Skip to main content
Note! These pages are generated automatically. The content may be incomplete or wrong, screenshots and videos especially. See About these help pages. We would really like your input: reach us through "Did you find what you were looking for?" at the bottom of the page, the chat at bottom right, or support@placepoint.no - we answer as fast as we can!

Examples of using Placepoint AI

Our users made the reports below with Placepoint AI, from a single prompt each. They range from a lookup on one property to analyses of entire streets and portfolios. If you want to start with something short, go straight to the example prompts at the bottom of the page.

Area analysis: a whole street in one prompt​

The prompt "Check the whole of John Colletts allé, where can you build new, make a report" produced a 15-page report that goes through all 72 land properties in the street, with a map and an assessment for each address.

The report found that the street is in practice zoned in three different ways, and that four addresses in the middle of the street are not covered by the building ban that stops all the neighbours. That exception is not in any map database. Claude found it with Placepoint AI by reading the planning documents and the decision on the extended building ban. The requirements are in the provisions, not in the colours on the zoning map. The method section lists which planning documents were actually read, and separates known gaps in the data from what was not analysed. The same plans are in the zoning map layers in Placepoint Fusion.

Front pageSummaryZoning map
Front page of the report on John Colletts alléThe summary: the street is three streets, with key figures per zoneZoning map extract with land use areas over the garden city

Download the full report: John Colletts allé - zoning analysis (PDF, 15 pages).

More reports and analyses​

Claude made the reports below with Placepoint AI, from one prompt each. The sources and the deliverable change from report to report, while the prompts are built the same way: geography, variables, level of detail and format.

A gym is considering Lørenvangen 22 in Oslo. The prompt asks for two catchment areas, ten minutes on foot and ten minutes by bike. For each area it wants population, age, households, daytime population, and income and wealth per grunnkrets (basic statistical unit), plus the competitors within the area. Everything is to be delivered as a PDF with maps. The answer is Lørenvangen 22 - demographics and catchment areas (PDF, 14 pages). The report calculates both isochrones, adds up the SSB figures per area and compares them with the Oslo average. A separate section says what the data does not cover. The full prompt, and how to do the same by hand in the map layers, are on Demographic analysis.

Here are six more, built the same way:

  • Sustainability report for a head office: A 25-page report on a well-known head office in Oslo, with matrikkel, BREEAM certification, ownership structure, all 25 tenants with a company dossier, registered ownership history, maps and a travel time isochrone. Prompt: "Make a sustainability report for [address] with BREEAM, energy label, ownership structure and tenants."
  • Due diligence with clear coverage: An 11-page due diligence report with legal, technical and financial screening. The report itself marks what is covered and what is not, so you can see where your own review has to start. Prompt: "Make a due diligence report for [address], and mark clearly what is not covered."
  • Climate risk and credit assessment: A climate risk report as a PDF magazine with physical risk, transition risk and biodiversity, built on the same sources as the risk map layers. A credit report for a bank followed the same recipe. Prompt: "Make a climate risk report for [address] with physical risk, transition risk and biodiversity."
  • Buyer assessment: From a completely neutral prompt, Claude concluded that the risk was high, because the buying company was a party to an ongoing court case. The prompt did not mention court cases. Claude found the case because it read the whole company picture. Prompt: "Assess organisation number [orgnr] as the buyer of the property [matrikkel number]."
  • Land registers for a whole facility: Claude went through the land registers for the whole facility. It found all the matrikkel numbers itself, retrieved the extracts together and summarised the encumbrances that could block development. Prompt: "Retrieve the land register extract for all the matrikkel numbers in [facility] and summarise the encumbrances."
  • Lead list for a roofing contractor: Claude found 1,517 properties based on year of construction and building type, linked them to the owning company and assessed the roofs on aerial photos. It is the same kind of selection that Filter gives you in Placepoint Fusion. Prompt: "Find buildings in [kommune] from before 1990 with large flat roofs, linked to the owning company, as a spreadsheet."

More prompts of the same kind, sorted by topic, are under Short example prompts.

Plot subdivision: which plots are large enough​

The question "Which plots in Bergsalléen in Oslo are large enough to be subdivided?" gives an answer in two parts: the rules that govern subdivision right now (area requirement per dwelling unit, %-BYA and the temporary ban with exceptions), and a table of the properties in the street that meet the area requirement, with matrikkel number and plot area. The answer also says what you must check with Plan- og bygningsetaten (the city planning and building authority) before you send a subdivision application.

Answer to the plot subdivision question for Bergsalléen, with rules and a table of plots that are large enough

This answer is from the Agent in Placepoint Fusion, which uses the same engine as Placepoint AI in Claude. The questions below work in both places:

  • "Check the whole of John Colletts allé, where can you build new, make a report."
  • "Which plots in Bergsalléen in Oslo are large enough to be subdivided?"
  • "Has any planning work been announced near Bergsalléen?"

How to get the most out of Placepoint AI​

We have already fine-tuned Placepoint AI itself: which tools Claude picks, how the lookups are narrowed, and how much data comes back in each answer. You do not set that yourself. What you control is the context Claude works in and the way you ask. That decides both the quality of the answer and how much of your Claude quota you use.

Keep conversations short​

Claude remembers everything that has been said and retrieved in a conversation, and uses it as working memory when it answers. The memory is large, but it gets worse the fuller it is. Anthropic writes in its documentation of the context window that precision and recall fall as the conversation grows, and calls this "context rot". Their advice is to keep the context small and relevant, with as little information as possible but enough to give the answer you want (Effective context engineering for AI agents). This is a property of Claude, not of Placepoint AI.

For you, this means two things. Start a new conversation when you change property, client or topic, even if the old one still answers. And give Claude good context from the start, such as instructions and standing documents in a project, instead of building it up through a long conversation. If a long conversation becomes unclear or slow, ask Claude to summarise what has been settled, and paste the summary into a new conversation.

Turn on memory, and Claude remembers you between conversations​

Placepoint AI remembers nothing itself. The client you use is what remembers, and that is where you control what gets stored. In Claude it is called Memory: Claude stores who you are and what you work on, and can search earlier conversations when you ask about something you have discussed before. Each project has its own memory, and Cowork running locally on your computer does not use it.

Whether memory is on from the start depends on the plan:

  • Free, Pro and Max: memory is on from the start in Claude Desktop, on the web and in the mobile app.
  • Team and Enterprise: an Owner must first turn it on for the organisation. Then each user must turn it on themselves under Settings, Memory, with the Generate memory from chats toggle. Until both are done, Claude remembers nothing from the previous conversation.

Under Settings, Memory you also see everything Claude has stored, and can correct or delete it. See Use Claude's chat search and memory to build on previous context.

If you use Placepoint AI from another client, the same principle applies, with different settings:

  • Microsoft 365 Copilot: memory is on from the start. You control it under Settings and more, Chat settings, Personalization, Saved memories. See Manage Copilot Memory in Microsoft Copilot.
  • ChatGPT: the settings are under Settings, Personalization, Memory, with separate toggles for saved memories and for using earlier conversations. See Memory in ChatGPT.

Give Claude instructions once, not in every prompt​

The simplest and most effective step is to tell Claude who you are, what you work on and what the deliverable should look like. Then you do not have to repeat it in every conversation, and the answers hit the mark from the first attempt. This applies to all use of Claude, not just Placepoint AI.

  • Your own profile: click your initials at the bottom left, select Settings and fill in Instructions for Claude. The instructions apply to all your conversations. Enter your role, company, field, the language you want answers in and the form the deliverable should take. See Understanding Claude's personalization features.
  • Projects: gather work that belongs together in a project, for example one client, one acquisition or one portfolio, and put the instructions and standing documents there. All conversations in the project inherit them. See What are projects? and How can I create and manage projects?.
  • The whole organisation: on Team and Enterprise plans, the Owner and Primary Owner can set instructions that apply to everyone in the business, under Organization settings, Organization and access. They take precedence over the individual user's own instructions where the two say something different. See Set organization instructions.

Anthropic's own advice is in Prompt engineering best practices, and the technical reference they keep updated is Prompting best practices. Four of the pieces of advice matter most: say what Claude should do instead of what it should not do, give reasons for the requirements so the model understands the purpose, let it say that something is missing, and keep the instruction short. Longer instructions do not give better answers.

Example instructions for a business​

A law firm that uses Placepoint AI for factual groundwork, not for advice, can enter something like this for the whole organisation. Anthropic recommends the tags because they keep the different requirements apart, and because you can later change one block without touching the rest.

<rolle>
You provide factual groundwork for lawyers. We make the legal assessment
ourselves, so the answer should be verifiable rather than conclusive.
</rolle>

<kilder>
State the register and the date for every factual claim about a property, and
separate information from Grunnboken (the Land Registry) from matrikkel
information. Mark your own reasoning with "Inference:", so we can see what the
register says and what you derive.
If something is not in the register, write that it is not registered.
</kilder>

<nar_noe_mangler>
If you are missing context that materially changes the answer, ask one precise
question before you answer. Reasonable inferences are fine when they are
marked as such. Numbers, names, quotes and source references must come from
data you have retrieved.
</nar_noe_mangler>

<personopplysninger>
You do not show personal names or national identity numbers: Placepoint AI removes them.
If we need an identified title holder, we look it up in Placepoint Fusion,
where access requires a legitimate interest, or in Grunnboken.
</personopplysninger>

<leveranse>
Answer with plain text or a table that we paste into our own templates. Ask first
if the task looks like it needs a report, PDF or dashboard: that takes
longer, and we rarely use it. End with findings, uncertainties and what should
be investigated further.
</leveranse>

The instruction does three things. It says what Claude should do, not what it should avoid. It gives reasons for the requirements. And it gives the model a way out when the basis is thin, so it does not force an answer. These three moves matter most.

Replace the content with what applies to your industry. A property company will for example ask for portfolio figures per SPV, a bank for a traffic light per risk topic.

Standard instructions by industry​

The blocks below use the same three moves in each industry, and are split into three parts: who you are and what Claude should deliver, what to prioritise, and the form the deliverable should take. Each example is written as if it came from one business in the industry. Pick your industry, and replace whatever does not fit you.

<rolle>
We are a regional bank with business customers, and you provide factual
groundwork for collateral and credit assessment of the properties put up as
security. We make the credit decision ourselves, so the answer should be
verifiable rather than conclusive.
</rolle>

<prioriter>
Title holder and form of ownership first, then encumbrances from Grunnboken:
mortgages with amount and priority, easements, restrictions on disposal and
bans on subdivision. Then natural hazards for each property: flood,
landslide, quick clay and storm surge, with awareness maps separated from
hazard zones. Finally tenants with lease terms where the rental income
services the loan. State the register and the date for each piece of information.
</prioriter>

<leveranse>
A table with one row per property and one column per topic: encumbrance,
natural hazard, tenant, source, date. Write "not covered by the register"
where coverage is missing, rather than leaving the row out. End with the
three things we should clarify with the customer before the case goes to the
credit committee. No recommendation on the loan or the loan-to-value ratio.
</leveranse>

If several of you work in different ways, put the shared part at organisation level and the industry block in each project.

Ask for what you actually need​

  • One task per conversation. Start a new conversation when you change topic. A long conversation carries along everything retrieved earlier, and that costs both time and quota without making the answer better; see Keep conversations short.
  • Say what form you want. If you need three numbers, ask for three numbers. A report, table, map or dashboard costs more than a short answer. Reports are delivered as HTML by default because that is fastest; ask for PDF when you actually need to send the document on.
  • Narrow it down. One matrikkel number, one street, one kommune or one organisation number gives faster and more precise answers than "Oslo".
  • Reuse the recipe. When an analysis works well, ask Claude to write the prompt down as a recipe you can run on the next property or portfolio. Put it in the project instructions.
  • Pick the model to fit the task. Use the most advanced model to build the structure and the recipe, and a faster model to run that same recipe many times.

Ask for sources, and read the caveats​

  • Require a source and a date for every factual claim. Claude will readily state which register the information comes from and when it was retrieved, and link to the map layer or the planning document. Make it a standing requirement, so the answer is verifiable.
  • Ask Claude to separate what the register says from what it infers. That is the most important distinction in factual groundwork.
  • Definitions vary between kommuner. "Undeveloped" and "sentrumsformål" (central area purpose) do not mean the same thing in every kommune. Ask Claude to flag which definition it has used when the answer covers more than one kommune.
  • If data is missing, the answer should say so. Placepoint AI tells you when there are no planning data for a property, instead of describing it as unzoned.
  • Claude can be wrong. The results are a starting point, not a definitive answer, and must be verified against official sources before they are used for decisions.

If an administrator has added the connector for the whole business, each user must still log in with their own Placepoint account the first time. Access follows the user, not the URL. The steps for connecting are on Placepoint AI.

Short example prompts​

Placepoint AI takes both short questions and complex tasks: one lookup on an address, or a whole street with a report at the end. The examples below are easy to write and still use a lot of Placepoint at once: the registers, the map layers and the analyses in the same answer. Feel free to start with "Use Placepoint AI", write in the language you use yourself, and replace the square brackets with a real address, a matrikkel number or an organisation number. The prompts are grouped by topic; most of them come from what our users have actually asked.

Property and land register​

  • "Retrieve the land register extract for [matrikkel number] and summarise which encumbrances are on the property."
  • "Find all the matrikkel numbers that belong to [facility], retrieve a land register extract for each of them and summarise the encumbrances that could block development."
  • "Make an energy report for [address] with energy grade, heating grade, date and validity for each building. List the buildings without a valid energy certificate separately."
  • "Who owns property in [kommune] with an average energy label of A?"
  • "What does BRA (usable floor area) mean, and how large is the BRA at [address]?"
  • "Make an investment report for [matrikkel number]: which tenants are there, what can the property be developed into, and where are you missing data?"
  • "Make a sustainability report for [address] with BREEAM certification, energy label, ownership structure and all tenants, and show 15 minutes' walking distance on the map."

The energy labels come from Enova and apply to certificates that are less than ten years old. A building without a valid certificate has no energy label, and Claude tells you so.

Zoning and plans​

  • "Which zoning plans and municipal plans apply to the property [matrikkel number]?"
  • "Retrieve the provisions and the planning description for the plan that applies to [matrikkel number], and summarise what is allowed in terms of land use ratio, heights and purpose."
  • "Which plans apply to [address], including the planning programme and advisory plans, and what do they say about land use ratio?"
  • "Check the whole of [street]: where can you build new? Make a report."
  • "Which plots in [street] are large enough to be subdivided?"
  • "Has any planning work been announced near [address]?"
  • "Find undeveloped properties in [kommune] that are designated for residential or central area purposes in the municipal plan."
  • "This is a new residential project on [matrikkel number]. What do the planning provisions say about housing mix, parking and noise?"
  • "Should the building on [matrikkel number] be demolished or extended? Go through zoning, risk and ownership, and say what you are sure of and what is uncertain."

What counts as undeveloped, and which land use purposes count as sentrumsformål (central area purpose), varies between kommuner. Ask Claude to say which definition the answer is based on.

Risk and climate​

  • "Check whether the property [matrikkel number] is in a flood or landslide exposed area, and show it on a map."
  • "Make a risk report for [address]."
  • "Which objections could Statsforvalteren (the county governor) raise against development on [matrikkel number]? Go through landslides, flood, storm surge, surface water, cultural heritage, nature conservation, biodiversity and contaminated ground, and make a report."
  • "Make a risk assessment of organisation number [orgnr] as the buyer of the property [matrikkel number]."

The last prompt shows what happens when Claude reads the whole company picture: an ongoing court case involving the buyer was included in the assessment without the prompt asking for it. The report Natural values and objections goes through the same sources as the objections prompt.

Company and tenants​

  • "Which properties does [company] rent in [kommune]?"
  • "Where is [company] based today, who is the landlord, and where were they before?"
  • "Show the largest tenants of [company] in Bjørvika."
  • "Make a tenant overview for [address] that shows how long each tenant has been there."
  • "Make a tenant report for [address] as a PDF, with one to three pages per company for the ten largest by number of employees: lease period, a short risk assessment and an ownership tree. The rest in a summary table."
  • "How many employees work at [address], and how many do the companies there have in total?"
  • "Which tenants are in BREEAM-certified buildings in [kommune]?"
  • "Which of the tenants of [group] are in buildings that are not offices?"
  • "Make a ranked list of commercial properties in [kommune] with only one tenant, where the tenant has weak finances."
  • "Make a list of IT companies in Bergen with more than 20 employees that sit in office buildings."
  • "Show all the [chain] stores in [kommune] with address and number of employees per location."
  • "Which companies with more than 20 employees own property in [kommune], and how much area do they own?"
  • "Which companies have obtained title to property in [kommune] in the past two years?"
  • "Which public bodies own property in grunnkrets?"
  • "Which companies being wound up own warehouse or logistics property?"
  • "Find the SPVs in [kommune] and who is behind them."

The number of employees exists at three levels: at the address, in the company and in the group. If the prompt is unclear, Claude asks which number you mean instead of guessing. A company being wound up is still listed as active in the registers, so ask for it explicitly when that is what you are looking for.

Portfolios and selections​

  • "How many active properties are there in [kommune]?"
  • "Make a portfolio report on all the properties [company] (organisation number [orgnr]) owns in [place/area]."
  • "Find all new build projects in [area] with a commencement permit, outline permission or temporary occupancy permit and at least 20 dwelling units."
  • "Find all the [chain] stores in Norway, link each one to the matrikkel and the energy label, and make a spreadsheet of the ones with energy label C or better."
  • "Find properties over 10,000 m² in [county] with the land use purpose commercial development in the municipal plan, less than 2 km from a transformer station and with no housing nearby."
  • "Make a full portfolio overview for organisation number [orgnr]: all properties with geography and a map of the main cluster, ownership structure from the ultimate owner down to the SPVs with flags for deleted companies and companies with negative equity, all tenants per building and how long each of them has been registered, and an ESG profile per building."

The new build projects come from the building status in the matrikkel, not from a project register, so check the count against your own sources. The transformer stations are in the map layers from Norges vassdrags- og energidirektorat (NVE, the Norwegian Water Resources and Energy Directorate), but free grid capacity is not included.

Location and travel time​

  • "Find the best places to set up a new grocery store near [place]."
  • "Which properties are within 300 metres of Jernbanetorget, and who owns them?"
  • "Find the ten best premises for a distribution centre for a company that has to make deliveries physically in [area], ranked by driving time."
  • "Which commercial buildings are within 25 minutes by public transport from [station]?"
  • "Find undeveloped plots within 30 minutes of [address] that are zoned for small houses or terraced houses, and check flood, landslide and quick clay on the best ones."
  • "Make isochrones for 10, 20 and 30 minutes by car around [address], and show them on a map."

Travel times are calculated with the same engine as Analysis in Placepoint Fusion.

Maps, aerial photos and export​

  • "Show all historical aerial photos around the property [matrikkel number], set up neatly with metadata."
  • "Show the density of detached houses in [kommune] on a map."
  • "Show the cultural heritage sites in [area] on a map."
  • "Export all properties with tenants in [kommune] to a spreadsheet."

Prompts that ask for images, maps or files only work in Claude Desktop, Claude Code and Cowork. The Agent makes neither images nor files, so there the answer comes as text and map links. The cultural heritage data comes from Riksantikvaren (the Directorate for Cultural Heritage).

Placepoint Dataset MCP​

Placepoint Dataset MCP works with the datasets in your projects, and only sees the projects your Placepoint account has access to. Before Claude writes anything, it says which project this concerns; datasets can have the same name in several projects. The result is a map layer in the project the next time you open Placepoint Fusion.

Create and upload​

  • "Make a dataset in the project [project] with all commercial properties over 5,000 m² in [kommune], with owner and year of construction."
  • "Retrieve the properties [company] owns, add them as a dataset in [project] and show the portfolio on the map."
  • "Make a dataset with all the grunnkrets areas in [kommune], so I can colour them afterwards."
  • "Make a dataset with all the postcodes in [city], assign them to our offices, and make office a dropdown list in the table."
  • "Upload our tenant list, link the rows to the matrikkel on address and show which ones got no match."

You cannot paste a file into the conversation. Claude creates an upload link, and you put the file there; in Claude Code, Claude does it itself. Rows Claude has found itself, such as the properties of a company, are added directly.

Link and enrich​

  • "Add kommune and grunnkrets (basic statistical unit) to each row in [dataset], and count rows per kommune."
  • "How many of the rows in [dataset] are in each bydel (city district) in Oslo?"
  • "Which rows in [dataset] got no match in the matrikkel, and why?"
  • "Add a column Status to [dataset] with the options New, In dialogue and Closed."
  • "Update [dataset] from this file, keyed on organisation number. Show what changes before you write."

Monitor​

  • "Which properties in [dataset] have a new title holder or a new mortgage in the past year?"
  • "Which properties in [dataset] are in a flood zone or a landslide awareness area?"
  • "Which of the companies in [dataset] are being wound up or are bankrupt?"

Route and travel time​

  • "Find the shortest car route from [address] that visits all the points in [dataset]."
  • "Make a 20-minute isochrone by car from [address], and save it as a dataset in [project]."
  • "What is the driving time from each office in [dataset] to each property in [dataset]?"

Style and sharing​

  • "Set the colour on [dataset] by year of construction, and hide the columns we do not use."
  • "Use our logo as the symbol for the points in [dataset]."
  • "Share the project [project] with the whole team as readers."

A symbol must be square; an oblong logo is stretched on the map. Claude shares at the lowest access level that is enough, and asks before it shares.

From list to map: one example​

Five prompts turn a customer list in Excel into a shared map layer:

  1. "I have an Excel file with our customers, one per row, with address. Make a dataset from it in the project [project]."
  2. "Link the rows to the matrikkel on address. Which ones got no match?"
  3. "Add kommune and bydel as columns, and count customers per bydel."
  4. "Set the colour by customer type, and hide the columns with internal notes."
  5. "Share the project with the team as readers."

You can do the same step by step in the panel; see Datasets. The difference is that Claude can also answer questions about the list based on the registers, for example which of the customers have sold, and who owns the neighbouring plot.

Where the prompts have a fixed page​

Several of the prompts have a fixed page in Placepoint Fusion. The tenants for one property are in Units and tenants, and for a whole portfolio in the tenant report. You make selections on industry, employees and building type in Filter. Driving time and proximity are in Analysis. The reports and maps are built on the same sources as the rest of Placepoint Fusion: see Data and sources and Placepoint in numbers.

Longer prompts for whole tasks​

Each of the prompts below solves a whole task and ends in a memo, a spreadsheet or a map layer. They are longer than the short example prompts, and they ask for the same things every time: a source and a link on every point, clear marking of estimates, and a note when something is not in the data. Replace the square brackets with a real address, a matrikkel number or an organisation number, and start a new conversation for each task (see Keep conversations short). The first time, run each of them on a plot or a company you know well, and compare the result with what you found yourselves; that shows you what hits the mark.

You can also save each of the prompts as a skill in Claude: a short instruction Claude brings up by itself when the task comes up, so you do not have to paste in the prompt every time. Anthropic explains what a skill is, and how to add one, in What are Skills? and Use Skills in Claude.

Plot assessment in five minutes​

The prompt answers the four questions you ask about every plot: who owns it, what it is zoned for, what is in the ground, and who the neighbours are. The basis is the same as in the property panel and Risk.

Use Placepoint AI. Do a plot assessment for [address or matrikkel number].

Cover:

  • The property itself: matrikkel number, plot area, buildings and title holder.
  • The owner and the ownership chain up to the ultimate owner, with the latest accounts for each company.
  • The applicable zoning and municipal plan, and whether the purpose is residential. Read the provisions, not just the name of the plan.
  • Risk: flood, landslide, quick clay, cultural heritage and contaminated ground.
  • The neighbouring properties, with owner and purpose.

State the source and a link on every point. End with a one-page memo: what speaks for, what speaks against, and what we have to check before we go further.

Variants:

  • "... compare [address A] and [address B] in the same answer, with the same structure."
  • "... include transactions and comparable sales in the area in the past [number] years."
  • "... deliver as a PDF I can attach to the decision memo."
  • "... include travel time to public transport, population and schools in the local area."

The answer is based on the applicable zoning plan and municipal plan, the title holder in the land register and the risk map layers. Sales pace and unit mix for new residential projects are not included; we do not have data on them.

Site search and prospecting​

The prompt finds development candidates based on criteria you set, instead of one lookup at a time. You can make the same selection in Filter; the difference is that Claude also follows the ownership chain and delivers the spreadsheet.

Use Placepoint AI. Find undeveloped plots within [30] minutes of [Oslo S] by car or public transport, where the zoning purpose or municipal plan purpose is residential, and the plot is over [area] m².

For each match: matrikkel number, address, plot area, the applicable land use purpose and which plan it comes from, the title holder and the ownership chain up to the ultimate owner.

Deliver as Excel sorted by plot area. Mark clearly what is an estimate, and flag matches where the land use purpose is uncertain or defined differently in the kommune.

Variants:

  • "... limit to [kommune] and sort by travel time to the nearest railway station."
  • "... only include plots where the title holder is a company."
  • "... only include purposes that allow small houses and terraced houses, not blocks of flats."
  • "... run our own plots [list] through the same check, so we see them side by side with the candidates."
  • "... extend to [neighbouring kommune] and sort all matches together."
  • "... add the ten best as a dataset in [project], so the team can see them on the map."

Claude reads "undeveloped" as meaning there is no completion certificate in the building cases in recent years. That is an indication, not proof. Whether the plan allows small houses or only blocks of flats is in the provisions, not in the land use purpose. Use the list to choose where to look more closely, not as a definitive answer.

Area overview around a plot​

The prompt puts together the picture around a project or a plot: which plans apply, what has been sold, who owns the properties around it, and what could stop the project. It is the same query that the area analysis is built on.

Use Placepoint AI. Make an area overview for [address or matrikkel number].

Cover:

  • The applicable zoning and municipal plan. Read the provisions, not just the name of the plan.
  • Transactions and comparable sales in the area in the past [number] years.
  • Ownership of the neighbouring properties, with company and ownership chain.
  • Risk: flood, landslide, quick clay and contaminated ground.
  • Location: travel time to public transport, population and services in the local area.

State the source and a link on every point. Finally, summarise in plain language what speaks for and what speaks against.

Variants:

  • "... compare [address A] and [address B] in the same answer, with the same structure."
  • "... deliver as a PDF I can attach to the decision memo."
  • "... include historical aerial photos that show how the area has developed."

The transactions come from the property sales in the land register, and the travel times from the same calculations that Analysis uses.

Portfolio overview​

The prompt gathers a company's portfolio, including the properties it owns through subsidiaries and SPVs. This is the query that most often uncovers properties people did not know were in the portfolio, because they sit several levels down. The Ownership tab in the company panel and the portfolio report show the same for one company at a time.

Use Placepoint AI. Give me a full overview of the properties of [organisation number], including those owned indirectly through subsidiaries and SPV chains.

Per property: matrikkel number, address, plot area, buildings and UFA, the company that owns it, and the applicable planning and zoning status.

Include a map of the main cluster, and the ownership structure from the ultimate owner down to the SPVs. Flag companies that have been deleted or have negative equity.

Deliver as Excel with one row per property, and a short summary at the top.

What has happened since last time:

Use Placepoint AI. For the properties of [organisation number], including those owned indirectly: what has changed since [date]? Look for changes of ownership, new building cases, new planning decisions and changes in the company structure. Deliver as a short list, newest first, with a source per point.

Variant:

  • "... add a risk sweep per property: flood, landslide, quick clay, cultural heritage and contaminated ground. A separate column per topic, with source."

The areas are BRA from the matrikkel, and the accounting figures come from Brønnøysundregistrene, as in the Accounts tab.

Risk sweep​

The prompt checks one plot, a list or a project against the risk map layers and reads what the plan says about the findings. Without it, you have to look things up in five map portals. Risk in the property panel does the same for one property at a time, and the risk report describes the content.

Use Placepoint AI. Run a risk sweep for [address, matrikkel number or list].

Per property:

  • Flood, landslide, quick clay and surface water, with zone and source.
  • Cultural heritage and consideration zones in the applicable plan.
  • Registered contaminated ground.
  • What the planning provisions require in terms of studies or measures on these points.

Deliver as a table with one row per property and one column per topic. Mark red where there is a registered finding, yellow where the map layer is uncertain or missing, and green where it has been checked and nothing was found. State the source in every cell.

Projects as a map layer:

Use Placepoint AI. Find [three] projects on the website of [company] that are completed or under way. Do a risk sweep on each of them and save the result as a dataset in [project], so the whole team can see it as a map layer in Placepoint Fusion.

Yellow means "not clarified", never "probably fine". An empty map layer means the hazard has not been mapped, not that it does not exist. If you get green across the board in an area you know has marine clay, ask Claude whether the quick clay map layer actually covers the area. Quick clay hazard outside mapped zones requires a ground investigation. Surface water is read from the drainage lines, cultural heritage and consideration zones from the zoning map layers, and contamination from Contaminated land.

Acquisition list in three steps​

The prompts build and maintain a list of candidates you follow over time, so that assessed plots do not disappear into old memos. Step three puts the list in a project as a dataset, so the whole team sees the history on the map and nobody assesses the same plot twice without knowing it. The acquisition report describes the content of the list.

Step 1, build the list:

Use Placepoint AI. Set up an acquisition list for [area or criteria].

Per candidate: matrikkel number, address, plot area, land use purpose, title holder and ownership chain, the last known transaction with price and date, and a short status line.

Add columns we fill in ourselves: our assessment, bid price, status and next step.

Deliver as Excel.

Step 2, check what has happened:

Use Placepoint AI. Here is our acquisition list [paste or upload]. Check each property against the registers and say what has changed since [date]: change of ownership, a new building case, a new planning decision, or a change in the company that owns the property. Deliver as a short list of what has actually changed. Do not repeat rows without changes.

Step 3, put the list on the map:

Use Placepoint AI. Upload this list as a dataset in [project], linked on matrikkel number, so the whole team can see the candidates as a map layer in Placepoint Fusion. Name the dataset [name].

Step 3 uses Placepoint Dataset MCP. You cannot paste the list into the conversation: Claude creates an upload link, and you put the file there.

Counterparty and ownership chain​

The prompts find out who actually owns a plot before you get in touch, and assess the counterparty in a transaction: company, group and accounts. The company panel and the ownership structure report show the same in Placepoint Fusion.

Who owns it:

Use Placepoint AI. Who owns [address or matrikkel number]? Follow the ownership chain from the title holder up to the ultimate owner, with the ownership share at each level. Include the board and the chief executive of each company, and which other properties the same owner has in [kommune or area].

State the source for each level. Say so if a level in the chain is a private person, without giving the name.

Assess the counterparty:

Use Placepoint AI. Make a buyer assessment of [company or organisation number].

Cover: group structure and ultimate owner, the last three years of accounts with revenue, net result and equity, the property portfolio with number of properties and UFA, recent transactions, and roles that recur in other companies.

End with five points on what I should know before I go into a negotiation. State the source for every figure.

Private persons are anonymised in Placepoint AI: you find out that a level in the chain is a private person, and how large the share is, but not who it is (see Personal data is not available). If you need the name, the land register extract is the source. The answer is working material, not a credit assessment, and Claude says so itself.

Sist oppdatert
Note! These pages are generated automatically. The content may be incomplete or wrong, screenshots and videos especially. See About these help pages. We would really like your input: reach us through "Did you find what you were looking for?" at the bottom of the page, the chat at bottom right, or support@placepoint.no - we answer as fast as we can!