# Chapter 5 · Writing pages (2): price pages and how-to-choose pages

> GEO Playbook · Dental & Aesthetics v1.0 · Canlah AI · CC BY 4.0 · Web page: https://canlah.ai/playbook/dental/price-pages/
> Markdown edition for AI assistants, same content as the web page. Figures are code blocks (wireframe / mermaid / bars / steps / split); "→" links open the matching section on the web, and the same URL with .md is its Markdown edition.

## 5.4 ② Price guide page: dental's first page

**What you'll do in this section**: put every service you currently offer into one master table, move the official public benchmark into its own section after the table, leave no room at all for peer clinics' price ranges, and change every price column to the final price incl. GST. When you're done, you'll have a dental price guide page ready to build straight away, with a reason for every missing item and a record of every revision.
```wireframe id=pt02 Figure: price guide page block order; AI most often copies the conclusion block and the price grid table under the first H2
title | <Category> Cost in Singapore (<Year>): <Price hook> | req
H1 | Statement, with the year; the Chinese version includes the country | req
Byline and date line | Named reviewer + month and year, or "Updated <date>" | opt
Conclusion block | 40–130 words: own sentence + market sentence, one of four formats | req,cite,law
Data basis line | One line right above the table: sample, source or check date | opt
H2 main price question | How Much Does <Category> Cost? | req
  First sentence | At <Brand>, <Category> costs S$X (incl. GST) | req,law
  Price grid table | <Type> / Cost / Suitable for, one figure per cell | req,cite,law
Official public benchmark | Its own section, with table number and update date; no own prices inside the section on the strict side | opt,law
H2 What affects the price | First sentence is the market sentence, 3–5 factors each with a numbered H3 | opt,law
Subsidies, insurance and tax H2 | Separate fact paragraph; government figures carry the table number and update date | opt
What's included and excluded H2 | Table or list, 4–8 items | opt
FAQ | 5–12 questions, 40–60 words each | opt
Related price pages, cross-links + CTA | List, only one CTA | opt
```
For the blocks marked law in the General Edition's blueprint, dental follows the five rules below.

1. The number of rows in the master table = the number of items currently offered on the signed-off price list. If a row is missing, you must state why (not offered / discontinued); never leave one out silently. Every row links to the matching single-service price page (see 5.5).
2. The official public benchmark goes in no column of the master table. Give it its own section after the table, stating the table number and update date; that section shows none of the clinic's own prices and never says "we are within the benchmark". The basis: an advertisement that lists prices must not contain a comparative listing of prices (**Statute text**; see Appendix A). As for the wider reading that "same table and adjacent both count as a price comparison": side by side in the same table is Statute text; the same paragraph or an adjacent paragraph is a Conservative line (not statute text) (see 1.3).
3. No peer clinic's price range anywhere on the page, not even one. Whether it is laid out side by side or mentioned in a single sentence, named or unnamed, makes no difference; putting the market range and the clinic's own price in the same table is itself a comparative listing of prices. This rule has no "publishable with written confirmation" exception: a written sign-off resolves only the three stop-and-escalate situations; it never turns something banned as Statute text into something publishable (rules and labels in chapter 1, 1.1, 1.3; rule text in Appendix A.1, A.2).
4. Remove prefixes such as From, starting price, After Medisave and CHAS Up to, and any post-subsidy price, from both the title and the H1; the title formula follows chapter 1, 1.2.
5. The H1 carries the year and is revised once a quarter; when you revise it, update `dateModified` and the visible "Updated" line on the page at the same time, and state on the page which rows changed this time. If the page hasn't been revised by the end of the quarter, treat it as an expired asset.

```split Figure: none of the three cited dental column-header sets can be used as is — the price column becomes the final tax-inclusive price, the subsidy is split out, and the whole peer-range column is deleted
Cited column headers || Dental rewrite
Crown type · price before GST · who it suits || change the price column to the final price incl. GST
Implant system · total price · Medisave-deductible · out of pocket || split into a price table + a separate payment worked-example section
Clinic type · estimated cost || delete: this is a peer price range
```


The payment worked-example section you split out must state that it is a payment method, not a discount (wording in chapter 1, 1.3).

Sample sentence for the standalone official-benchmark section: "MOH lists a CHAS fee benchmark of S$750–S$1,400 (before GST) for a permanent crown (Guide to Dental Treatment Costs, updated 28 Nov 2025)." This sentence appears only in the benchmark section, never in the section holding the price table.

Build the Chinese version as its own separate page, written per 5.3. For the general writing pattern (why the first page is always this one, the blueprint, column headers, sample sentences, most common mistake), see （→ 通用版 5.12 价格指南页（一）：为什么第一页永远是它（pt02））, （→ 通用版 5.13 价格指南页（二）：施工图、列头、句子（pt02））.

## 5.5 ① Single-service price page: one URL per service

**What you'll do in this section**: give every service its own page, show the final tax-inclusive price for each tier right in the first screen, rewrite the cited "range + depending on" sentence pattern into fixed prices, and price aesthetic items by their unit of measure. When you're done, you'll have a copyable single-service price page template that works for both dental and aesthetics.
```wireframe id=pt01 Figure: Single-service price page; AI most often copies the own-price sentence and the price table right after it
H1 | Statement, with the year | req
Byline and date line | Named reviewer + month and year, or "Rate card reviewed <month year>" | opt
Own-price sentence | 1–3 sentences, 40–130 words, opens with the figure | req,cite,law
Price table | Right after the price sentence: service tier + defining condition + price + what's included | req,cite,law
Tier criteria H2 | First sentence states what decides the tier; one figure per tier | opt
Included vs charged-separately table | Item / Fee / Notes | opt
Subsidies and payment H2 | Separate fact paragraph, apart from the price table | opt
Alternatives H2 | First sentence gives your own price for both options; the market sentence can go here | opt
FAQ | Restate the billing rules as complete sentences | opt
Related price pages, cross-links | List | opt
```
```split Figure: in dental and aesthetics, no cell of the most-cited range sentence can be copied as is — each cell has a fixed-price rewrite to put in its place
Cited pattern (banned in this industry) || Dental rewrite
"S$X to S$Y depending on…" || Mild S$X / Moderate S$Y / Complex S$Z (incl. GST)
depending on tooth condition || state the defining condition for each tier clearly (e.g. degree of crowding)
the tier is only known after an examination || "tier confirmed after examination" + a fixed price for the examination itself
the RESULT column of the spec strip || delete: it implies results or a timeframe
aesthetics quoting a range per ml || the final tax-inclusive price per ml
```


Five things the figure cannot show:

1. For the tier that is only "confirmed after examination", the cost of the examination itself is also written as a fixed price — never "varies by patient".
2. For aesthetics, price by unit of measure (e.g. per 1 ml) as a final tax-inclusive price, not a "from" starting price; the quantity used is confirmed at the consultation.
3. In the spec strip (PRICE / RESULT / FREQUENCY / DOWNTIME / PAIN), delete RESULT: it implies results within a given time, and the regulator lists claims of that kind as soliciting (**Statute text**; see Appendix A). The DOWNTIME and PAIN columns conflict with the hard rule against writing recovery time or pain level, so keep only content quoted word for word from the device's instructions for use (IFU) or an NEA document, with the source cited, and delete everything else in those columns (**Conservative line (not statute text)**).
4. The market sentence is optional; if used, rewrite it as a variable sentence carrying no peer figures and put it in the alternatives block or the FAQ. Sample (not measured): "Whitening price depends on method (take-home or chairside), number of sessions and whether a check-up is needed first."
5. List prices row by row, one item per row; do not build tiered package cards or rank options with a recommended pick. This is also why dental does not build an official pricing page; see the stand-in table in chapter 0, 0.3.

> **Example** (dental) One clinic's whitening page was cited by ChatGPT for this sentence: take-home kit S$400, in-chair whitening S$900, gel refill S$200. The shape is right: the two tiers each have a single fixed price, with no "from", and they are not the two ends of one range. But the original sentence didn't say whether GST was included. When dental writes to this shape, every amount must be the final tax-inclusive price, marked "incl. GST".

For dental price lists, use the column headers `Treatment / General dentist / Specialist` or `Item / Fee / Notes`, with the price column always the final tax-inclusive price. For the general writing pattern (structure, sample sentences, most common mistake), see （→ 通用版 5.11 单项目价格页（pt01））.

## 5.6 ⑰ Subsidy page, ㉗ parameter page, ㊳ calculator page, and a stand-in for ㊷

**What you'll do in this section**: for the subsidy and parameter pages, write only two sections — the clinic's own final price, and the official limit facts; the calculator only estimates the process, never a price; and rewrite the time-limited promotion you wanted to run into a standing bundle-price page. When you're done, you'll have a minimum viable version of each of these three types, plus one rewrite path that turns a time-limited promotion into a lawful bundle price.
```wireframe id=pt17 Figure: Subsidy and limit rules page; AI copies cells from the itemised limit table
H1 | scheme's full name with abbreviation | req
First-screen summary + update date | one summary sentence, update date in the first screen not the footer | req
H2, three to four questions | What is it / Benefits / Who is eligible / How to use | req
Itemised limit table | one row per item + upper limit amount | req,cite
Old and new tables side by side | when limits change: one current table, one for the new effective date | opt
Eligibility table | which group can use which tier, listed by group | req
Numbered procedure steps | download form → fill in → submit | opt
Your own three-column table | item / limit / what you pay; availability depends on the side | opt,law
```
```wireframe id=pt27 Figure: Parameter and rate basis page; AI copies the current value given as a tax-exclusive and tax-inclusive pair
title | the original-question form, or the category + location form | req
First-screen basis sentence | who sets the price, how often it is adjusted | req
Current value as a pair | tax-exclusive and tax-inclusive values fixed in the same sentence | req,cite
Time-tiered table | parameter name / current value / effective period | req
Single current-value sentence | one sentence stating the current figure and the year it took effect | req,cite
Formula and worked example | total formula + a three-column worked example (not measured) | opt
Units H2 | unit and average usage | opt
Everything in static HTML | the numbers are still there with JS off | req
```
```wireframe id=pt38 Figure: Calculator page; what gets copied is the first-screen tier conclusion and the static reference table, not the widget
title = H1 | tool name + year + what it calculates + site name | req
First-screen tier conclusion | one sentence with a number range: tier A maps to X, tier B maps to Y | req,cite,law
Calculator widget | JS, cited 0 times, don't invest in it | opt
Large static reference table | pre-computes and lays out the results flat | req,cite
FAQ | 5 questions, restating the definition of the calculation | opt
Definition sentence for the calculation | one sentence: how it's added up, what the range is | req
```
### Subsidy and limit rules page (pt17)

1. When AI is asked "how much do I pay", it pieces together three things on the spot (the subsidy cap, the fee benchmark and whether the treatment can be claimed) and does the maths itself. The only thing it lacks is your clinic's actual price, so that price must get a section of its own, written as final tax-inclusive prices.
2. The own three-column table in the General Edition's blueprint (item / limit / what you pay) cannot be used in dental. Split it into two sections instead: one for the clinic's own final tax-inclusive prices, one for the official limit facts (with the table number and update date). The two sections must not share a table or sit directly next to each other, and must not use prefixes such as After Medisave or CHAS up to (rules in chapter 1, 1.2, 1.3).

The General Edition's own sentence that rolls the actual price charged, the subsidy offset and "what you actually pay" into one sentence cannot be used in dental either: it is simply the three-column table written as a sentence. The clinic's own price section contains only "<item name>: S$<final tax-inclusive price> (incl. GST), including <mandatory inclusions>". How much Medisave / CHAS can offset goes into the payment worked example in a separate section, which states that it is a payment method, not a discount (wording in chapter 1, 1.3).

> **Example** (dental) On MOH's CHAS page, the sentence giving the limits (up to $830 a day, $240–$5,290 for surgical items; paraphrased) was lifted whole from a cell of the itemised limit table and placed straight into the answer. Write the official limit facts section to the same shape: one item per row, one cap per cell, with the table number and update date attached.


### Parameter and rate basis page (pt27)

1. The parameter must be the final tax-inclusive price, never a range or a "From" price; write "which tier applies is confirmed after examination" as its own separate sentence.

### Calculator page (pt38)

1. Build only process-type calculators, such as number of visits or duration, and the output must never be a price commitment. The static table beside it holds only items that already have a fixed price; a total that can only be set after an examination is never written as a final price.

### ㊷ A stand-in for the time-limited promotion page

1. Dental and aesthetic clinics never offer a discounted price with an expiry date, a discount off the original price or a free examination; none of the three has an exception (**Statute text**; rules in chapter 1, 1.2; see Appendix A).
2. The stand-in is a standing bundle-price page: one URL per bundle, written per the single-service price page in 5.5.

```split Figure: strip the expiry date and "nett" out of the cited promotion sentence, and the same bundle of items can still be built as a standing, copyable price
Cited pattern (banned in this industry) || Dental rewrite
Scaling + polishing + 3D scan, S$109 nett (paraphrased) || "S$<final tax-inclusive price>, including <mandatory inclusions>"
time-limited, has an expiry date || remove it, make it a standing price
the word "nett" and any offer wording || remove them
posted under a "latest news" section || one URL per bundle
participating clinics || state exactly which clinics it applies to
```

Why the stand-in is worth building: for that one question, the cited promotion page was the only citation that came from the clinic itself, carried a price and was written into the answer by AI as a positive fact; the same clinic's homepage contributed only neutral facts (measured sample). Turn it into a standing price and the copyable price sentence stays; it just no longer touches the promotion rules. Fill the price slot with the final tax-inclusive price from the signed-off price list.


For why AI never clicks the calculator and just takes the unit price and the limit and does the maths itself, and for the general writing pattern of all five types, see （→ 通用版 5.21 价格类档 B：补贴限额、参数口径、品类列表、计算器、限时促销）.

## 5.7 ④ Comparison page: compare procedures, not clinics

**What you'll do in this section**: keep the comparison page to procedures, treatment options and materials only, put the verdict up front, and write out the scenarios where "the other option fits better"; organisations, doctors, success rates and testimonials must not appear at all. When you're done, you'll have a procedure-comparison page structured by scenario, not by lining up candidates.
```wireframe id=pt04 Figure: comparison page; AI most often copies the conclusion block and the verdict in the first sentence of the first H2
title / H1 | title is a question, H1 is a statement, both carry the year | req
Comparison-basis line | Check date + both sides' version or tier | req
Conclusion block | 60–120 words, or a by-scenario verdict table | req,cite,law
H2 main question | First sentence gives the verdict with its condition, followed right away by the scenario table | req,cite
Criteria H2 | Numbered H3s, each first sentence states the threshold that flips the verdict | opt
Master comparison table | In the first 25%, one checkable value per cell | req,law
Per-criterion H2 × N | The verdict goes into the heading; the first sentence repeats it with a figure | opt
Price comparison H2 | Same tier, same currency, same billing cycle | opt
Scenarios where the other one fits better H2 | ≥2 specific scenarios, each paired with one public fact about the other side | req
Not-a-fit conditions H2 | Skip it if… | opt
Data sources and dates | Source + check date for every figure | req
FAQ | H3 questions, first sentence answers directly | opt
Related comparison pages, cross-links | 3–5 entries | opt
```
```split Figure: everything a dental comparison page can compare is the procedure itself — the moment it touches organisations, a track record or "choose us", it has crossed the line
Can compare || Cannot do
Procedure: implant vs bridge || comparing organisations, comparing doctors
Options and materials || success rate, rescue rate
Conditional sentences comparing indications || any variant of switcher testimonials
Official benchmark: its own separate section after the table || benchmark in the same table as the clinic's own price
System name: a factual statement || linking to the manufacturer's page
Verdict placed up front, stating the scenario where the other option fits better || steering the conclusion toward "choose us"
```

Four things the figure cannot show:

1. What can be compared: implant vs bridge, porcelain crown vs zirconia, clear aligner treatment vs traditional braces. Rewrite claims like "better" or "more advanced" as "suitable under which conditions".
2. No variant of a testimonial (anonymous, initials only, blurred, with a disclaimer, with the patient's written consent) can save this section. Rewrite it in place as neutral conditional sentences comparing indications: which objective conditions must be met before moving from option A to option B, when the original option stops being suitable, and what records to bring for a referral (X-rays, past records, medication list). Write all of it as conditional sentences, with no specific patient and no account of an outcome (rules in chapter 1, 1.4).
3. The official-benchmark section has no "put it side by side once confirmed case by case" exception (rules in chapter 1, 1.3).
4. The most common mistake: turning a procedure comparison into an organisation comparison ("why our implants beat the chain clinics") — this is a compliance violation, not a layout issue.

Comparison pages are mainly cited on the AI Mode leg. When ChatGPT meets a "vs" question, it goes to the procedure pages on each side and builds the table itself: for the dental "implant vs bridge" question, it cites the procedure-education pages of two medical institutions. So every one of the clinic's own figures on a comparison page must come from the same source and carry the same value as on the single-service price page (the same-source point is not measured).


> **Example** (dental) Asked "implant vs bridge, which suits a single missing tooth better", both legs generated their own comparison table, with different column headers: ChatGPT used `Option / Best when / Main advantage / Main downside`; AI Mode used `Feature / Dental Implant / Dental Bridge`, with rows such as Lifespan and Impact on Adjacent Teeth. Both tables compare the procedures themselves. The "Best when" header hits the official list of laudatory terms (see chapter 1, 1.5), so send it to the compliance officer for a ruling; the safer wording is "Suitable when" (**Conservative line (not statute text)**).

The cited dental "Invisalign vs braces" comparison table put the Singapore market range and the clinic's own price in the same table — do not copy the market-range column. For the general writing pattern, see （→ 通用版 5.15 对比页（pt04））.


## 5.8 ㊸ Criteria-based selection guide and ㊹ concept pillar page

**What you'll do in this section**: replace the "which clinic is good" citations that dental cannot get into with a criteria-based guide to "how to choose a clinic", which lists criteria, ranks no one and names no one; use the concept pillar page only to compare procedure concepts, never organisations. When you're done, you'll have a minimum viable version of each of these two types, plus one boundary you must never cross.
```wireframe id=pt43 Figure: Block order for the Buyer's selection framework (not measured as a whole; spec taken from an external teardown; no block is marked cite); on the strict side and when switch E is on, the two blocks marked law are removed entirely
title | How to Choose a <provider type> for <scope> | req
Quick answer | 2–3 sentences answering the title question directly, followed by a Disclosure statement | req
H2 What does <thing> measure? | Opens with one punchy sentence, then 5–6 bold labels | opt
H2 What a useful <thing> should show | Numbered, 5–7 items, one sentence each | opt
H2 Research snapshot | Evidence paragraph: what the status quo is missing | opt
H2 Evaluation dimensions | 5 "Label:" paragraphs, no bullets | req
H2 Provider comparison matrix | Column headers: Provider / Dimension 1 / Dimension 2 / Best for | opt,law
  Per-provider card | One card per provider; write each limitation as a buyer task | opt,law
H2 How to interpret common results | Short sub-sections without bullets, each opening with a label | opt
H2 How to run a defensible baseline | Numbered, 5–10 imperative steps | opt
H2 What a free diagnosis cannot prove | 100–160 words, spelling out what the diagnosis cannot prove | req
H2 Questions to put in the RFP | 8 items, each a complete question ending in a question mark | req
H2 Final recommendation | When to switch from X to Y, 2 paragraphs | opt
FAQ + Method and sources | 5–6 questions, at most 2 naming your own brand | req
```
```wireframe id=pt44 Figure: block order for the concept pillar page (the whole diagram is not measured, spec taken from an external teardown, no block marked cite)
title | {NEW} vs {OLD}: the concept gap + the outcome noun | req
intro | one-line axiom: the two complement each other, they're not substitutes | req
H2 Why AI doesn't recommend you | 200–280 words, 4–5 buyer questions inline | req
H2 The difference in 30 seconds | 180–250 words + comparison table | req
H2 Why it matters now | 250–350 words | req
H2 What is OLD / What is NEW | 250–320 / 320–420 words respectively | req
H2 The core difference | 250–320 words + difference table | req
H2 Why you need NEW when you already have OLD | 450–600 words + figure + 3 "Pattern" subheadings | req
H2 The most common problems in real projects | evidence section, 550–700 words + figure + numbered subheadings | req
H2 Four common problems | 350–450 words: doesn't exist / known but not recommended / wrong information / buried under old information | req
H2 Priority differs by engine | 300–400 words + figure + comparison table | req
H2 Will it replace it → How they work together → Which to do first | 180–250 / 300–400 / 300–400 words + 3 "Case" subheadings | req
H2 Summary | 250–320 words + figure: OLD governs search results, NEW governs AI answers | req
FAQ | 8–11 items, answers must have no bold, no links, no bullets, no brand name | req
```
Neither blueprint is measured at all: pt43's specs are taken from an external teardown, and pt44 currently has 0 cited samples. ㊸ is dental's stand-in for the ⑤ list page it does not build; the full stand-in table is in chapter 0, 0.3.

### ㊸ Criteria-based selection guide

1. Delete the two blocks marked law in the blueprint (the provider comparison matrix and the per-provider cards) entirely. The body keeps only three blocks: evaluation dimensions, the RFP questions and "What a free diagnosis cannot prove". The brand name appears only in the four places set out in 5.2 [Conservative line (not statute text)].
2. List criteria, don't rank, don't name peers: this is currently dental's only lawful route to the share of citations that goes to list-type pages. Not measured, 65% confidence; schedule one page as a control. The figure of 19.6% of citations going to list-type pages is an all-industries figure; in the healthcare sub-sample that cell is 0% (from deltaV's page-type breakdown by industry). There is no guarantee yet that this route can win list-type citations.
3. Four boundaries for a long-form educational piece: it must not steer towards efficacy, outcomes or success rates; it must not contain any item on the ban list in chapter 1, 1.5; it must carry the byline of the organisation or a licensed practitioner; it must not recommend the clinic over its peers anywhere in the text.


For how to write each block, see （→ 通用版 5.30 档 C 四型：未验证页型）.

### ㊹ Concept pillar page

1. You may build a page like "implant vs bridge: differences, indications, cost gap": compare procedures only, never organisations; write the cost section as final tax-inclusive prices per 5.5; do not link to the manufacturer's page; carry a reviewer line reading `Medically reviewed by Dr <name> · <month year>`. A byline on medical pages is the only byline requirement that held up in our measurements.
2. The title wording decides whether a page belongs to ④ the comparison page or to this type; how to judge it is in the figure at （→ 通用版 5.30 档 C 四型：未验证页型）. Before you write, fix which side the title falls on; each keyword may be taken by one page only.