The short answer: integrating CRM and ERP means making sure the customer record, the orders and the payment status exist once only and can be read by both systems. It isn't an IT project: it's a decision about which system is in charge of which piece of data. In Italian companies the ERP is almost always there and the CRM almost never: in 2025, 49.5% of companies with at least 10 employees used ERP software, only 21.7% CRM software (Istat, the Italian national statistics institute, Imprese e ICT, 2025).
The result is predictable. Finance works off one database, sales off another, and neither can answer simple questions: what is this customer worth, what did they buy last time, why have they stopped buying.
This article explains what integrating the two systems means in a company of 10 to 50 people, which data has to move and in which direction, in what order to proceed, and when it is better not to.
What integrating CRM and ERP means
Definitions first, because these words get used loosely.
- ERP (Enterprise Resource Planning): the system that governs orders, stock, invoicing and accounts. The system of things that have already happened.
- CRM (Customer Relationship Management): the system that governs leads, deals and sales activity. The system of things that have yet to happen.
- Integration: a connection that moves data between the two automatically and repeatably. Not a monthly export to Excel.
- Master system: the system that holds the official version of a given piece of information. For the legal customer record it is almost always the ERP; for the status of a deal it is the CRM.
- One-way synchronisation: data travels in a single direction. Two-way: it travels both ways, which is why it needs a rule on who wins in a conflict.
The hard part isn't technical. It's deciding, for each piece of information, which system is in charge. The integrations that fail are almost always the ones where that decision was never taken and two systems overwrite each other.
Why the ERP is there and the CRM isn't
The ERP is compulsory in practice: without one you can't issue invoices. The CRM is optional, so it lags behind. The figures bear this out.
| Software | Companies 10+ employees | SMEs (10–249) | Large (250+) |
|---|---|---|---|
| ERP | 49.5% | 48.8% | 85.9% |
| CRM | 21.7% | 21.1% | 56.5% |
Source: Istat, Imprese e ICT, 2025 (press release, 15 December 2025).
The European picture is structurally similar. In 2025, across the European Union, 46.45% of companies used an ERP to share information between business functions and 28.51% used a CRM; among small companies of 10 to 49 employees the shares were 41.08% and 24.69% (Eurostat, E-business integration, 2025 data). A direct comparison with the Italian figure should be treated with care, because the indicators are not defined identically, but the direction is the same: the system that records orders is widespread, the one that governs the commercial relationship far less so.
There is a second reason, less statistical. The Politecnico di Milano's Digital Innovation in SMEs Observatory finds that the main obstacles to digitalisation include competition with other business priorities and a lack of resources, and that only 27% of SMEs report having met no obstacles at all (press release, 21 May 2026). Translated: integration doesn't get rejected, it gets postponed.
Which data has to move between the two systems
Not all of it. A useful integration moves a few pieces of information, always the same ones, in a direction someone decided. This is the most common configuration in a company of your size.
| Information | Who is in charge | Direction | What it's for |
|---|---|---|---|
| Legal customer record (company name, VAT number, registered address) | ERP | ERP → CRM | Avoiding duplicates and customers spelled three different ways |
| New lead acquired online | CRM | CRM → ERP (on first sale) | Not creating legal records for people who haven't bought yet |
| Orders and revenue by customer | ERP | ERP → CRM | Giving the salesperson the customer's real value |
| Status of deals and quotes | CRM | CRM → ERP (on confirmed order) | Going from deal to order without re-keying the data |
| Overdue payments and blocked accounts | ERP | ERP → CRM | Stopping sales from selling to people who don't pay |
| Lead source (campaign, channel) | CRM | Stays in the CRM, linked to the order | Knowing which activity produced which sale |
The last row is the one SMEs skip most often, and it's the one that makes marketing measurable. Without the lead source surviving all the way to the order, every discussion about ad budget stays an opinion. I've written separately about the order you need in your data before any AI project.
In what order to proceed
Five steps. The order matters more than the technology you pick.
1. Map how the data moves today. Where a customer is born, who enters it, how many times the same information is typed in by hand. Two or three manual re-entries nobody had counted usually surface.
2. Decide the master system for each piece of data. A table like the one above, written down and agreed with finance and sales. It's the step people try to skip, and it's the one that decides the outcome.
3. Clean the customer records before connecting anything. Connecting two dirty archives produces one larger dirty archive, harder to fix. Deduplication comes first, with a written rule on who creates new records.
4. Start with one direction only. In most cases it's best to begin with ERP → CRM: customer records and revenue flowing down to sales. It's the lowest-risk flow, because it doesn't touch accounting data.
5. Test on a sample, then extend. Fifty customers, two weeks of watching, corrections, then the rest. An integration switched on across the whole database without a test is the fastest way to have to switch it off again.
When integration isn't worth it
This is my judgement, not a figure: in at least one case in three of those I've seen, integration wasn't the first thing to do.
- The CRM isn't being used. If the salespeople don't work in it, integration fills a system nobody opens. Make it useful first, connect it second.
- The ERP is about to be replaced. Building a connection on a system on its way out is work you'll do twice. Better to wait and fold the integration into the new project.
- Volumes are low. With a few dozen orders a month and one person handling them, a well-made periodic export costs far less and holds up.
- There's no one inside the company who owns it. An integration needs someone to decide on customer records when an ambiguous case comes up. Without that person, the mess reforms within months.
If, on the other hand, the CRM is already in use and the problem is that it lives apart from everything else, connecting it is almost always the intervention with the best ratio of cost to result. If you don't have a CRM yet, the question comes earlier: I've written about it in CRM for SMEs.
What it costs and how long it takes
There is no reliable public figure for the average cost of a CRM–ERP integration in Italian SMEs: I haven't found one in primary sources and I'm not going to invent one. What I can say is how the spend is structured, based on my own experience.
The variable that weighs most is not the number of systems, it's the quality of the customer records you start from. A company with tidy archives and an ERP that exposes its data in a standard way is a job of a few weeks. A company with three archives out of sync and a closed ERP is a job of months, in which the technical part is the smaller one. That's why any quote given without having looked at the data is a guess.
Frequently asked questions
Do CRM and ERP need to come from the same vendor?
No, and it isn't even the decisive factor. What matters is that both systems expose their data in an accessible, documented way. A CRM and an ERP from two different vendors that talk to each other work better than two products from the same vendor that only share a logo. Check the technical documentation before you sign.
Can we integrate the online store as well?
Yes, and the logic doesn't change: you decide which system is in charge of each piece of data. Usually the online store generates the order, the ERP records it and holds the legal customer record, and the CRM receives the customer and their purchase history. The delicate point is the customer record: whoever buys online is often not the person the invoice is made out to.
How much time does it take from the company, not from the supplier?
Less than people fear, but not none. It takes a few hours of the owner's or the sales director's time for the decisions on master data, and one person inside the company available for the checks on the sample. The work you can't delegate is the deciding: no supplier can settle for you which system is in charge.
Better to change CRM or integrate the one we have?
In most cases it's better to integrate what you have. Replacing a CRM means migrating data, retraining people and rebuilding habits: a high cost for a problem that is usually about connection, not product. Replacement is justified when the system doesn't expose its data, or when your real sales process doesn't fit inside it.
In short
Half of Italian companies with at least 10 employees have an ERP, one in five has a CRM, and the two almost never talk. Integrating them isn't a technology project: it's the decision, taken once and written down, about which system holds which piece of information. Once that's done, the connection is the easy part.
If you want to find out where the flow between your systems breaks today, the first step I propose is an analysis: two weeks, from €2,500, fixed scope. It produces a map of how leads, orders and data move, the break points quantified, and a 90-day plan with priorities and costs. The document is yours and you can have anyone carry out the work.
To talk it through, get in touch. You'll find how I work and who I am on the rest of the site.
Sources
- Istat, Imprese e ICT — 2025, press release of 15 December 2025. Population: companies with at least 10 employees.
- Eurostat, E-business integration, 2025 data (updated May 2026). Country-level detail is not given on that page.
- Digital Innovation in SMEs Observatory, Politecnico di Milano, press release of 21 May 2026, 2025–2026 research edition.
A note on the figures: the percentages quoted are primary-source data with the year stated. The guidance on the order of the steps, on when integration isn't worth it, and on how costs are structured is the author's judgement based on direct experience, and should be checked against your own case.
Author: Sergio Selvelli, SE2marketing — marketing systems and operations for SMEs. Published 14 September 2026.