The Decision You Can't Undo: Dismantling Lodging's Listing Illusions
WEEK 95 :: POST 1 :: GOOGLE GEMINI
Directions Given To The A.I. This Week+
Instructions Given to each A.I. — Please provide 3 prompt variations that share this objective:
Each A.I. also received two static attachments: the blog post template (structure) and the authoring instructions (voice and standards). The text below is the week-specific assignment as sent — reflowed for the web; wording unchanged.
I'd like you to write this week's Ketelsen.ai post. Two files are attached: the blog post template (the structure to follow) and the authoring instructions (context, voice, and standards). Please read both before you begin, then produce the complete post in a single response.
This week's theme: "Where You Sleep Changes Everything" — Lodging and Neighborhood Intelligence.
This is Week 4 of an eight-week series on planning a vacation with AI. Week 1 established the reader's real constraints — a validated budget ceiling and a constraint profile. Week 2 chose the destination. Week 3 locked the flights, which fixed the dates and the arrival airport. This week they decide where they will actually wake up every morning.
Lodging is the decision travellers most often get wrong in a way they cannot undo. A flight that disappoints is four hours; a badly chosen neighbourhood is the whole trip. It is also the part of travel where the listing is written by someone whose interest is opposed to the reader's: the price shown is not the price paid, the photographs are chosen and cropped, the reviews are a mixture of the genuine, the incentivised, and the fabricated, and the words that sound most reassuring — cozy, vibrant, up-and-coming, steps from — are the ones most often doing concealment work.
The job this week is to give the reader a defensible lodging decision: this property, in this neighbourhood, at this true all-in cost, with these cancellation terms, chosen for reasons they can state.
The three prompts should help a reader work through:
- True total cost, not nightly rate. Resort fees, cleaning fees, service fees, occupancy taxes, parking, security deposits, mandatory "destination charges" — the drip-pricing pattern from Week 3's airfare work reappears here in a different costume, and it is usually worse. Two properties whose headline rates differ by 15% can invert once everything is added.
- Review forensics. How to read a review set rather than a review score: weighting recent reviews far more heavily than old ones, spotting clusters of suspiciously similar phrasing or timing, reading the three-star reviews where the honest detail lives, and noticing what a stream of positive reviews never mentions. A property whose complaints are all about the same thing is telling the reader something a 4.5 average conceals.
- Neighbourhood, decided by block and not by city. Walkability to what the reader actually plans to do, transit access at the hours they will use it, noise, and how the character of an area changes between a Tuesday afternoon and a Saturday night. The right question is never "is this a good area" but "is this a good area for this trip, these travellers, and these hours."
- Cancellation-policy risk as a priced feature. A non-refundable rate is a discount in exchange for accepting a risk. The reader should be able to say what that risk is worth to them rather than defaulting to the cheaper number.
- Decoding listing language. What cozy, charming, lively, convenient for transport, and partial view reliably mean, and which specific photograph or floor-plan absence should prompt a direct question to the host.
The output a reader should walk away with is a ranked shortlist they can act on: two or three properties scored on true cost, review credibility, location fit, and cancellation risk, with the reason each one ranks where it does.
A note on the strongest version of this week: at the advanced end, this is a lodging dossier matrix — candidate properties as rows; true all-in nightly cost, review-pattern credibility, location score against this trip's actual itinerary, and cancellation-policy risk as columns; with the reader's own weightings applied and the trade-offs made explicit. That structure is worth reaching for, and it is the natural successor to Week 3's fare decision framework.
A hard constraint, carried forward from Week 3 and just as binding here. AI models cannot see live listings, current rates, or today's availability, and their knowledge of specific properties is stale, thin, and — for anything below the level of a famous hotel — frequently invented. No prompt in this post may ask the AI to recommend a named property, quote a current rate, describe a specific listing it has not been shown, or assess a named hotel's present condition. A confidently hallucinated hotel recommendation is worse than a hallucinated airfare: the reader books it.
Neighbourhood character deserves particular care. A model will happily describe a district's safety or atmosphere from training data that is years stale and was never reliable — and this is the one place in the series where a wrong answer can put someone somewhere they should not be. Prompts should have the AI generate what to check and where to check it — the questions to ask, the hours to look at, the sources to consult — rather than deliver a verdict.
Design the prompts so the AI does what it is genuinely good at: structuring the comparison, exposing the fee layers, teaching the reader to read a review set, and naming what to verify. The reader supplies the listings; the AI supplies the judgement framework. Posts that blur that division should expect to be marked down on Practical Utility, exactly as in Week 3.
Series dependency chain, for the Metadata block: Week 4 consumes the confirmed routing and dates from Week 3, the destination from Week 2, and the budget ceiling from Week 1 — lodging is scored against what the flights left of the budget, and a nightly rate that breaks the ceiling is a signal to revisit the property tier or the neighbourhood, not to quietly raise the budget. Week 4 produces the booked lodging and its location, which Week 5's itinerary assumes as its starting point every morning: the itinerary is built outward from where the reader wakes up.
Because readers may arrive at this post without having read Weeks 1 to 3, the prompts should work for someone who knows their destination, dates, and rough remaining budget, while making clear they get far more from them with a real constraint profile, confirmed flights, and a live set of candidate listings in hand.
Three difficulty tiers as always — Beginner, Intermediate, Advanced — each a genuinely different approach to the same problem, not the same prompt at three lengths.
On examples: this is a consumer travel topic. The template lists tech startup / retail / freelance as suggested industry examples — those are marked MAY, and this week you should almost certainly adapt them. Families needing two bedrooms and a kitchen, couples choosing between a central hotel and a quieter rental, solo travellers weighing safety and walkability, older travellers for whom stairs and lift access decide everything, and anyone booking around a fixed-date event are the right contexts here. Choosing them over the suggested business examples is correct behaviour and will not be scored against you.
A note on supplied figures. Anything marked `[SUPPLIED — use as given]` above came from Ketelsen.ai's own research brief. Use it freely — you are not fabricating by repeating it, and you will not be marked down for leaving it uncited. Do not attach an invented source to it. (No supplied figures this week. Given the live-listing constraint above, this is a bad week to invent any — if you find yourself reaching for a typical resort fee or an average nightly rate, that is the signal to restructure the prompt so the reader supplies the real number instead.)
## BEFORE YOU SUBMIT — STRUCTURAL CHECK
(This block is identical every week. It exists because these specific items are the ones posts drop, and a dropped structural item costs compliance points for something that takes one minute to add.)
Your post is parsed by a script before any human reads it. Confirm all seven:
1. ☐ Response begins with `PLATFORM: <your name>` and `WEEK: 4` 2. ☐ `## Lead` present once, at the very top, before Variation 1 3. ☐ `## In one line` present in all three variations 4. ☐ `## What this prompt gives you` present in all three variations 5. ☐ `## The Prompt` present in all three variations, with the prompt in double quotes beneath it 6. ☐ `## Introductory Hook` and `## Current Use` present in all three variations (three of each — not one) 7. ☐ Every template heading written as `##`, none bolded instead; prompt breakdown is running text split on ` : `, with no `###` headings inside it
A complete post has 57 `##` headings. If your count is well short, a section is missing or was bolded instead of hashed.
One extra check this week: confirm no prompt asks the AI to name, rate, or price a specific property, or to pronounce on a neighbourhood's current safety. Those must be things the reader goes and verifies.
Lodging is the travel decision you can never undo. A disappointing flight is over in four hours, but a badly chosen neighborhood or a deceptive rental defines the entire trip. This week, we dismantle the listing illusions—drip pricing, curated review scores, and vague location promises. Using three prompt tiers, you will move from simply exposing hidden checkout fees to running forensic analysis on reviews, and finally building a full lodging dossier matrix to score properties against your actual itinerary. The AI won't guess where you should stay; it will give you the framework to ensure the choice you make is bulletproof.
The Fee De-Cloaker
Uncover the true total price behind deceptive nightly rates.
The nightly rate on a travel platform is often a work of fiction. Between cleaning fees, service charges, occupancy taxes, and mandatory "destination fees," two properties with vastly different headline prices can easily swap places once you reach the checkout screen. Doing this math manually across ten open tabs is exhausting, which is exactly what the platforms are banking on. This prompt strips away the drip-pricing camouflage and hands you the only number that matters.
Why this matters now
With travel platforms facing increasing pressure over junk fees but still burying the final numbers deep in the checkout flow, travelers need a fast way to normalize costs. This prompt cuts through the noise today, allowing you to instantly compare the real financial impact of different properties without being fooled by a low teaser rate.
I am evaluating a vacation rental. I will paste the raw checkout page text and fee breakdown below. Calculate the true nightly cost by dividing the grand total by the exact number of nights. Then, list every fee that is not the base room rate, such as cleaning, service, resort, or tax charges. Present the final numbers clearly so I know exactly what I am paying per night.
How the AI reads this prompt
"I am evaluating a vacation rental. I will paste the raw checkout page text and fee breakdown below."
This sets the operational boundaries. Without this, the AI might try to search the web for average rental prices or hallucinate a listing. By explicitly stating that you are providing the text, you constrain the model to act solely as a calculator for your specific data.
"Calculate the true nightly cost by dividing the grand total by the exact number of nights."
This forces a specific mathematical operation. If left vague (e.g., "tell me the cost"), the AI might just repeat the advertised nightly rate and list the fees separately. This instruction ensures you get the normalized baseline needed to compare properties side-by-side.
"list every fee that is not the base room rate, such as cleaning, service, resort, or tax charges."
This requires the model to isolate and categorize the junk fees. Without this directive, the model might just give you the final sum, hiding the context of why the property is so expensive. It teaches you to always force the AI to show its work on hidden variables.
Practical examples from different industries
The Traveling Family:
A parent booking a beach house for five nights sees a listing for $200 a night. They paste the checkout page into the prompt. The AI calculates that a $250 cleaning fee, a $150 platform fee, and local taxes push the true nightly cost to $315, completely changing the budget equation. The Solo Freelancer: A freelancer looking for a month-long city stay wants to avoid hidden deposits. They paste the terms of a serviced apartment. The prompt highlights a mandatory weekly linen fee and a non-refundable administrative charge, allowing the freelancer to realize a standard hotel is actually cheaper.
Creative use case ideas
- Bachelor/Bachelorette Party Splits: Pasting the final group booking costs to get a clean breakdown of the true per-person, per-night cost to share in a group chat.
- Conference Budget Justification: Proving to an accounting department that an "expensive" hotel next to the venue is cheaper overall than a budget rental with massive service fees and daily transit costs.
- Long-Term RV Park Pricing: Decoding the confusing utility hookup fees, pet fees, and resort taxes that RV parks often layer onto their base monthly rates.
Adaptability tips
You can easily adapt this prompt to compare rental cars, event spaces, or even coworking memberships. Just change "vacation rental" to the service you are evaluating, and the AI will apply the same fee-isolation logic to whatever checkout text you provide.
Pro tips
Paste the checkout screens for two properties at once and add, "Tell me which one is actually cheaper and by what percentage." This turns the AI into a comparative engine, saving you a step.
Prerequisites
You must have navigated far enough into the booking platform's checkout process to reveal the final, all-in price and fee breakdown.
Required tools
Any standard AI text model (ChatGPT, Claude, Gemini). No specialized integrations required.
Frequently asked questions
Why can't I just ask the AI to find the price of a specific hotel?
AI models do not have reliable access to live availability or current rates. Even if they browse the web, they frequently pull stale data or hallucinate numbers. You must supply the real price from your own screen. Will the AI understand a messy copy-paste from a website? Yes. Modern models are excellent at parsing unstructured, messy text. You can drag your cursor over the entire checkout column, copy it, and paste it directly. The AI will extract the numbers accurately. What if the platform hides fees in a dropdown menu? You have to expand those menus before copying the text. The AI can only analyze the text you provide; if the platform visually hides a fee from your clipboard, the AI cannot factor it in.
Recommended follow-up prompts
If the price works, the next step is location. Look to the intermediate prompt in this post to analyze the neighborhood and the reviews to ensure the property is actually worth the true cost.
Tags and categories
Tags:
budgeting, hidden fees, rentals, cost calculation Categories: Budget Management, Deal Analysis
Citations
NOT APPLICABLE
The Review Forensics Engine
Decode review patterns to spot hidden property flaws.
A 4.5-star average tells you almost nothing. Property listings are written by people financially motivated to obscure the negatives, and reviews are a minefield of the genuine, the incentivized, and the fabricated. The words that sound best—lively, charming, up-and-coming—are usually doing heavy concealment work for loud, cramped, and gentrifying. To know what you are actually booking, you need to stop reading the score and start reading the pattern. This prompt turns the AI into a forensic analyst that reads between the lines.
Why this matters now
Travelers frequently arrive at rentals only to discover construction next door or an elevator that has been broken for months. By running recent reviews through this prompt, you can identify localized, current issues that a multi-year average score deliberately masks.
I am considering a lodging option and will paste a batch of reviews below, specifically focusing on recent stays and 3-star ratings. Read them as a forensic analyst. Do not give me an average sentiment or summarize the positive fluff. Instead, identify specific, recurring complaints (e.g., street noise, broken amenities, poor temperature control). Tell me what is conspicuously missing from the positive reviews. Output a bulleted list of the actual risks I face if I book this property.
How the AI reads this prompt
"Read them as a forensic analyst. Do not give me an average sentiment or summarize the positive fluff."
This explicitly blocks the AI from acting like a polite concierge. Without this negative constraint, the model will likely output a balanced summary ("Guests loved the location, but some noted noise"). By defining the persona as a forensic analyst, you force it to look for discrepancies and problems.
"identify specific, recurring complaints"
This tells the AI to look for clusters of data. An isolated complaint about a bug might be a fluke; three complaints about Wi-Fi drops over two months is a structural issue. Without this, the AI might weigh a one-off grievance equally with a systemic failure.
"Tell me what is conspicuously missing from the positive reviews."
This is the most powerful part of the prompt. It forces the AI into second-order thinking. If 50 positive reviews praise the "cozy decor" but absolutely no one mentions the "fast Wi-Fi" or "comfortable bed," that silence is data. Without this instruction, the model only analyzes what is present, missing what is deliberately omitted.
Practical examples from different industries
The Older Traveler:
An older couple is looking at a highly rated historic hotel. They paste the reviews. The AI notes that while everyone praises the charm, no one mentions an elevator, and multiple 3-star reviews complain about dragging luggage up narrow stairs. They decide to book elsewhere. The Business Traveler: Someone needing reliable sleep before a major presentation runs the reviews of a central downtown rental. The AI flags a recurring pattern: recent guests consistently mention "lively street energy" on weekends, translating the coded language to mean severe noise pollution from the bar downstairs.
Creative use case ideas
- Evaluating "Up-and-Coming" Neighborhoods: Feeding the AI neighborhood descriptions to translate real estate speak (e.g., "transit-adjacent" means a train runs right behind the house).
- Checking Pet-Friendly Claims: Analyzing reviews to see if "pet friendly" just means they tolerate dogs, or if there is actually green space nearby for walking them.
- Accessibility Verification: Parsing reviews to see if a property claiming wheelchair accessibility actually has flat thresholds and roll-in showers, as noted by previous guests.
Adaptability tips
You can use this same forensic framework for software purchases, Amazon products, or hiring contractors. Just swap "lodging option" for the product, and instruct the AI to look for recurring technical glitches or missing praise regarding customer support.
Pro tips
Sort the reviews on the platform by "Newest" first, rather than "Most Relevant," before you copy them. The AI needs chronological data to tell you if the air conditioning broke last week, regardless of how good the reviews were in 2023\.
Prerequisites
You need to copy a substantial block of raw reviews—ideally a mix of recent 5-star, 1-star, and especially 3-star reviews, where the most honest trade-offs are usually documented.
Required tools
Any standard AI text model.
Frequently asked questions
Why specifically 3-star reviews?
Five-star reviews are often brief or incentivized, and one-star reviews are often emotional rants. Three-star reviews are usually written by reasonable people who liked the property but felt compelled to document a specific, factual flaw. How many reviews do I need to paste? Aim for at least 15-20 reviews. If you only provide three, the AI cannot establish a pattern. Do not worry about formatting; just highlight, copy, and paste the whole block. Will the AI know if a review is fake? It cannot know for certain, but it can spot unnatural phrasing or repetitive language clusters that indicate a bot or paid review farm. The instruction to look for "what is missing" helps bypass fake praise.
Recommended follow-up prompts
Once you have vetted the reviews, use the Advanced prompt in this post to measure the surviving candidate properties against your actual daily schedule.
Tags and categories
Tags:
review analysis, forensic reading, risk assessment Categories: Deal Analysis, Qualitative Research
Citations
NOT APPLICABLE
The Lodging Dossier Matrix
Score candidate properties against your actual itinerary and risk tolerance.
Comparing properties by toggling between browser tabs is how critical details slip through the cracks. You might find a rental with a great nightly rate, only to realize the cancellation policy is draconian and it requires a 45-minute commute to the things you actually want to see. To make a defensible decision, you need all the variables in one place, weighed against your specific travel behavior. This prompt acts as your personal logistics coordinator, processing multiple files and outputting a rigorous decision matrix.
Why this matters now
Travelers are increasingly using local files and AI desktop integrations to manage complex planning. By utilizing local files instead of pasting massive blocks of text, you can feed the AI the full terms and conditions, your actual itinerary, and the fee structures of multiple properties to generate a single, mathematical ranking.
Act as a critical travel analyst. I have attached local files containing the listing details, fee structures, and cancellation policies for three candidate properties, along with a document detailing my specific daily itinerary. Create a Lodging Dossier Matrix. Score each property on: 1\. True all-in nightly cost, 2\. Commute time to my itinerary's first stop each day, 3\. Review credibility based on the text provided, 4\. Cancellation-policy risk. Weight commute time and cancellation risk heavily. Recommend the best fit and explicitly state the trade-offs I am making with each option.
How the AI reads this prompt
"Act as a critical travel analyst. I have attached local files containing the listing details, fee structures, and cancellation policies for three candidate properties"
This leverages advanced desktop integrations (like Claude's local file access) to process complex, multi-source data. Without defining the input format as files, the AI might get confused by a massive wall of pasted text. Setting the role to "critical travel analyst" ensures the output is analytical, not promotional.
"Score each property on: 1\. True all-in nightly cost, 2\. Commute time to my itinerary's first stop each day, 3\. Review credibility based on the text provided, 4\. Cancellation-policy risk."
This establishes the exact columns of the matrix. If you don't define the criteria, the AI will invent its own, likely focusing on superficial things like decor rather than logistics. This forces a structured, apples-to-apples comparison.
"Weight commute time and cancellation risk heavily. Recommend the best fit and explicitly state the trade-offs I am making with each option."
This dictates the evaluation algorithm. It prevents the AI from simply picking the cheapest option. By forcing it to state the trade-offs, you ensure you understand exactly what you are sacrificing (e.g., "Property B is cheaper, but you accept a non-refundable risk and lose an hour a day in transit").
Practical examples from different industries
The Conference Attendee:
A professional needs to balance a strict corporate budget ceiling against proximity to a convention center. They attach the hotel options and their session schedule. The matrix reveals that a slightly more expensive hotel across the street actually saves money when factoring in the cost and time of twice-daily rideshares. The Family Reunion Coordinator: A family booking a large house attaches the terms for three properties. The AI highlights that Property C requires a massive, non-refundable deposit six months out—a critical risk for a group of 15 people whose plans might change. They choose Property A despite a higher nightly rate because of its flexible cancellation terms.
Creative use case ideas
- Event-Based Travel: Booking around a marathon or festival where morning transit times are critical and road closures might impact location scores.
- Medical Travel: Patients traveling for specialized care can use this to score properties based on walkability, lack of stairs, and proximity to a specific hospital campus.
- Digital Nomad Hopping: Scoring long-term stays based strictly on desk setups, verified internet speeds from reviews, and proximity to transit hubs for weekend trips.
Adaptability tips
You can customize the matrix criteria based on your specific trip. If you are traveling with a toddler, swap out "cancellation risk" for "proximity to parks and grocery stores." If you are on a strict budget, instruct the AI to heavily weight "true all-in cost."
Pro tips
Include a text file of your constraint profile from Week 1\. Ask the AI to immediately eliminate any property that violates a hard constraint (e.g., "Must have dedicated parking") before it even builds the matrix.
Prerequisites
You must have gathered the listing texts, fee breakdowns, cancellation policies, and your rough daily itinerary, and saved them as text or PDF files ready for upload.
Required tools
You need an AI model capable of file uploads and document processing, specifically utilizing local files (such as Claude desktop integration).
Frequently asked questions
How do I save a listing as a local file?
The easiest way is to use the "Save as PDF" function in your browser's print dialog, or simply copy the text and paste it into a plain text document (.txt). Can the AI calculate the commute time accurately? The AI uses general knowledge of the city to estimate commute times based on the neighborhoods in your documents. It is highly accurate for general logistics, but you should verify exact transit schedules for the specific dates of your trip. What if none of the properties score well? That is a successful outcome. The matrix is doing its job by saving you from a bad booking. If all fail, it is a signal to revisit your budget ceiling or look at a completely different neighborhood.
Recommended follow-up prompts
With your lodging secured and your basecamp established, you are ready to build out the day-to-day timeline. Keep an eye out for next week's post, where we build the itinerary outward from where you wake up.
Tags and categories
Tags:
decision matrix, logistics, risk assessment, itinerary alignment Categories: Advanced Workflows, Strategic Planning
Citations
NOT APPLICABLE
Which of the three should you use?
The Beginner prompt is a tactical tool; it exists to solve one immediate problem—exposing the real price of a single property so you aren't fooled by drip pricing. It requires zero setup and is the fastest way to check your budget reality. The Intermediate prompt shifts from quantitative to qualitative, teaching you to read review patterns like an auditor to expose structural flaws a property manager is trying to hide. The Advanced prompt pulls everything together into a strategic workflow. Instead of looking at properties in a vacuum, it forces you to score them against your actual life—your itinerary, your budget constraints, and your tolerance for cancellation risk. You should start with the Beginner prompt to filter out overpriced options, use the Intermediate to vet the survivors, and run the Advanced matrix to make your final, defensible booking decision.
Two notes on Gemini's post this week, both about what it produced rather than what it argued.
The prompt breakdowns are formatted differently from the other posts. The site's "Prompt Breakdown — How A.I. Reads the Prompt" format is running text — a quoted fragment, then a colon, then the explanation — which renders as the two-column "teaching ledger" you see in the Claude and ChatGPT posts. Gemini instead wrote each quoted fragment as its own ### subheading with the explanation beneath it, so its breakdown renders as headings-and-paragraphs. The content is all there; only the presentation is downgraded. The judge (ChatGPT this week) cited this and Gemini's bolded section headings under Template Compliance, where Gemini scored 5 of 10 — but Gemini placed third on the strength of its content, by a wide margin, so the formatting did not decide its finish.
A few escaped characters landed in the reader prompts. Gemini's Advanced prompt numbers its scoring criteria as 1\., 2\., 3\. — escaped periods — in the block you copy, four such escapes in all. They are harmless; every model reads past them. But the prompts are the product, so we flag them.
We have not corrected the post. Ketelsen.ai is an experiment in what these models actually produce from one identical brief, so the format each one chooses and the stray marks it leaves are part of the finding. Reformatting Gemini's breakdown to match the others, or cleaning the escapes, or telling it in advance how to format, would quietly delete exactly the differences this series exists to surface.
TAGS:
