Generative AI in marketing: how to do it right
Erwin Berkouwer
Your marketing team loses hours each week writing copy, adjusting campaigns and creating reports. Generative AI takes this work over, without lowering quality.
Your marketing team works hard. Yet writing one campaign takes three days, social media falls behind and the monthly report takes another afternoon. The problem isn't the people, it's the way of working.
Generative AI changes that. Not with empty promises, but with concrete time savings that SMEs are realizing right now. In this article, you'll read what it delivers, how it works and how to get started.
The problem: marketing costs too much time for too little result
Marketing is a time-consuming task for many SMBs that never seems to finish. You write a newsletter, adjust the text three times, ask for feedback, correct again and send it out late. Meanwhile, the customer is waiting for a quote and a pile of other tasks waits to be done.
The costs of this pattern are concrete. An employee spending ten hours a week on content creation costs you over 25,000 euros per year in labor costs, excluding the missed opportunities from slow execution. Companies that change nothing now see competitors using AI pull further ahead. Not in five years, but now.
What most articles on generative AI marketing don't tell you: the biggest obstacle isn't the marketer, it's the director or operations manager who has to approve the investment without understanding what it concretely delivers. 85% of Dutch marketers are positive about generative AI, but at many companies the decision stalls due to lack of concrete ROI numbers. We'll solve that below.
The solution: generative AI as a productive colleague
Generative AI is software that produces text, images and data itself based on an instruction. You specify what you need, and the system delivers a usable first draft within seconds.
That sounds simple, and it is. The power lies in scalability. One employee can produce four times as much marketing material with AI as before, without quality loss. A digital marketing agency that applied this quadrupled its content production while keeping the team the same size.
Practical tip: Don't start with all applications at once. Choose one task that costs the most time now, execute it with AI and measure the difference after four weeks. This builds internal support and proves value with your own numbers.
The difference from traditional tools: generative AI doesn't write fixed templates. It adjusts tone of voice, incorporates your brand guidelines and tailors content to the channel, audience and campaign goal. That's the level where it adds real value.
A point that almost no article mentions: generative AI only works well when connected to the systems your company already uses. An AI disconnected from your CRM, customer data or content library delivers average results. An AI working from HubSpot, Salesforce or your own database produces content that's immediately usable.
Practical applications with ROI numbers
These are the most common applications for SMEs, with concrete time savings per week:
Website and blog content creation
Current time investment: 6-8 hours per piece. With AI: 1-2 hours (first draft in minutes, editing after). Savings: 5-6 hours per article. Direct benefit: more content, higher search visibility.
Social media posts and campaign copy
Current time investment: 3-4 hours per week. With AI: 30-45 minutes. Savings: 2-3 hours per week. Direct benefit: consistent presence on all channels, even when busy.
Email campaigns and newsletters
Current time investment: 4-5 hours per edition. With AI: 1 hour. Savings: 3-4 hours per edition. Direct benefit: higher open rates through better subject lines and personalization.
Reports and performance analyses
Current time investment: 3-4 hours per month. With AI: 30 minutes. Savings: 2-3 hours per month.
Google and LinkedIn ad copy
Current time investment: 2-3 hours per campaign. With AI: 20-30 minutes. Savings: 2 hours per campaign.
SMBs using generative AI consistently save 20 hours per week on marketing tasks. Translated into costs: at an hourly rate of 50 euros, that's 1,000 euros per week or 52,000 euros per year in reclaimed capacity. The investment pays for itself within 3 to 6 months.
Practical tip: Measure not just time savings, but also output. How many more campaigns did you run? How many more blog posts published? How many more leads generated? Those numbers make the business case concrete for your next budget round.
An insight most sources miss: generative AI works best when connected to your existing marketing software. Unify AI works with more than 40 systems, including HubSpot, Salesforce and Trengo. This means AI-generated content goes directly into the right workflow, without anyone needing to copy or retype data.
Here's a concrete example. An SMB with two employees each spending eight hours per week on marketing content saves an average of fifteen hours per week with AI. At an internal hourly rate of 45 euros: 15 hours x 45 euros x 52 weeks is 35,100 euros per year. Implementation costs for a standard AI marketing agent with CRM integration: one-time 8,000 euros plus 500 euros per month. Year 1 costs: 14,000 euros. Net benefit year 1: 21,100 euros. Payback period: four months.
How to get started: four steps
Step 1: Map your biggest time drains
Ask your marketing employee or team to track how much time they spend per task over two weeks. You'll see that content creation, social media and reports take the most time. Those tasks are your starting point. Make the list concrete: not 'content' but 'write blog post', not 'report' but 'create monthly PowerPoint summary'.
Step 2: Choose one application and test for four weeks
Start with email campaigns or blog posts. These are tasks where AI delivers results fastest and quality control is simple. Your employee reviews the output and makes small adjustments, but the writing time largely disappears. Set a measurement point upfront: how many hours per week saved, what's the output quality on a scale of 1 to 5.
Step 3: Connect AI to your existing systems
Standalone AI tools provide temporary results, but the real advantage is integration. When AI works from your CRM data, it knows who the customer is, what they bought before, and what's the next step in the sales process. That delivers personalized content that converts, not generic text anyone can write. This is the distinction almost no overview article makes: a generic AI tool is a word processor. An integrated AI agent is a marketing employee who understands your company.
Step 4: Scale based on results
After four weeks, you have data. You know how much time you saved, which content performs better, and where improvements are needed. Based on that, you decide which other marketing tasks to tackle with AI. Standard AI agents go live in 2 to 4 weeks. Custom agents integrating deeply with your systems go live in 4 to 6 weeks.
Practical tip: Involve your marketing employee actively in the implementation. AI doesn't replace people, it strengthens them. Someone who understands how it works gets four times as much out of it as someone who sees it as a threat.
Common mistakes to avoid
Starting too broad. Companies buy a large AI platform, roll it out across the company and discover three months later that no one uses it consistently. Better: start small, prove the value, then expand.
Seeing AI as a replacement. Generative AI works best as a supplement to people, not a replacement. A marketer using AI as an assistant produces more and better content. AI without human direction produces generic content no one reads.
Not connecting to existing data. AI that doesn't know who your customers are writes for nobody. Connect AI to your CRM, customer database and past campaign results. Then the output becomes truly usable. Dutch companies working with Exact Online, AFAS or e-Boekhouden can connect these systems directly to their AI marketing setup.
Not building in a measurement point. Without pre-set goals, you won't know after four weeks if the investment is worthwhile. Set a concrete target: five hours saved per week, three extra campaigns per month, or a newsletter open rate that rises by ten percent. That provides direction and makes the business case clear.
What generative AI marketing doesn't do
A realistic picture is at least as important as the benefits. Generative AI is not a complete marketing strategist. It generates text based on instructions, but it doesn't set campaign goals, analyze market positioning or determine budget allocation. That's still people's work.
AI also produces no original insights. It combines and reformulates based on what it knows. If you want to build a campaign around a unique customer insight or a new product not yet in AI training data, you still need a person to set the direction.
What AI does do: it takes over the repetitive execution work. Writing text variants, adjusting tone per channel, incorporating SEO elements, A/B testing campaign copy, writing social media responses. Those are the tasks where AI is consistently fast and good enough. By removing those tasks, you free up space for what really matters.
How SMBs use AI marketing in practice
Here are three concrete examples of how SMEs are already using generative AI:
Wholesaler with 45 employees
The company sent a weekly product newsletter to 3,200 customers. The sales manager spent four hours per week writing it. After implementing an AI agent connected to the product catalog system and HubSpot: the newsletter is ready in 40 minutes, including personalized product recommendations per customer group. Open rate rose from 22% to 31%.
IT service provider with 28 employees
The company wanted more LinkedIn content but nobody had time. With an AI agent generating three posts weekly based on project updates and industry news, taking the marketing manager 20 minutes per week to review and publish, LinkedIn reach grew 180% in three months.
Construction company with 80 employees
Quote follow-up communication cost two employees five hours per week each on follow-up emails and status updates. An AI agent connected to the CRM now automatically sends personalized follow-up messages at the right time. Quote to contract conversion rose 12 percentage points.
These examples share three things. They started with one specific process, they connected AI to existing systems, and they measured results after four weeks. That's the pattern that works.
Generative AI marketing: what it delivers for your company
Generative AI for marketing is no future talk. It's a concrete investment that SMEs are already getting 20 hours per week from, a 3 to 6 month ROI payback, and marketing capacity that scales without needing to hire immediately.
The difference is how you approach it. Standalone tools deliver limited results. Integrated AI agents working from your data, your systems and your brand guidelines deliver structural advantage. That's the investment that counts.
Want to know what AI can deliver for your company? Schedule a free consultation at unify-ai.nl/contact.
