AI by Industry: Applications, Opportunities and Examples

AI applications by industry are the specific ways in which artificial intelligence is deployed within a sector — from automatic triage in healthcare to inventory optimisation in retail and contract analysis in the legal sector. The impact of AI differs fundamentally by industry: what counts as a quick win in logistics (route optimisation) is a complex compliance project in healthcare. This article provides a complete overview of AI opportunities per sector, with concrete examples, cost indications, and links to our detailed industry articles.

Not every industry is in the same position. Some sectors have a head start thanks to the availability of structured data. Others lag behind due to strict regulations or traditional working methods. This overview helps you determine where your industry stands and which AI applications deliver the highest ROI.

How do you assess AI readiness by industry?

Before diving into specific sectors, it is useful to understand which factors determine how quickly an industry can adopt AI. Four criteria are decisive:

Data availability: Does the sector already have large quantities of digital data? Financial services and e-commerce score high here — they have been running on databases for decades. Construction and hospitality score lower — much information still lives in heads, spreadsheets, and paper receipts.

Regulatory pressure: Sectors with strict compliance requirements (healthcare, finance, legal) require extra steps in AI implementation. The EU AI Act classifies AI applications by risk, and most medical and financial applications fall into the "high risk" category — requiring additional documentation and human oversight.

Process repeatability: The more standardised, repetitive processes a sector has, the more AI can automate. Accounting and logistics score high here. Creative services and strategic consulting score lower.

Labour market pressure: Sectors with large staff shortages have a stronger business case for AI. Healthcare, hospitality, logistics, and construction all face structural shortages — AI can partially relieve that pressure.

AI in healthcare: triage, administration, and diagnostics

The healthcare sector is under extreme pressure: staff shortages, rising care demand, and increasing administrative burdens. AI offers concrete relief on three fronts.

Top 3 applications:

1. Automatic triage and referral: AI systems analyse incoming patient requests (via phone, chat, or portal) and categorise them by urgency. This shortens waiting times and ensures that GPs and specialists spend their time on the cases that need it most.

2. Administrative relief: Medical record keeping, reporting, and claims processing cost healthcare workers an average of 35% of their working time. AI solutions such as automatic conversation summaries (speech-to-text + summarisation) and claims assistants bring this down to 15–20%.

3. Image recognition and diagnostics: AI models for X-rays, MRI scans, and pathology images achieve accuracy of 94–97% in controlled studies — comparable to experienced specialists. They function as a "second pair of eyes" that flags abnormalities.

Dutch example: Erasmus MC uses AI for analysing CT scans for pulmonary embolisms, reducing assessment turnaround from 30 minutes to 5 minutes per scan. UMC Utrecht deploys AI for automatic classification of pathology images.

Cost indication: EUR 15,000–60,000 for an AI triage or administration system, depending on integration with the EHR (Electronic Health Record). Ongoing costs: EUR 500–2,000/month.

Note: Medical AI applications fall under the EU AI Act as high-risk systems and must comply with the MDR (Medical Device Regulation). Human oversight is mandatory.

Read our full article: AI in healthcare: applications and opportunities.

AI in hospitality: reservations, procurement, and staffing

The hospitality sector faces structural staff shortages and thin margins. AI helps not by replacing cooks or waitstaff, but by reducing administrative and logistical burdens.

Top 3 applications:

1. Reservation management and no-show prediction: AI analyses reservation patterns and predicts no-shows with 80–85% accuracy. Based on this, you can apply controlled overbooking and prevent empty tables. A restaurant with 60 covers experiencing 15% no-shows loses EUR 200–500 per evening.

2. Procurement optimisation: AI predicts ingredient needs based on reservations, weather forecasts, events, and historical data. This reduces food waste by 15–25% and prevents running out on busy evenings.

3. WhatsApp and chatbot reservations: A chatbot takes reservations, answers frequently asked questions about the menu, allergies, and opening hours, and sends automatic confirmations and reminders. This saves 5–10 hours per week in phone traffic.

Dutch example: Restaurant chain The Harbour Club uses AI-driven procurement predictions that reduced food waste by 22%. Hotel chain Van der Valk deploys chatbots for reservations and room service via WhatsApp.

Cost indication: EUR 3,000–15,000 for a chatbot or reservation system. Procurement optimisation: EUR 5,000–25,000 depending on integration with POS systems and suppliers.

Read our full article: AI for hospitality and restaurants.

AI in retail and e-commerce: personalisation and inventory

Retail and e-commerce are the sectors where AI is most advanced. The reason: there is an enormous amount of available data — transactions, search behaviour, click patterns, stock levels, return reasons — and processes are largely digital.

Top 3 applications:

1. Product personalisation: AI analyses buying behaviour and shows customers products that match their preferences. Amazon generates 35% of revenue via AI-driven recommendations. For Dutch webshops, revenue uplift from personalisation ranges between 8–15%.

2. Dynamic pricing: AI adjusts prices based on demand, stock, competition, and seasonal patterns. Bol.com and Coolblue have been using this for years. For smaller webshops, tools like Prisync and Competera now make this accessible from EUR 200/month.

3. Inventory management and demand forecasting: AI predicts which products will sell when and optimises order timing and quantities. This reduces both overstock (capital tied up) and stockouts (missed revenue) simultaneously.

Dutch example: Wehkamp uses AI for personalised product recommendations that increased conversion rates by 12%. Hema deploys AI for demand forecasting per store, improving inventory accuracy by 18%.

Cost indication: EUR 5,000–30,000 for a personalisation engine. Inventory optimisation: EUR 8,000–40,000. SaaS solutions start from EUR 200–1,000/month.

Read our full article: AI for retail and e-commerce.

AI for accountants: invoice processing and anomaly detection

The accounting industry processes enormous volumes of structured financial data — the ideal feeding ground for AI. The staff shortage in the profession makes automation not a luxury but a necessity.

Top 3 applications:

1. Invoice processing: AI-based OCR recognises purchase invoices, extracts supplier, amount, VAT, and invoice number, and automatically posts to the accounting system. Accuracy: 95%+. A firm processing 500 invoices/month saves 25 hours per month.

2. Bank reconciliation: AI matches bank transactions with open invoices based on amount, description, and payment history. For recurring payments, the system learns which posting is standard. Result: 80–90% automatic reconciliation.

3. Anomaly detection: AI flags duplicate invoices, unusual amounts, fraud risks, and missing documents. What takes hours per client per quarter manually, AI does continuously.

Dutch example: Firms on Exact Online and Yuki already use AI-driven booking suggestions. Silverfin automates reporting and compliance for larger firms.

Cost indication: EUR 3,000–15,000 for invoice processing, EUR 5,000–25,000 for a full AI package including anomaly detection and automatic annual report preparation.

Read our full article: AI for accountants.

AI in logistics: routing, inventory, and demand forecasting

Logistics is a sector where minutes and kilometres directly cost money. AI optimises the three biggest cost drivers: routes, inventory levels, and workforce planning.

Top 3 applications:

1. Route optimisation: AI calculates optimal routes based on delivery addresses, traffic data, time windows, and vehicle capacity. Savings: 10–20% on fuel costs and 15–25% on driving time.

2. Demand forecasting: AI predicts how many products are needed when, based on historical data, seasonal patterns, promotions, and external factors (weather, economy). This reduces both overstock and out-of-stock situations.

3. Warehouse automation: AI optimises warehouse layout, determines pick routes, and predicts when restocking is needed. Combined with WMS (Warehouse Management Systems), this delivers 20–30% efficiency improvement.

Dutch example: PostNL uses AI for parcel volume prediction and route optimisation. DHL deploys AI for demand forecasting in their supply chain. De Greenery (fresh produce wholesaler) uses AI to predict freshness and shelf life.

Cost indication: EUR 10,000–50,000 for route optimisation. Demand forecasting: EUR 15,000–60,000. SaaS platforms like Ortec and AIMMS offer solutions from EUR 1,000/month.

Read our full article: AI in logistics.

AI in manufacturing: quality control and predictive maintenance

Manufacturing combines two characteristics that accelerate AI adoption: abundant sensor data (IoT) and high costs during downtime. AI here saves not just time but also prevents expensive production interruptions.

Top 3 applications:

1. Visual quality control: Cameras with AI image recognition inspect products on the production line and detect defects that the human eye misses. Accuracy: 99%+ with trained models. Speed: 10–50x faster than manual inspection.

2. Predictive maintenance: Sensors measure vibrations, temperature, pressure, and sound from machines. AI recognises patterns that indicate wear or impending failures and warns before the machine breaks down. This reduces unplanned downtime by 30–50%.

3. Production scheduling: AI optimises production planning based on orders, machine capacity, raw materials, and delivery agreements. This maximises utilisation rates and shortens lead times.

Dutch example: ASML uses AI for quality control in lithography machine production. FrieslandCampina deploys AI for predictive maintenance on dairy production lines. VDL Nedcar used visual inspection systems on the assembly line.

Cost indication: EUR 15,000–75,000 for a visual inspection system. Predictive maintenance: EUR 20,000–100,000 depending on the number of sensors and machines. Payback period: 6–18 months through reduced downtime costs.

Read our full article: Manufacturing automation.

AI in construction: planning, safety, and estimation

The construction sector is one of the least digitalised industries in the Netherlands. That means two things: there is substantial room for improvement, and the first steps need not be complex to have significant effect.

Top 3 applications:

1. Project planning and estimation: AI analyses historical project data (lead times, material quantities, staffing) and generates more accurate quotes and schedules. Contractors using AI estimation report 15–20% more accurate projections.

2. Safety monitoring: Cameras with AI image recognition detect unsafe situations on the construction site: missing helmets, people in danger zones, unsecured scaffolding. The system warns in real-time via an app or intercom.

3. Document management and BIM integration: AI classifies and searches the enormous volume of project documentation — drawings, specifications, change requests, RFIs — and links information to the BIM model (Building Information Modelling).

Dutch example: BAM Infra uses AI for risk analysis on infrastructure projects. Heijmans experiments with AI-driven scheduling that integrates weather forecasts. VolkerWessels deploys drones with AI image recognition for progress monitoring.

Cost indication: EUR 5,000–20,000 for AI estimation. Safety monitoring: EUR 10,000–40,000 per site. BIM integration: EUR 15,000–50,000. ROI is highest on larger projects (>EUR 1M) where estimation errors and delays represent the biggest cost items.

Read more about AI opportunities in construction: AI for construction and installation.

AI in the legal sector: contract analysis and due diligence

The legal sector runs on text — contracts, case law, legislation, correspondence. AI language models are particularly good at analysing and summarising large volumes of text, making the legal sector a natural candidate.

Top 3 applications:

1. Contract analysis: AI reads contracts, identifies deviating clauses, compares against standard terms, and flags risks. What costs a lawyer 2–4 hours per contract, AI does in 5–10 minutes — with 90–95% accuracy on trained models.

2. Due diligence: During acquisitions, mergers, or large transactions, thousands of documents must be reviewed. AI searches data rooms, classifies documents, and flags relevant passages and red flags. Time savings: 60–80% compared to manual review.

3. Case law research: AI searches court databases, EUR-Lex, and other sources to find relevant rulings that apply to the specific case. Including a summary of the core of each judgment and its relevance to the current matter.

Dutch example: Pels Rijcken (State Attorney) uses AI for case law research. Loyens & Loeff deploys contract analysis AI for due diligence in mergers and acquisitions. Legalloyd offers AI-driven legal intake for SMBs.

Cost indication: EUR 5,000–25,000 for a custom contract analysis tool. SaaS solutions like Luminance, Kira Systems, and Legalis start from EUR 500–2,000/month. Due diligence automation: EUR 15,000–50,000 per project.

Read more: AI for the legal sector.

AI for financial services: fraud, advisory, and compliance

Financial institutions were the first large-scale adopters of AI. Banks, insurers, and asset managers have been using AI for years for fraud detection, risk modelling, and customer analysis. These applications are now becoming accessible to smaller financial service providers as well.

Top 3 applications:

1. Fraud detection: AI analyses transaction patterns in real-time and flags suspicious activity. Machine learning models detect fraud with 95–99% accuracy and reduce false positives by 50–70% compared to rule-based systems.

2. Client advisory and financial planning: AI-driven tools analyse a client's financial profile and generate personalised advice — from investment strategies to mortgage optimisation. Not as a replacement for the advisor, but as preparation that automates 60–70% of the groundwork.

3. Compliance and reporting: AI automates KYC (Know Your Customer), AML (Anti-Money Laundering) checks, and regulatory reporting. What takes days per dossier manually, AI does in minutes — with a structured overview of findings.

Dutch example: ING uses AI for real-time fraud detection on more than 10 million transactions per day. Rabobank deploys AI for mortgage acceptance. Nationale-Nederlanden uses AI for damage assessment on insurance claims.

Cost indication: EUR 20,000–100,000+ for custom fraud detection. Compliance automation: EUR 10,000–50,000. SaaS solutions for smaller players: EUR 500–3,000/month.

Read more: AI for financial services.

AI for marketing and content: creation, analysis, and SEO

Marketing is one of the most accessible sectors for AI. The tools are mature, prices are low, and results are quickly visible. Virtually every marketing department — whether a team of 20 or a one-person operation — can benefit from AI.

Top 3 applications:

1. Content creation: AI generates blog articles, social media posts, ad copy, product descriptions, and newsletters. Not as a final product, but as a first draft that an editor refines. Time savings: 50–70% on content production.

2. SEO analysis and optimisation: AI analyses search trends, competition, and content gaps, and generates optimisation recommendations. Tools like Surfer SEO, Clearscope, and MarketMuse use AI to determine which topics you should cover and how to improve existing content.

3. Campaign optimisation: AI optimises ad campaigns on Meta, Google, and LinkedIn by automatically adjusting bid strategies, audiences, and creatives based on performance. Google's Performance Max and Meta's Advantage+ are examples of AI-driven campaign tools.

Dutch example: Coolblue generates thousands of product descriptions with AI. KLM uses AI for personalised email campaigns. Bol.com deploys AI for A/B testing of ad copy at scale.

Cost indication: EUR 50–500/month for AI writing tools (Jasper, Copy.ai). SEO tools with AI: EUR 100–500/month. Campaign optimisation is built into the advertising platforms.

Read more: AI for marketing and content.

More industries: education, real estate, insurance and maintenance

Beyond the sectors discussed above, four industries are seeing rapid AI adoption:

• Education and training: AI saves teachers 15-20 hours per week on grading, administration and lesson planning. Read more: AI for education and training.
• Real estate: From automated valuations to lead scoring and virtual tours — AI accelerates the entire sales process. Read more: AI for real estate agents.
• Insurance: Claims processing, fraud detection and risk assessment automated with AI. The insurance sector is among the most AI-ready industries thanks to its vast structured data. Read more: AI for insurance claims processing.
• Maintenance companies: HVAC, plumbers and electricians benefit from AI for scheduling, dispatch and customer communication. Read more: AI for maintenance companies.

Quick-win matrix: where should SMBs start?

Not every AI application is equally complex or expensive. Here is a matrix that ranks applications by implementation effort versus impact.

The golden rule: Start with applications that have low implementation effort and deliver measurable impact. Prove AI's value with a quick win, and use that success to build support for more complex projects.

How do you choose the right AI application for your industry?

The choice depends on three factors you can fill in concretely:

1. Where do you lose the most hours?

Have your team track for two weeks which tasks take the most time and require the least thinking. The tasks most frequently labelled as "mindless work" are your best candidates. This is the same approach we describe in our guide about implementing AI in SMBs.

2. Where do you make the most errors?

Errors are expensive — not just in direct costs but also in customer loss and reputational damage. AI is consistent: it does not make mistakes from fatigue, distraction, or rushing. Processes with a high error rate and high error costs are strong candidates for AI.

3. Where does the data sit?

AI without data is like a car without fuel. If you already use a CRM, ERP, accounting system, or webshop, you probably have enough data to get started. If your data still lives in spreadsheets, emails, and people's heads, the first step is digitalisation — not AI.

Want to find out if your business is ready? Read our article about determining if your business is ready for AI. Or read our guide on hiring AI consulting for a structured approach.

The cost of doing nothing

Finally, a perspective that is often overlooked: the cost of not deploying AI. While you are still deliberating, your competitors are automating their invoice processing, personalising their customer communications, and optimising their inventory. The productivity gap grows every month.

A concrete example: a wholesaler manually processing 2,000 invoices per month at EUR 35/hour spends EUR 35,000 per year on it. With AI, that becomes EUR 7,000 per year. That is EUR 28,000 per year in lost competitiveness — every year you wait.

The question is not whether AI is relevant for your industry — it is. The question is which application delivers results fastest. Regardless of your industry, customer experience is the common thread. Learn how AI improves CX in our complete guide. Check the costs in our overview of AI costs for SMBs, or schedule a conversation with our team for tailored AI consulting.