Digital platforms and artificial intelligence have dramatically reduced the cost of sales research.

Today, an SME can quickly identify hundreds of potential customers, distributors or partners. It can analyse their websites, find the right people to contact and collect information that, until only a few years ago, would have required weeks of work.

But finding company names is no longer the real problem. The real challenge is deciding which ones deserve attention, developing a credible approach for each of them and moving them forward consistently through the sales process.

Having 500 companies listed in an Excel file does not mean having 500 business opportunities.

From an export project to sales organisation

This issue emerged during the final stage of Export Booster, a project promoted by Confindustria Vicenza’s FarExport desk, in which I supported Lanulfi Moulds in developing its international business.

The planned activities had been completed: market selection, prospect research and qualification, initial contacts and sales visits. We still had two working days available.

The company decided to use them to address an organisational issue:

How can we better structure our internal sales process and integrate digital tools, automation and artificial intelligence where they can add value?

We did not start by choosing a software platform. First, we mapped the entire journey from a new contact to an order, identifying activities, responsibilities, the tools currently in use and the steps that could be improved.

From contacts to opportunities: selection, personalisation and method are essential.

The principle was simple:

Process + tools = results

First, you need to understand how the sales department works. Only then can you decide where to introduce a CRM system, automation or an AI platform.

Digitising an unclear process simply makes confusion move faster.

A company name is not yet a prospect

A company listed in a contact file is just a name. It becomes a prospect only when there are concrete reasons to believe that it may need our products or expertise.

Before contacting the company, we should therefore determine:

• whether it actually operates in the sectors we are targeting;
• whether it purchases products or services similar to ours;
• whether its size, applications and markets are compatible with our offering;
• which problem we could help it solve;
• who the most appropriate person to contact is.

Not all company names have the same value. It is therefore useful to classify them, for example as A, B and C prospects, based on shared criteria such as technical fit, economic potential, reliability, accessibility and likelihood of interest.

AI can speed up the collection and summarisation of information. The final assessment, however, must remain a commercial decision.

Standard emails are becoming even less effective

Once the prospect has been selected, a second issue arises: how should we approach them?

Many companies use standard introductory emails, copying and sending them to dozens of recipients while changing, at most, the name of the person and the company.

From the sender’s perspective, this is efficient. From the recipient’s perspective, however, it is usually obvious that the message is standardised.

Buyers receive many commercial proposals every day. With the spread of AI, producing formally correct messages has become easier, and the number of messages is likely to increase further. Gaining someone’s attention, however, is becoming increasingly difficult.

When making first contact, one simple rule applies:

You never get a second chance to make a good first impression.

Personalisation does not mean inserting the recipient’s name. It means studying the prospect and developing a realistic hypothesis about the value we could create for them.

Study the prospect before talking about your company

The Value Proposition Canvas can be a useful tool at this stage.

Before writing, we can try to understand the potential customer’s objectives, the problems they face and how our offering could provide a concrete benefit.

Information can be gathered from the company’s website, LinkedIn profiles, social media, published news, completed projects, target markets and technologies used.

The message will no longer begin with the usual “We are a leading company”, but with something relevant to the recipient:

We noticed that you are developing applications for sector X. In similar projects, we have helped manufacturers address problem Y through solution Z. Would it be useful to explore whether there may be an opportunity to work together?

This is not yet a sales proposal. It is a well-founded hypothesis that gives the prospect a reason to reply.

Of course, personalisation does not mean spending hours researching every company. The level of analysis should be proportionate to the prospect’s potential.

For a priority company, it may be worth studying its products, decision-makers and possible needs. For a lower-priority prospect, identifying one or two relevant elements may be enough.

AI can also help at this stage by collecting information and preparing an initial draft. It should not, however, be used to generate hundreds of messages automatically. Otherwise, we would simply replace the old standard email with a standard email that only appears to be personalised.

Every prospect needs a next action

If the recipient does not reply, what happens next? When will they be contacted again? Which channel will be used? Who is responsible?

Many opportunities are not lost because the prospect has said no, but because nobody has decided what should happen next.

Every priority prospect should have a defined stage in the process, a person responsible, a next action and a deadline.

The next action could be another email, a phone call, a LinkedIn connection request, the sending of a case study or a sales visit. It could also be a decision to pause the contact for a few months.

What matters is that it is a deliberate choice, not an oversight.

From a database to a sales pipeline

A database answers the question:

Who have we found?

A sales pipeline answers different questions:

Who have we qualified? Who have we contacted? Who has shown interest? With whom have we scheduled a meeting? Which opportunities require further action?

This is what transforms sales research into a manageable and measurable activity.

CRM systems, automation and AI can reduce manual work and provide reminders. But they cannot decide which prospects should be prioritised, what value should be proposed or who should carry out the next action.

For this reason, the question to ask the sales department should not be:

How many prospects do we have in our database?

It should be:

For how many priority prospects have we defined a next action, a person responsible and a deadline?

This is when a list of companies begins to become a real export sales pipeline.

Pier Paolo Galbusera