left on the floor in this market, because the store has outgrown hand-editing but nobody has yet moved it onto patterns. The symptom is recognisable: a marketing calendar full of individual page tasks and a template that has not been touched since launch.
Above roughly 2,000 products the ranking changes character. The question stops being what should this page say and becomes which pages should exist at all.
1. Deploy the index policy, in the correct sequence. Remove from the index first, wait for the pattern to drain, then block crawling, and strip the internal links pointing at the pattern at the same time. Getting the sequence backwards is the most common self-inflicted injury on catalogues of this size.
2. Segment the sitemaps by template type. Categories, subcategories, products, guides, each in its own file. You are not doing this for discovery, you are doing it so coverage reporting tells you which layer is failing.
3. Check the rendering dependency. Crawl with scripts off and on and compare the link counts. If the rendered crawl finds far more, your discoverability depends on execution, and that is a first-rank finding.
4. Put template changes behind a release process. Staging, a named approver and a change log. One template edit now reaches thousands of pages, which cuts in both directions.
5. Consolidate near-duplicates. Cross-listed products, near-identical variants and supplier-fed descriptions repeated across ranges. You are not aiming for uniqueness prizes, you are aiming to stop your own pages competing.
6. Generate internal links systemically. Rules, not decisions. Which modules appear on which templates, how many links each emits, and what they prioritise.
7. Promote qualified filters into real categories, selectively. Only where demand for the qualified phrase is genuinely there and the resulting page has enough stock to be worth landing on.
8. Write 10 to 30 products a month in strict revenue order. Everything below that tier lives on the template and that is the correct answer, not a compromise.
9. Automate the structured data check. Manual field checking stops scaling here. Validate on a schedule and alert on error counts.
10. Run a regression crawl after every release. Same as the small store version, larger sample, same reasoning.
This is the band where a store first needs genuine technical SEO capacity rather than marketing capacity, and recognising that early is considerably cheaper than discovering it in month six.
At this size the store is a system and the tips become systems work. Very few Singapore retailers operate here, but regional distributors, industrial suppliers and B2B parts catalogues do.
1. Read the logs and shape the crawl. Where crawl attention actually lands, which templates absorb it, and how much goes to URL space you do not want. Nothing else at this size gives you as much direction.
2. Report index coverage by template, never in aggregate. An aggregate number across 400,000 URLs is a statistic with no decision attached to it.
3. Partition the sitemaps finely enough to isolate a problem. If a coverage number moves, you want to know which 5,000 URLs moved, not which 400,000.
4. Treat server response time as a search factor. At this scale time to first byte and rendering cost directly limit how much of your catalogue gets seen.
5. Cluster duplicates programmatically. Manual review is not available to you. Similarity clustering across titles, descriptions and attributes, then a merge or canonical rule per cluster.
6. Design the internal link architecture deliberately. Hub pages, faceted entry points and cross-links between related ranges, specified once and generated.
7. Guard data integrity at the feed level. Price, availability and identifiers flow from the same source into your pages, your markup and your shopping feed. Fix it upstream or fix it forever.
8. Formalise governance. A pattern table, an owner per row and a quarterly re-verification, because rules at this scale stop being enforced quietly.
9. Accept selective depth. Only the top commercial slice gets human attention. Trying to give 60,000 products individual care is how large catalogue programmes stall.
10. Retire ranges properly. Discontinued lines need a redirect decision at range level, on a schedule, or they rot into tens of thousands of dead URLs.
The same twenty or so fixes appear in every band. Only the order changes, and this is the table the rest of this post exists to produce.
| Rank | Under 200 products | 200 to 2,000 | 2,000 to 20,000 | Over 20,000 |
|---|---|---|---|---|
| 1 | Category copy above the grid | Template title and meta pattern | Index policy deployed in order | Log analysis and crawl shaping |
| 2 | Specifics on the top 30 products | Facet and parameter rule | Sitemaps segmented by template | Coverage reported by template |
| 3 | SGD price and availability accuracy | Category rebuild in revenue order | Rendering dependency check | Fine sitemap partitioning |
| 4 | Merge thin categories | Structured data against rendered page | Template release process | Server response and render budget |
| 5 | Title pattern set once | Internal link modules | Near-duplicate consolidation | Programmatic duplicate clustering |
| 6 | Four buying guides | Variant canonical decision | Systemic internal linking | Designed link architecture |
| 7 | Out-of-stock and discontinued rules | Pagination without scripts | Selective filter promotion | Feed-level data integrity |
| 8 | Local trust layer | Category template speed | Product tier writing | Formal pattern governance |
| 9 | Block internal search results | Breadcrumbs marked up | Automated markup validation | Selective depth by commercial value |
| 10 | Post-release crawl | Redirect chain cleanup | Regression crawl per release | Scheduled range retirement |
Read the table across rather than down. Category copy is rank one on a small store and does not appear in the top ten above 20,000 products, because at that size you are not writing categories, you are governing the system that generates them. Crawl shaping is rank one at the top and absent at the bottom for the mirror-image reason. A fix disappearing from a column does not mean it stopped mattering. It means something outranks it. The same reordering logic drives how we sequence work on e-commerce SEO in Singapore engagements, where the first question is always the product count rather than the platform.
A short list survives every band, and if you do nothing else from this page, do these.
Know which URLs earn revenue. One sorted list of landing pages by organic revenue, refreshed monthly. Almost no store owner we meet has looked at this list, and it is almost never what they expect, because the range people assume carries the business is frequently the price-shopped commodity one.
Keep the displayed price and the declared price identical. Every size, every platform, every time. It breaks after promotions and it breaks silently.
Never delete a product URL without a decision. Successor, parent category or genuine not-found. Deletion without a decision throws away accumulated history for nothing.
Crawl after every release. Twenty minutes for a small store, an hour for a large one. Theme and app updates break category indexability more often than anything else, and the gap between the break and the discovery is normally a full reporting cycle.
Write the change down. One line per change: what, which templates, who, when. It costs nothing and it is the only document that will explain next quarter’s numbers when nothing else does. We ask for this log on every handover, and the accounts that have one are visibly cheaper to fix.
Common advice on this subject is not false so much as unranked, and unranked advice is how stores spend a year on the wrong thing.
“Publish more.” On a store with unresolved facet problems this actively hurts, because you are adding URLs to a site that is already competing with itself. Four genuinely useful guides that link properly into categories beat twelve generic posts on essentially every Singapore store account we have run.
“Fix your crawl budget.” Below roughly 2,000 clean URLs this is not a real constraint, and the money belongs in category differentiation instead. Above 20,000 it is the top of the list. The advice is identical and the correct response is opposite.
“Add 800 words below the product grid.” We would argue against this at every catalogue size. Almost nobody reads it, it pushes the products down on mobile, and it is transparently there for a machine. Put two or three genuinely useful paragraphs above the grid instead and accept the lower word count.
“Target each product with its own keyword research.” Reasonable at 80 products, impossible at 8,000, and the attempt consumes a budget that belonged in the template. Research the category layer and the qualified variants. Let the template carry the rest.
Four local realities change which of these tips pay, and they are the ones most imported advice misses.
Marketplace share changes where you should compete. If a meaningful part of your volume goes through Shopee or Lazada, your product pages carry the same supplier text as a far stronger domain and will rarely outrank it. That pushes the guide layer and the category layer up your ranking and product page micro-optimisation down it. Treat it as a constraint on where to fight rather than a problem to be solved.
Fulfilment detail is commercial content. Same-day windows, locker collection, islandwide delivery thresholds and the actual returns process are what local shoppers check before buying. Putting them in the template rather than a policy page attached to nothing is a small change with a real conversion effect.
A share of orders never touches your checkout. WhatsApp enquiries, phone calls and PayNow transfers after a conversation. If that describes you, categories that generate enquiries instead of transactions will look unprofitable in the revenue list and get defunded, which is the wrong call made on correct-looking data.
Cross-border storefronts need a decision, not a default. Singapore and Malaysia variants of the same catalogue, or a regional store with a currency switcher, generate duplicate patterns quickly. Decide on separate storefronts with language and region annotation, or one storefront with clean currency handling. What does not work is drifting between the two. Where a store sells into several verticals at once, our industry SEO work shows how the same catalogue rules get re-weighted sector by sector.
Field notes: In our ecommerce case study, the store sat just above the smallest band in this list, with 200+ products, and its priorities followed the ranking closely. Category copy did not exist at all: category pages with content stood at 0. Before that could pay off, though, three years of faceted navigation URLs had to be brought under control, because only 34% of product pages were indexed. Robots.txt rules, a clean sitemap and resolved redirect chains moved indexation to 79% by the end of Month 2. The 15 top-revenue categories were then rewritten as content-first landing pages, averaging 720 words each, and page 1 keywords climbed from 12 to 38 in Months 3 to 5. The 20 blog guides came alongside and after that work, not ahead of it, and organic monthly revenue moved from S$8,400 to S$28,600 over nine months. Whatever your band, do the top-ranked fix for your size before adding articles.
Tips are cheap and ordering is not. The ranking in this post comes from three factors you can apply yourself when your catalogue changes: how many pages an hour of work reaches, how likely the problem is to cost you money without telling you, and whether the fix depends on a decision you have not made yet. Run those three against any advice you are given and the order sorts itself.
If you take one thing from the table, take the fact that rank one changes completely between bands. The store owner with 150 products who spends the quarter on crawl efficiency and the store owner with 9,000 products who spends it writing product descriptions are making the same mistake in opposite directions, and both of them are following advice that is perfectly sound somewhere else.
Start with the band you are actually in, do rank one properly before rank two, and re-read the table when your product count crosses a boundary. If you want to see how the ordering plays out across a full engagement, our e-commerce SEO results write-up follows the sequence on a mid-sized catalogue, and our pricing page shows where each band of work usually sits in hours.
We rank these fixes by catalogue size for a reason: in our experience, advice written for a 20,000-SKU marketplace is often actively wrong for a 150-product boutique store, and vice versa. Our team starts every new ecommerce client with the same five checks regardless of size, because those five consistently produce the fastest visible movement before anything more catalogue-specific gets attention.
Pull a list of your landing pages sorted by organic revenue, then check that the top five are returning a 200 status code, are indexable and appear in your sitemap. That is under an hour and it is the single check most likely to find something expensive. If everything is healthy, spend the rest of the afternoon writing two or three real paragraphs above the grid on your highest revenue category, which is rank one for most Singapore catalogues.
Count sellable items in your admin, not URLs. A store with 400 products and 9,000 crawlable URLs is a 400 product store with a facet problem, not a large catalogue, and treating it as the latter leads to the wrong fixes entirely. If your product count sits near a boundary, use the higher band for template and index decisions and the lower band for content decisions, since template work is cheap to do early and expensive to retrofit.
The ranking is the same, the effort per item differs. Hosted platforms constrain robots and URL control, so some index fixes need workarounds or an app rather than a setting. Self-hosted stores give you full control and hand you plugin conflicts and performance work instead. Custom builds can do anything and often launch missing the basics because nobody specified them. Expect the same order of priorities and a different number of hours against each one.
Not as a priority. If your filters generate a handful of parameterised URLs and your total crawlable count sits under about three times your product count, the search engine is seeing your whole store comfortably and there is nothing to recover. Set a simple rule so it does not sprawl as you grow, then spend the hours on category differentiation. If your ratio is above ten to one, you are no longer a small store in the sense that matters and the band above applies.
Rank one to four for your band, properly, beats all ten superficially. A realistic quarter for an in-house marketer with a few protected hours a week is the top four plus the five universal fixes, with the post-release crawl becoming a habit rather than a task. Attempting the whole list produces a lot of started work and very little finished work, which is the most common pattern we see on stalled in-house programmes.
They matter for a revenue-ranked tier and almost not at all below it. Original writing on your top 10 to 30 products a month earns its keep through differentiation and conversion. Rewriting the long tail of a 5,000 product catalogue by hand is arithmetic that never closes and the budget belongs in the template instead, where attribute display, titles, headings and markup reach every product at once.
Deliberately absent from the top ten in every band, because on most Singapore catalogues it is not the binding constraint. Authority helps a store compete on contested category terms, and it is worth doing once the category layer is genuinely good, the index policy is deployed and the guide layer exists. Building links to a store that is competing with its own filter URLs is paying to amplify a problem.
When the product count crosses a band boundary, when you replatform, and when the constraint changes from judgement to capacity. In practice most stores re-rank about once a year. The fixes that never leave the top are the five universal ones, and the ones that move most are crawl shaping, which climbs steadily with size, and individual page writing, which falls steadily.
Individual page micro-optimisation above roughly 2,000 products, publishing volume while facet problems are unresolved, keyword research on individual products at scale, and adding copy below the product grid. Each of those is a defensible activity somewhere and a misallocation at the size most Singapore stores actually operate at. Stopping them frees the hours that the top of your band’s list requires.
Three triggers. A replatform, where the risk is concentrated into a short window. A catalogue above roughly 5,000 products, where the category queue exceeds what one person clears in a year. And any point where your fix list turns out to need code rather than settings, since your timeline becomes a developer queue rather than your own calendar. The B2B version of that pattern appears in our B2B e-commerce results write-up.
If you want to know which band you are actually in and what rank one looks like for your store specifically, we will run a free initial review and send back this ranking rewritten against your own catalogue, with the first three fixes named and the reasoning for the order. Keep the findings either way. Get in touch with your store URL, product count and platform, and tell us what you have already done, because the order changes considerably once the early fixes are in.
Natalie leads SEO strategy at Singapore SEO Agency, helping local and regional businesses build organic search programmes that drive qualified leads. She specialises in technical SEO and content-led authority building for Singapore SMEs.
Get a free SEO audit for your Singapore website — we'll show you exactly where you stand, what's holding you back, and what it would take to rank on page 1.
Get Your Free SEO Audit →
Whats bounce rate, and how is it different from exits, engagement time or dwell time? Learn what each one really answers and what to ask your SEO agency.
Google Analytics website traffic is a consented sample, not a head count. See where visits come from, why tools disagree and which number to trust each month.
Google analytics consulting services for Singapore SMEs: what consultants do, indicative SGD costs, deliverables to demand and red flags. Buy the right scope.
Bounce rate in GA4 is not the old Universal Analytics number. Learn what it measures now, why old benchmarks fail and how to judge each page by its job.
Yoast SEO helps most when its site-wide settings are right. See which settings to choose, what the traffic lights really mean and the mistakes to avoid.
Which free keyword research tool should a Singapore clinic, firm, contractor or tutor use? Combine five free tools to find your first 30-50 keywords. See how.
Fast, no obligation. We reply within 24 hrs.
Singapore’s specialist SEO agency for SMEs. We rank your business on Google — and only Google. No distractions, just results.
© 2026 Singapore SEO Agency. All rights reserved.