AI in 2025: Marketing and Sales Are Driving Adoption
One in six companies used AI in 2025, roughly double from two years ago. Marketing and sales are the top use cases, based on provisional figures from Statistics Netherlands (CBS). If you work in revenue, this is your cue: adoption is no longer a bet, it's table stakes.
Adoption by company size (provisional)
Growth is broad-based, with the sharpest jump in mid-sized firms. Companies with 50-249 employees surged from 20% in 2023 to 45% in 2025. Large enterprises lead, but smaller teams are catching up fast.
- Total (2+ employees): 17% in 2025, 13% in 2024, 8% in 2023
- 2-9 employees: 14% in 2025, 11% in 2024, 7% in 2023
- 10-49 employees: 27% in 2025, 19% in 2024, 11% in 2023
- 50-249 employees: 45% in 2025, 31% in 2024, 20% in 2023
- 250+ employees: 66% in 2025, 53% in 2024, 42% in 2023
Where AI is used most
Among companies that use AI, 35% apply it in marketing and sales. 32% use it for business administration and management, and 25% for R&D or innovation. Logistics is still early at 4%.
Sector notes: nearly half of trade companies and about 40% of information and communication, renting and leasing, and other business support activities use AI for marketing or sales. Around half of financial services firms apply AI to business administration or management.
The tech that's sticking
- Text mining: 12% (up from 4% in 2023)
- Natural language generation: 8% (up from 3% in 2023)
- Speech recognition: 6% (tripled since 2023)
- Machine learning: 4%
- RPA: 3%
- Image recognition/processing: 2%
- Autonomous robots/drones: 1%
Translation: text and voice are where most teams are getting results today-content, conversations, documents, and data you already have.
Practical plays for marketing and sales
- Text mining for demand signals: Pull themes from reviews, chats, and social comments. Turn the top 10 objections and triggers into ad angles and landing page sections.
- Speech-to-text on calls: Transcribe and tag discovery and support calls. Feed common questions into better scripts and objection handling.
- Natural language generation for volume: Draft first-pass ad variations, email sequences, product copy, and SEO briefs. Keep human editing in the loop.
- Lead routing and scoring: Light ML on form data and interaction history to score fit and intent. Start with rules, add models once you have labeled outcomes.
- RPA for handoffs: Automate CRM updates, follow-up tasks, and quote creation to cut delay between interest and response.
Metrics that actually move
- Conversion rate lift by segment and channel
- Time-to-first-response and time-to-meeting
- Cost per qualified lead and sales cycle length
- Email reply rate and call outcome rate after script updates
- Content production time saved vs. baseline
30-day rollout plan
- Week 1: Pick one pipeline stage with friction (e.g., lead response). Map the workflow and define one success metric.
- Week 2: Pilot text mining on past chats/calls. Build a short list of winning angles and a new reply framework.
- Week 3: Add speech recognition on new calls. Auto-tag objections and outcomes. Route hot intents faster.
- Week 4: Use generation to produce controlled variants (ads/emails), A/B test, and keep only what lifts the metric.
Lightweight stack ideas
- Data in: Call recordings, chat logs, CRM notes, form fills
- Analysis: Text mining for topics, sentiment, and intent; simple scoring rules
- Action: CRM updates, instant responses, content drafts, task automation
- Review: Human approval gates for outbound copy and pricing
Risk and guardrails without the headache
- Keep customer data scoped. Mask PII in transcripts and logs.
- Log prompts/outputs for audit and quality checks.
- Document use cases and risks with a simple checklist. The NIST AI Risk Management Framework is a solid reference.
- If you sell in the EU, track obligations under the EU AI Act for higher-risk use cases.
Skill up fast
If you want structured, marketing-first training, see the AI Certification for Marketing Specialists. You can also browse by role here: Courses by Job, or explore toolkits for copy and content here: AI Tools for Copywriting.
The takeaway
Marketing and sales are where AI is earning its keep right now. Start with text and speech workflows you already run every day, measure the lift, and double down on what proves itself.
Your membership also unlocks:
