EXAMPLE REPORT · PLACEHOLDER PROFILE

Small project-development office · 8 employees · Excel stack · AI beginner

AI READINESS REPORT

Maturity analysis

Maturity Exploring — First touchpoints exist, but no systematic use.

32/100

Tool stack

28/100

Processes

18/100

AI experience

45/100

Team

Summary

Your company is at the start of AI adoption — and at your size that is the norm.

You run a pragmatic tool stack (Excel, Outlook, Teams, DATEV), but the underlying processes are not yet digitised enough for AI to plug in meaningfully.

Your biggest lever is not buying tools but covering three recurring tasks — investor reporting, site manager meetings and authority correspondence — with Claude or Microsoft Copilot.

Within four weeks, saving 6-8 hours per week is realistic.

About this profile: Owner-led project-development office focused on residential construction in NRW. Microsoft 365 + DATEV + Excel. First ChatGPT attempts in the team, but no systematic use.

Score in detail · Where to start

Tool stack maturity

32/100

You only use Microsoft 365 and DATEV — solid, but without project-management or CRM software. Pragmatic at your size, but it limits the AI lever: without structured data in a CRM or PM tool, AI always has to be fed manually.

Improvement steps

→Stay with Microsoft 365 — but activate Copilot for 2-3 power users in the team (€30/user/month).
→Build Excel sheets for project costing and status to a clean structure so AI can read them.
→Only invest in a CRM (e.g. Pipedrive from €14/user/month) after 3-6 months of AI experience.

Process digitisation

28/100

Meetings are minuted manually in Word, status reports to shareholders are assembled ad hoc. Both are perfect candidates for the first AI pilot — recurring, text-heavy, not regulatory-critical.

Improvement steps

→Select a single pilot process: the weekly status report to shareholders.
→Standardise the existing template before introducing the tool — otherwise you feed AI inconsistencies.
→Run the pilot for 2 weeks, then measure honestly: how much time saved, what errors did AI make.

AI experience

18/100

No systematic AI use yet. Earlier attempts were not followed up — that is typical and not a problem if you now start in a focused way.

Improvement steps

→Start with exactly one use case — not three in parallel.
→Claude Pro (€20/month) or ChatGPT Plus (€22/month) for management — deliberately where the most text gets written.
→Weekly 30-min learning slot: one team member shows what they did with AI.

Team adoption

45/100

Middle-of-the-road adoption experience — earlier tool rollouts were mixed. That is actually better than "enthusiastic but chaotic": you have learnt that introducing tools needs planning.

Improvement steps

→Name a clear champion (typically you or a tech-savvy junior) — they try things first, the team follows.
→Set the data privacy ground rule early: which data may go into which tool (default: no tenant/buyer data in consumer AI).
→After 4 weeks of pilot: an honest retro with the team — what worked, what did not.

Top 3 immediate actions · What to do this week

Draft the shareholder status report with Claude

What exactly

Gather bullet points from the 3-5 active projects in a simple list and have Claude write the weekly report in your usual style. You correct — Claude tends to nail the tone on the second iteration.

How to implement

Open claude.ai/new, upload the last good example update as a template, then send only the current week's bullet points.

Time investment

30 min setup + 10 min per week

Expected result

Saves 90-120 minutes per week at the same quality.

Minute the site manager jour fixe with Microsoft Teams Premium

What exactly

Teams Premium (€10/user/month extra) transcribes the meeting automatically. You hand Claude the transcript and get a protocol with action items.

How to implement

Activate Teams Premium for the two people who are always in the jour fixe. After the meeting: export transcript → Claude → finished protocol.

Time investment

15 min per meeting instead of 60

Expected result

Saves 45 minutes per site meeting; action items get documented consistently.

Pre-draft authority correspondence with ChatGPT

What exactly

Applications, statements and queries to the building authority are formulaic. ChatGPT hits the official register very reliably.

How to implement

Send case number, facts and request in bullet points to ChatGPT. First version as a basis, personally cut down to one page.

Time investment

5-10 min instead of 30 per letter

Expected result

Saves 20-30 minutes per letter and reduces wording mishaps.

Quick wins · Implementable today or this week

Sanity-check Excel costings with Copilot

You give Copilot your project costing and ask for anomalies, missing line items, unusual ranges.

How: In Excel: Copilot icon → "Check this costing for plausibility and name 3 possible gaps".

Effort: 5 min per costing· Impact: Finds 1-2 forgotten line items per costing on average.

Skim a development plan

Instead of 60 minutes of skim-reading, Claude extracts the relevant provisions in 10 minutes.

How: Paste the PDF text into Claude with the prompt: "Extract all provisions a residential developer needs to know."

Effort: 10 min per development plan· Impact: Saves 45-60 min and reduces the risk of missing critical provisions.

Buyer update letter during the build phase

Every 6-8 weeks a buyer update goes out — a perfect recurring writing task.

How: Send status bullets + 2-3 site photos to Claude. Prompt: "Write a friendly buyer update letter in the style of the attachment."

Effort: 15 min instead of 60· Impact: Saves 45 min per update with better continuity.

Site risk quick-check

For every new site idea, 10 minutes of risk structuring with Claude — legal, technical, commercial, ESG.

How: Send address + intended use to Claude. Prompt: "List 8-10 typical risks by category and estimate likelihood."

Effort: 10 min per site· Impact: A structured first check before LOI — prevents expensive surprises.

Structure a defect list after a site walk

Voice notes or bullet points become a sorted defect list with trades and deadlines.

How: Send a voice memo of the walk to Claude. Prompt: "Sort by trade, formulate concise defect descriptions, suggest deadlines."

Effort: 5-10 min per walk· Impact: Saves 30-45 min and makes the list more legally robust.

Your tool stack in detail · What stays, what goes

Microsoft 365

Augment

Privacy: safe

A solid base for your size. Outlook, Teams, Word, Excel cover 80 % of your communication. Weakness: without Copilot the AI capabilities go unused.

AI integration — how concretely

Activate Microsoft 365 Copilot for 2-3 power users (€30/user/month extra). Stays in the EU tenant, GDPR-compliant under your existing M365 agreement. Quick impact in Outlook (email replies) and Word (investor updates).

Privacy in detail

Safe: M365 with EU Data Boundary, DPA already part of the master agreement, data does not leave the tenant.

Alternatives

Currently no better alternative for your size.

DATEV

Keep

Privacy: safe

The standard for German accounting. Your tax advisor runs on DATEV — switching makes no sense. The UI is old-school but technically robust.

AI integration — how concretely

AI directly inside DATEV is not relevant today. But: export BWA reports and evaluations as CSV → comment with Copilot in Excel → send as investor explanation.

Privacy in detail

Safe: DATEV data centre Nuremberg, ISO 27001, highest GDPR compliance in Germany.

Alternatives

No better alternative in the German Mittelstand.

Microsoft Excel

Augment

Privacy: safe

Currently carries the entire project costing — pragmatic at your size, risky as you grow. Your biggest single point of failure.

AI integration — how concretely

Copilot in Excel makes plausibility checks routine. "Stress test" tables before every send.

Privacy in detail

Safe: locally on your machine or in the M365 tenant.

Alternatives

Specialised cost-estimating software (ORCA, California) only worthwhile from 6+ parallel projects.

Microsoft Outlook

Augment

Privacy: safe

Pure email management. With Copilot it becomes an efficiency tool for investor communication and authority correspondence.

AI integration — how concretely

Copilot in Outlook summarises long threads and drafts replies in the style of your past emails. Setup 5 minutes.

Privacy in detail

Safe: M365 standard.

Alternatives

None.

Recommended AI tools · Which ones for you, with privacy rating

Microsoft 365 Copilot

Privacy: safe

Best for

Directly inside Outlook, Word, Excel, Teams. The most straightforward entry point for your setup — no data migration, no new UI.

Privacy

Safe: EU Data Boundary, DPA in the M365 contract, no training with your data.

Pricing

€30/user/month on top of your existing M365 licence.

How to start

Pick 2-3 power users, activate licences in the M365 admin. First impact in Outlook within the first hour.

Claude Pro (Anthropic)

Privacy: conditional

Best for

Longer texts, investor updates, authority letters, site meeting protocols. Reliably hits the formal German business register.

Privacy

Conditional: Pro plan with no training on your data, US hosting. For sensitive contracts: anonymise data or move to the Business plan with zero-retention.

Pricing

€20/month per user.

How to start

claude.ai/pricing → sign up for Pro. First application: the next investor status report.

ChatGPT Plus (OpenAI)

Privacy: conditional

Best for

Quick research, brainstorming, alternative wording. Strong on short texts and image analysis.

Privacy

Conditional: Plus plan without training on your data (activate opt-out), US hosting.

Pricing

€22/month per user.

How to start

chat.openai.com → Plus plan. Start with one employee, then expand.

Microsoft Teams Premium

Privacy: safe

Best for

Automatic transcription of site and investor calls — the basis for AI-generated protocols.

Privacy

Safe: M365 EU Data Boundary.

Pricing

€10/user/month extra.

How to start

Activate in the M365 admin — only for the 2 people who are almost always in the meeting.

Your 4-week plan

WEEK 1

Status report pilot

Sign up for Claude Pro for management. Upload the last good status report as a template. First test with this week's bullet points. Share output with the team.

Tools: Claude Pro

Outcome: The first AI-generated status report has been sent.

WEEK 2

Site manager protocol

Activate Teams Premium for 2 people. Record the next site meeting, send transcript to Claude for protocol generation.

Tools: Microsoft Teams Premium + Claude

Outcome: The first AI-generated site protocol with action items is in circulation.

WEEK 3

Authority correspondence

Pre-draft the next three authority letters with ChatGPT. Compare one with the architect (human vs AI).

Tools: ChatGPT Plus

Outcome: Tone is on point. Time savings documented.

WEEK 4

Review and scale

Team retro: what worked? Measure time saved per use case. Finalise the privacy rule. Plan the next 90 days.

Tools: Excel + Notion or Word

Outcome: A clear plan for which two processes come next.

What to watch out for

Using AI tools on consumer plans (ChatGPT Free) for business data — data is used for training.
Starting three use cases at once — adoption fails. One must actually work first.
Buying tools without a clear use case — classic Mittelstand trap.
EU AI Act AI-literacy duty: every employee working with AI must be trained (since 02.02.2025).

Get your own report in fifteen minutes.

The report follows the structure shown above, but is built from your own answers about tool stack, processes and team. The result and recommendations are directly tailored to your real estate company.

Start the AI Advisor

Free · no account · GDPR-compliant