
Free Keyword Research Tool: The Free Stack That Works for Service Businesses
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.
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: Featured snippets are boxed answers pulled from a page already ranking on the first results page, shown above the standard listings. Google draws them from four formats: paragraph, list, table and video. Winning one raises visibility, but on fully answerable queries it can reduce clicks rather than increase them.
Few search features are chased as enthusiastically or understood as poorly as the boxed answer at the top of the results. The usual framing is that featured snippets are free traffic you simply need to format for, and that framing is half right at best. This post covers the mechanism properly: which formats Google draws from and what triggers each, how to structure a page so your text is the one extracted, why a win sometimes costs you clicks rather than adding them, and how the whole feature now interacts with generated answers sitting above it. It is written for Singapore businesses with small addressable markets, where the difference between a vanity win and a commercial one is not academic. If you want the underlying service context, our SEO services page explains how we prioritise this kind of work inside a retainer.
The box at the top is not a separate ranking system. Google selects it from pages that already rank, in practice almost always from the first page of results and very often from the top five. You cannot win a snippet for a query you do not already rank for. That single fact reorders most snippet strategies, because it means the work starts with ranking and only then moves to formatting.
Once a page is in contention, the selection is an extraction problem. The system is looking for a passage on the page that answers the query directly, completely and in a shape that fits the box. It lifts that passage, attributes it to your URL, and displays it. You do not submit anything, you do not mark anything up for it, and there is no structured data type that makes a page eligible. The widely repeated advice that schema markup wins featured snippets is simply wrong, and it sends teams off to install a plugin when the actual fix is a paragraph rewrite.
The second mechanical point is deduplication. When your page is shown in the snippet, your standard listing is removed from the results below it rather than appearing twice. You occupy one slot, promoted to the top of the page. This is the detail that turns the entire tactic from a pure gain into a trade, and it is why the rest of this post spends as much time on cost as on capture.
Google draws snippets in four shapes, and the shape is determined by the query, not by your preference. Writing a table for a definition query wins nothing.
Paragraph snippets are the most common by a wide margin. They answer what, why, who and definitional queries with a block of text, typically running somewhere around 40 to 60 words before it truncates. If your answer takes 200 words to arrive, it will not be selected.
List snippets come in two forms. Ordered lists answer sequence and process queries, the how-to-do-something class. Unordered lists answer set queries: types of, examples of, what to include. Google will frequently show only the first several items with a link to see the rest, which is the one snippet format that reliably drives a click because the box is deliberately incomplete.
Table snippets answer comparison, specification and pricing queries. They are pulled from genuine HTML table markup, which means a comparison laid out with styled divs or an image of a table cannot be extracted no matter how good the content is. This is one of the few cases where the underlying HTML element itself decides eligibility.
Video snippets surface a clip with a timestamped moment, usually for demonstrative how-to queries. For most Singapore service businesses this is not a practical target, though it matters for trades and product demonstrations.
| Format | Typical trigger | Structure that wins it | Click value |
|---|---|---|---|
| Paragraph | what is, why does, who is, definitions | Question as a heading, then a direct 40 to 60 word answer | Low, often resolved in the box |
| Ordered list | how to, steps to, process for | Numbered heading sequence or a clean ol element | Medium, rises with step count |
| Unordered list | types of, examples of, checklist | Short bulleted set under a matching heading | Medium to high when the list is long |
| Table | cost, comparison, specifications | Real HTML table with a header row | Medium, high when rows exceed the box |
| Video | demonstrations, visual how-to | Chaptered video with accurate transcript | High for the channel, low for the site |
The pattern worth extracting from that table is this: the more completely the box satisfies the user, the less the win is worth to you. A definition snippet is prestigious and frequently commercially empty. A twelve-row pricing table that gets truncated at four rows is a much better asset.
Assume you already rank in the top ten for the query. The work from there is structural and takes less time than people expect.
Put the question in a heading, phrased the way users phrase it. An H2 reading “How much does a corporate secretarial service cost in Singapore” is a far stronger extraction target than one reading “Our fees”. The heading tells the system that what follows is an answer to that specific question.
Answer immediately under the heading, in one self-contained paragraph. No preamble, no throat-clearing, no “before we get into that”. The answer must make sense lifted out of the page entirely, because that is exactly what happens to it. Include the subject of the question in the answer, since a passage beginning with “it depends on” is meaningless out of context.
Keep the extractable answer inside roughly 40 to 60 words, then elaborate below it. This is the shape that matches the box. Write the concise answer first and the nuance second. Many pages do the reverse and lose on length alone.
Use the correct HTML element for the format. Real ol and ul elements for lists, real table markup with a header row for tables. Page builders frequently render visual lists as styled paragraphs, which look identical to a human and are invisible as a list to a parser. We have found this exact problem repeatedly on sites built by web vendors with no search involvement, and it is one of the cheapest fixes available.
Cover the adjacent questions on the same page. The People Also Ask box surfaces related queries, and those questions are a free map of what else the system considers part of this topic. A page answering the main question plus five adjacent ones is in contention for several boxes rather than one.
And keep the answer current, because staleness is visible. Snippets on price, regulation and process queries turn over faster than the rest, and a page quoting a figure from three years ago will eventually be replaced by one that does not. Our small business SEO clients get more from refreshing six existing answers than from writing six new pages.
This is the part most guides leave out, and it is the part that decides whether the tactic is worth your time.
Because your standard listing is removed when the snippet is shown, you are trading a blue link with a title and description for a box containing your answer. For a query the box can fully satisfy, that trade loses. The user reads the definition, understands the thing, and leaves. You were shown more prominently and earned less.
The clearest way to think about it is to ask what the user still has to do after reading the box. If the answer is “nothing”, the snippet is a branding impression rather than a traffic source. If the answer is “choose a provider, see the full price list, check whether this applies to their situation, book something”, the snippet works as intended and the prominence is a real gain.
There is a second cost that is easy to miss. Chasing snippets encourages a particular writing style: clipped, definitional, optimised for extraction. Applied to a commercial page, that style converts worse, because the humans who do click through wanted context, evidence and reassurance, not a dictionary entry. Conventional wisdom says to make every page snippet-friendly. In our experience that advice quietly degrades the pages that actually earn revenue.
Our working rule is straightforward. Format aggressively for extraction on informational supporting content, where a click was always unlikely and visibility has brand value. Keep commercial service pages written for the buyer, with at most one clean answer block near the top. The pages that pay for the engagement should be built for the person, not the parser.
There is also a defensive dimension that gets ignored. If a competitor holds the snippet for a query where the answer naturally leads to a purchase, they are occupying the most prominent position on the page and framing the topic in their terms. Taking that box away from them has value beyond the clicks it delivers to you. In competitive professional categories this is often the stronger argument, and it shapes how we prioritise in law firm SEO work where a handful of queries carry most of the commercial weight.
The landscape changed when search engines began placing generated summaries above the results, and the honest position is that the two features overlap heavily without replacing each other.
On many informational queries, a generated answer now appears where a snippet used to. The functional effect on your traffic is similar: the user’s question is resolved above the links. The difference is that a generated answer draws from several sources and names them, rather than lifting one passage from one page, so the visibility is shared and the click-through is generally lower still.
The structural work is almost entirely the same, which is the useful news. A page with a clear question heading, a direct self-contained answer, real list and table markup, and specific factual statements is the ideal raw material for both features. Nothing about optimising for extraction is wasted, because both systems are solving the same problem of finding a quotable, trustworthy passage.
What changes is the target. Where the old goal was to own the box, the sensible goal now is to be the source the answer draws on, across both features, while concentrating your commercial effort on queries where a summary cannot finish the job. A generated paragraph can explain what stamp duty is. It cannot tell a buyer what your firm charges to handle their specific transaction, and it cannot book a consultation.
And measurement has to change with it. Impressions holding steady while clicks fall is the signature of an answer being read above your listing. That is not a ranking problem and chasing it as one wastes budget. We see this most clearly in research-heavy categories such as finance sector SEO, where a large share of query volume is definitional and was never going to convert in the first place.
A short prioritisation exercise saves a great deal of wasted formatting work.
First, list the queries where you already rank between positions two and eight. These are your realistic targets. Anything outside the first page is a ranking project, not a snippet project.
Second, mark each one for what the user must still do after reading a complete answer. Discard the ones where the honest answer is nothing, unless the query carries strong brand value for you.
Third, check whether a snippet currently exists and what format it is in. If Google is showing a table and your page has prose, you know the exact gap. If no snippet is shown at all, the query may not trigger the feature and no amount of formatting will change that.
Fourth, weight by commercial value rather than volume. In Singapore, a query with 40 monthly searches from people ready to appoint a provider is worth more than one with 900 searches from students and overseas researchers. This is the single most common misallocation we correct in local accounts, and it applies far beyond snippets.
The same filter applies whatever the sector, and it is how we sequence work in education SEO accounts where definitional volume dwarfs commercial volume.
Our car dealer case study shows the pattern on a commercial site. For an authorised Toyota dealership with 3 showrooms, the programme built 8 model landing pages alongside COE and financing content. The COE content was indexed in Months 3 to 4 and was appearing in People Also Ask boxes by Months 5 to 6, while 5 model terms reached the top 3 and monthly organic leads moved from 11 to 41 over six months. The answer-box visibility was useful because it sat next to model pages built for buying queries, with a test drive booking form on each one. A snippet that has no commercial page behind it is visibility without a path to an enquiry.
Field notes: In our family law case study, the firm published a library of 14 legal guide articles, including process guides on contested divorce, matrimonial property, and HDB and CPF issues, with FAQPage schema on all of them. During Months 3 to 4 those process guides began appearing in People Also Ask, and monthly consultation enquiries moved from 4 to 8. By Month 7 consultations reached 16. The guides worked because each one led to 8 rebuilt practice area pages and lawyer profiles with consultation booking forms, so a reader who saw the answer box had somewhere to go next. The lesson we take is not that answer boxes are good or bad, but that format and query intent, plus a clear next step, predict the outcome far better than the win itself.
Featured snippets are worth pursuing selectively and not worth pursuing as a programme. The box rewards you in proportion to how much work the user still has to do after reading it, which means the queries that feel most prestigious to win are usually the ones that return the least.
We’ve seen a business win a snippet, celebrate it, and then notice traffic barely moved, because the query it won answered the question completely inside the box and nobody needed to click through. Our clients get the best return from snippets on queries that are a genuine first step toward a decision, not the ones that are the whole answer on their own.
Our practical advice for a Singapore business is to spend an afternoon identifying the queries where you already rank on page one and where a complete answer still leaves the user needing something from you. Format those properly: a question as the heading, a direct answer in about 50 words, real list and table markup where the query calls for it, and current figures in SGD. Leave your core commercial pages written for the buyer.
Then measure the right thing. Not whether you hold the box, but whether the pages holding boxes produce enquiries. The same structural discipline that wins a snippet is what makes a page usable inside a generated answer, so none of the work is stranded, but the metric that matters has not changed and never was visibility for its own sake.
If you want an external read on which of your queries are worth the effort, an SEO audit and consulting engagement is the quickest way to get one.
Our pricing page sets out how we scope that work.
If you want to see the prioritisation applied in a category where a handful of queries carry most of the commercial weight, the law firm SEO results write-up follows one through.
Effectively no. Google selects the snippet from pages already ranking for the query, overwhelmingly from the top five results and essentially never from beyond the first page. That makes snippet work a second-stage tactic: earn the ranking first through relevance, depth and links, then format the answer so your passage is the most extractable one available. Teams that reverse this order spend weeks reformatting pages that were never in contention.
Not directly. Snippet selection is a text extraction process that reads your visible page content, and there is no schema type that makes a page eligible. Structured data does other valuable work, including rich result eligibility and entity clarity, but it is not the mechanism here. The things that actually influence selection are heading phrasing, answer position, answer length and whether you used genuine list and table elements rather than styled approximations.
Aim for roughly 40 to 60 words in the extractable block, which is about the length the box displays before truncating. Write it so it makes complete sense removed from the page, because that is what happens to it. Then continue below with the detail, caveats and local specifics for the reader who clicks through. The common failure is burying the answer three paragraphs into a warm-up that no extraction process will ever reach.
Because your standard listing is removed when the snippet is shown, so you traded a blue link for a box. If the box fully answers the question, the user has no reason to continue. This is the expected outcome on definitional queries and it is not a fault in your page. The response is to check whether that query was ever commercially valuable, and if it was not, to move your formatting effort to queries where the user still needs something from you.
Yes. The max-snippet robots meta directive limits how much text can be displayed, and setting it to a small value removes you from paragraph snippet eligibility. The data-nosnippet attribute excludes specific passages within a page. Both are blunt instruments and we would rarely recommend them, since the same directives affect how your description displays elsewhere. Reformatting the page is almost always the better lever.
They are separate features drawing on similar extraction. People Also Ask expands into short answers pulled from various pages, and its questions are a useful map of what the system considers adjacent to your topic. Covering several of those questions well on one page puts you in contention across a cluster rather than a single query. It is also a cheaper way to earn visibility on queries where the main snippet is locked up by a large publisher.
Yes, though less consistently on the query classes where a generated summary appears instead. The two features coexist and the mix varies by query type and by market. The practical implication is not to choose between them. The page structure that wins a snippet, a clear question heading with a direct and self-contained answer, is the same structure that makes your page usable as a source in a generated answer, so the work serves both.
On informational content, usually yes, because a click was always uncertain there and the visibility has value. On commercial service pages, be more careful. A page written primarily for extraction reads thinly to the human buyer who does click, and those buyers are the ones who generate revenue. One clean answer near the top is fine. Converting the whole page into question and answer blocks is a trade we would not make.
A rank tracker configured for Singapore results will flag which of your tracked queries display the feature and whether you hold it. Failing that, search the query yourself in an incognito window with location set locally, since results differ by country and personalisation. Then cross-reference against Search Console to see which of those queries you already rank on page one for. The overlap between those two lists is your actual target list.
Selectively. National search volumes here are small, so a snippet on a high-volume definitional query may represent very few in-market users. The exercise that pays is filtering by whether the user must still act after reading the box. A table snippet on a pricing query with 40 monthly searches can produce more enquiries than a paragraph snippet on a definitional query with 900, and we have watched exactly that happen in local accounts more than once.
If you want to know which featured snippet opportunities are genuinely worth your time, we will pull the queries where you already rank on page one, mark which ones trigger a snippet and in what format, and tell you plainly which we would not chase and why. It takes us about an hour and there is no charge for it. Talk to us and we will send the shortlist across regardless of what you do next.
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 →
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.

SEO vs SEM is usually the wrong question. Learn what the terms really mean and how to use ads and Search Console data to decide which searches to earn or buy.

Google Search Console login problems usually start with verification and ownership. Learn how to get in, fix access errors and offboard agencies safely.

The click through rate formula is clicks divided by impressions. Learn what each platform counts, the averaging trap and how to set it up in Google Sheets.

A click through rate means nothing on its own. Learn to compare CTR by position, query type and SERP features, and see what low CTR is really telling you.

What a google analytics certification proves, what it misses, how to verify one and the practical questions to ask before you hire a marketer or freelancer.
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.