- Is it really worth moving from traditional to AI automation for your company?
- What's really happening in the market: the end of fixed-rule automation
- What changes in practice: who does what in the new AI + human model?
- Three real-world scenarios: how much does an Italian company gain by migrating?
- Which platforms and paths can you choose for AI automation?
- What risks and rules should you keep under control before migrating?
- Why everyone's talking about it now: is the Italian market really ready?
- How do you actually get started: the steps to migrate from classic to AI automation
- What are the most common mistakes people make when taking this leap?
- How do you tell whether the new system is really working?
- So, is it worth it for your company?
- How we see it at SHM
- Frequently Asked Questions
- Rule-based automation (built on fixed rules) is no longer enough: the new AI systems read context and act on their own, without waiting for you to write every single rule.
- In Italy only 30% of large companies have adopted intelligent automation; SMEs are even further behind, with a real risk of losing ground to competitors.
- Migrating pays off if you have medium-to-high contact volumes and repetitive flows: the real gain is time given back to your team, not just faster campaigns.
- The winning model isn't "AI instead of people" but AI that proposes and humans who approve: the so-called human-in-the-loop.
62% of large Italian companies already use automation tools, but only 30% have integrated AI features. The rest are still running on the autopilot of the 2010s, and the gap widens every month.
Is it really worth moving from traditional to AI automation for your company?
It depends on how many contacts you manage and how much time your team loses writing and updating rules. If you have many flows and they change often, migration pays for itself in a few months.
Traditional automation works like this: you write a rule ("if the contact opens the email, send the follow-up after 3 days"), and the platform runs it. It worked well when channels were few and behavior was predictable. Today touchpoints, the points of contact with your customer, keep multiplying and rules pile up by the hundreds. Maintaining them is a full-time job.
The new AI systems don't wait for a rule: they read customer data, work out the right moment and decide the action. The human team supervises, corrects and approves. This is the human-in-the-loop model that MarTech discusses in its analysis of the future of automation.
| Your situation | Should you migrate? | why |
|---|---|---|
| You have more than 5,000 active contacts and complex flows | Yes | AI quickly pays back the cost of migration |
| You run an e-commerce store with abandoned carts and upsells | Yes | Contextual personalization increases conversion rates |
| You have fewer than 1,000 contacts and simple flows | No, for now | Classic automation is still enough and costs less |
| You've used a legacy platform for more than 4 years | Yes | You risk losing integrations with new channels |
| Your team doesn't have time to manage rules | Yes | AI measurably reduces the operational workload |
What's really happening in the market: the end of fixed-rule automation
Systems built on static rules are giving way to platforms that use AI language models to read customer context and act in real time, without waiting for manual instructions.
According to MarTech, the core problem of classic automation is "context rot". Data goes stale, rules don't get updated, and the result is that customers receive the wrong messages at the wrong time. A welcome email sent three weeks after signup, for example.
The new AI systems tackle this problem differently: instead of running rules, they continuously analyze signals (opens, clicks, purchases, site visits) and adapt their behavior. But it's not enough to add a "thin layer of AI" on top of an old platform: you need to redesign the data flow.
Data quality is central. Large language models (LLMs, the AI engines behind tools like ChatGPT) can't be pointed at billions of raw data points without cleaning and structuring them first. Garbage in, garbage out.
The data is clear: 85% of European companies already use AI tools in marketing, according to IAB Europe. Italian companies are lagging, with only 30% having adopted intelligent automation, according to the Osservatori Digital Innovation of Politecnico di Milano. That gap is a risk, but also an opportunity: those who move now start with data they've already accumulated and can speed ahead.
What changes in practice: who does what in the new AI + human model?
In the human-in-the-loop model, AI proposes and optimizes, while the human team validates strategic decisions and handles exceptions. Roles don't disappear: they change.
- 1 · AI platformCollects and cleans dataAggregates signals from your CRM, website, email and social channels. Identifies outdated data and automatically excludes it.
- 2 · AI modelGenerates the recommended actionSuggests the best message, channel and timing for each contact, based on recent behavior.
- 3 · Marketing teamApproves or correctsValidates high-impact proposals and sets ethical limits and brand guidelines. No more writing every single rule.
- 4 · AI platformExecutes and monitorsLaunches the communication, measures results in real time and updates the model for the next cycle.
- 5 · Marketing teamAnalyzes strategic resultsFocuses on high-level decisions: segments, offers, positioning. Day-to-day operations are delegated.
This setup solves one of the problems marketers cite most: AI churning out content nonstop, creating more clutter than value. With human oversight in the loop, every output is filtered before it reaches the customer. As we explored in our article on MoEngage AI agents and the new face of marketing automation, real personalization always needs a human keeping watch over data quality and message quality.
Three real-world scenarios: how much does an Italian company gain by migrating?
Benefits vary a lot by industry and contact volume, but the main gain is always the same: hours of work given back to the team and more relevant messages for the customer.
The numbers below are illustrative examples: they show the order of magnitude of the benefit and are not real client data.
| Company type | Current situation | With AI automation | Estimated benefit |
|---|---|---|---|
| Fashion e-commerce (50k contacts) | 20 manual flows, 18% open rate | Adaptive flows, 26-28% open rate | ~+8 pp opens, -15 hrs/week of management |
| B2B professional firm (2k contacts) | Monthly newsletter + 3 fixed sequences | Contextual sequences by funnel stage | Fewer unsubscribes, more qualified appointments |
| Retail store chain (8k contacts) | Same promotions for every customer | Offers personalized by purchase behavior | Fewer wasted coupons, higher average basket |
Which platforms and paths can you choose for AI automation?
The market offers three main paths: upgrade the platform you already use, switch to an AI-native solution, or build a custom AI layer on top of your data.
| Approach | Suitable for | Pros | Cons |
|---|---|---|---|
| Upgrading your existing platform | Those who have already invested in HubSpot, Salesforce, ActiveCampaign | Less disruption, data already in place | AI features often superficial or paid extras |
| Migrating to an AI-native platform | Those starting from scratch or with highly fragmented data | Architecture designed for AI from the start | Migration cost and team training |
| Custom AI layer | Companies with complex flows and proprietary data | Maximum flexibility and control | Requires in-house technical skills or a partner |
| Agentic automation (Agentic AI) | Early adopters with a budget and a technical team | Maximum autonomy, continuous learning | Still experimental for most SMEs |
Agentic Automation — systems based on language models that can use external tools and learn on their own — is still in its early days in Italy: only 8% of large companies have started experiments, while 25% plan to in the coming months, according to the Osservatori del Politecnico di Milano. To learn how these autonomous agents work, the article on autonomous AI agents and the Monks model offers a practical overview.
What risks and rules should you keep under control before migrating?
There are three main risks: data quality, privacy compliance and dependence on a single vendor. All three can be managed with a plan before you switch on any system.
The first risk is dirty data. An AI system fed with duplicate contacts, outdated emails or expired consents delivers worse results than classic automation. Before migrating, a database audit is mandatory, not optional.
The second risk is GDPR compliance (the European data protection regulation). AI systems that profile user behavior must respect the legal bases for processing. If you use behavioral data to personalize, make sure the consent you collected covers this use too.
The third risk is vendor lock-in — dependence on a single vendor. Choose platforms with open APIs (interfaces that let you connect other systems) and always check that your data can be exported. On the topic of AI agent governance, the analysis on OpenAI and AI agents out of control is also worth reading.
| Risk | Warning sign | How to manage it |
|---|---|---|
| Outdated or duplicate data | Email bounce rate > 5% | Database audit and cleanup before migration |
| Insufficient GDPR consent | Consents collected before 2018 and never updated | Re-opt-in campaign before enabling AI profiling |
| Vendor lock-in | No data export API | Choose platforms with open standards |
| AI hallucinations | Out-of-context messages or ones inconsistent with the brand | Mandatory human review for high-impact content |
| Over-automation | The team no longer knows what's running | Centralized log of all active flows |
Why everyone's talking about it now: is the Italian market really ready?
The AI market in Italy is worth €1.8 billion and is growing 50% a year, but real adoption among SMEs is still low. Those who move now gain a concrete competitive advantage.
The market is growing, but real adoption is concentrated in large companies. Among SMEs, only 8% have started at least one AI project, according to the Osservatori Digital Innovation del Politecnico di Milano. Italy ranks last in Europe for intelligent automation adoption (23%), well behind the United Kingdom (32%), Spain and Germany (26%), according to InnovationCity data. The gap is real, but those who start now still face little competition in advanced personalization.
Italian B2C e-commerce topped €62 billion in 2025, with 35.2 million digital consumers, according to the Osservatori del Politecnico di Milano and Netcomm. At this scale, personalized automation isn't a luxury: it's infrastructure.
Where do you start? The steps to migrate from classic automation to AI
Migration happens in phases, not all at once. The first step is always cleaning your data, not choosing a platform.
The most common mistake is picking the tool before understanding your own data. A poorly fed AI system delivers worse results than a well-configured classic automation. Start from the ground up.
- 1 · Internal team / consultantAudit your contact databaseRemove duplicates, update GDPR consents, segment by source and behavior. Goal: a reliable database.
- 2 · Marketing teamMap your existing flowsList every active automation flow. Identify the high-volume ones and the ones that change often: they're your priority migration candidates.
- 3 · IT team / partnerAssess the integrations you needMake sure your CRM, e-commerce platform and analytics tools talk to each other. Without unified data, AI has no context.
- 4 · Marketing team + managementChoose your platform or approachBased on your priority flows and budget, decide whether to upgrade your current platform or migrate. Test on a small segment before scaling.
- 5 · Marketing teamSet up the human-in-the-loop modelDefine which decisions AI can make on its own and which require human approval. Write these rules down before you switch the system on.
- 6 · Marketing / analytics teamMeasure and iterateTrack your KPIs week by week for the first three months. Adjust the model's parameters based on real results.
What are the most common mistakes people make when taking this leap?
90% of mistakes happen before you even choose a platform: dirty data, wrong expectations and a team that isn't involved.
According to the Omnichannel Customer Experience Observatory at Politecnico di Milano, about one in three companies had not yet launched any personalization initiative in 2025. Among those that had, most use basic approaches. The problem isn't technology: it's method.
| Error | Why it happens | How to avoid it |
|---|---|---|
| Starting from the platform, not the data | The urge to see something running right away | Audit your database before any purchase |
| Thinking AI will replace the team | Expectations inflated by vendors | Define human roles in the process from day one |
| Launching too many flows at once | Initial enthusiasm | Pilot on one segment, then scale |
| Not updating GDPR consents | Assuming old consent is good enough | Run a re-opt-in campaign before migrating |
| Not measuring results for the first 90 days | No KPIs defined up front | Pick 3 key metrics before you start |
| Ignoring content quality | Believing AI writes on its own | Strategic copywriting stays human |
How do you tell whether the new system is really working?
Positive signals show up within 60-90 days: rising open rates, fewer unsubscribes, fewer hours spent managing flows manually.
| Indicator | What it measures | Alert threshold |
|---|---|---|
| Email open rate | How relevant the message is to the recipient | Below 20% after 60 days: revisit your segmentation |
| Click-through rate (CTR) | Quality of the offer and send timing | Below 2%: the content isn't contextual enough |
| Unsubscribe rate | Annoyance felt by the contact | Above 0.5%: you're sending too much or out of context |
| Weekly hours spent managing flows | Team operational efficiency | If it doesn't drop within 90 days, the system isn't configured well |
| Conversion rate by segment | Effectiveness of personalization | Compare with the pre-migration period, same segment |
| Errors or off-brand messages | Quality of human oversight in the loop | Even a single case calls for an immediate review of the process |
So, is it worth it for your company?
Worth it if you have more than 5,000 active contacts, flows that change often and a team losing hours managing rules. The return in time saved and message relevance is measurable within three months.
A middle path if you're an SME with mid-sized volumes and a platform that already works. You can start by adding AI features to the platform you use, without migrating everything. Test on one flow, measure, then decide.
Not yet if you have fewer than 1,000 contacts, simple and stable flows, or no clean database yet. Strengthen the basics first: a well-configured classic automation beats a poorly fed AI.
How we see it at SHM
For us, AI in automation only makes sense if it gives real time back to the team, not if it adds one more tool to manage. We believe technology should free people from routine, not create new layers of complexity to oversee.
In our experience with companies that manage complex catalogs and contact flows across multiple channels, the critical point is never the platform: it's data quality and clarity on who decides what. An AI system configured without these foundations produces noise, not results.
If you're weighing whether and how to evolve your automation, our starting point is always an audit: understand where you are now before deciding where to go. You can explore the topic in our hub on AI and process automation or go straight to our services for strategic AI integration and data processing and business intelligence. If you want to train your internal team, we also offer a business process automation training program.
The article on OpenAI AI agents for SMEs offers another perspective on how these tools are changing day-to-day marketing.
Frequently Asked Questions
Not necessarily. Many platforms you may already use (HubSpot, ActiveCampaign, Salesforce) are adding AI features. The first step is to check whether the features in your current plan are enough for your priority flows. A full migration only makes sense if your current platform can't support the integrations you need.
The first signals arrive within 60-90 days: open rates, clicks and unsubscribes are the quickest to measure. The benefits to your team's operational efficiency (fewer hours spent managing rules) show up in the first few weeks. Revenue results need at least one full purchase cycle.
It depends on how it's configured. Systems that profile user behavior need a valid legal basis (consent or legitimate interest, depending on the case). Before you turn on any AI personalization system, check that the consents you've collected also cover this type of processing, and update your privacy policy.
It's a real risk, especially in the early stages. The human-in-the-loop model exists for exactly this reason: high-impact messages (offers, sensitive communications, tone of voice) are approved by a person before sending. Decide upfront which flows need human approval and which can run on their own.
With fewer than 1,000 contacts and simple flows, classic automation is still enough and costs less. AI pays off when you have many flows, they change often, or they require personalization that would be impossible to manage by hand. Our advice: consolidate your database and core processes first.
It's the problem of the data used by automation going stale without being updated: a customer who bought six months ago still gets messages meant for a new buyer, for example. Modern AI systems try to solve this by continuously updating the contact profile based on recent signals. But it only works if the starting database is clean.
Automation has never been a tooling problem: it has always been a problem of data, processes and who decides what. AI changes the speed, not the substance.
Related articles
Discover more articles exploring similar topics, selected to offer you a more complete and stimulating perspective. Each piece of content is carefully chosen to enrich your experience.
