Back to blogPractical Guides

From Excel to your first lead: import thousands of companies into the CRM in a minute

A real-life case with nearly eight thousand digitalising agents: upload the file, review the mapping, declare consent, tag, segment, score and launch the first sequence. All within the NAiOS CRM.

N
NAiOS.net Team
13 de septiembre de 20265 min read
Compartir:
Del Excel al primer lead: importar miles de empresas al CRM en un minuto
In this article
  1. 1. Choose the source
  2. 2. Review the mapping
  3. 3. Declare consent and tag
  4. 4. Confirm and view the summary
  5. 5. Segment and score
  6. 6. The first sequence
  7. What's done by the end

You have an Excel sheet with thousands of companies: name, CIF, province, categories, website, email, phone number. You downloaded it from a public registry, an association shared it with you, or you built it by hand. And there it sits, because moving it to a "real" CRM is usually a whole afternoon's work. With the NAiOS CRM, it takes a minute. Here is the complete walkthrough, using a real-life case: the digital agents from Kit Digital, nearly eight thousand rows.

1. Choose the source

In the CRM, go to Import (sidebar). You can upload the file directly, choose a sheet you already have in "My Documents" in NAiOS Office, or a scannable Excel sheet from your Knowledge. All three paths lead to the same place. Nothing is saved until you confirm.

2. Review the mapping

The importer recognises common headers in Spanish, Catalan and English: name, CIF, domain, email, phone, job title, address, company, province, notes. Anything it doesn't recognise is suggested as a custom field, with the type inferred from the data: text, number, date, or a list of options if there are few distinct values. In the Kit Digital Excel sheet, "Solution categories" and "Provinces of operation" end up like this, and are later used for segmentation.

If the file has both the company and the contact person in the same row, choose the company + contact mode: it creates the company by CIF (or by name if there isn't one) and a general contact with the email and phone number. If the company already exists, it doesn't duplicate it: it fills in the blanks and never overwrites what you already had.

CRM import screen with Excel columns mapped to fields and a summary of rows to be created
Each Excel column mapped to its field. Any leftover columns are converted into custom fields.

Before confirming, the importer asks you something other CRMs don't: what legal basis of consent these contacts have. Explicit, customer, one-to-one professional contact, or none. For a list of companies collected from a public registry, the honest answer is "one-to-one professional contact": you will be able to write to them personally from your inbox, but not add them to a mass newsletter. The platform records who declared it and when.

In the same step, you add the tag ("Digital agents"), the initial lifecycle stage (prospect) and the source. This way, in six months' time, you will know where each record came from.

4. Confirm and view the summary

The summary tells you how many rows will be created, how many will update existing records, how many are duplicates within the file itself, and how many are invalid, along with the reason. You confirm, and the CRM processes them in batches while you watch the progress. The nearly eight thousand rows in our real-life case took about eleven seconds. If something isn't right, the batch can be rolled back: what was created is deleted (anything that already had activity is archived instead of deleted).

5. Segment and score

Now the useful part begins. In Segments, you define filters that are evaluated every time you use them: tag "Digital agents", province Girona, category containing "Customer management", no activity in the last 90 days. The counter updates in real time. You save the segment, and it's ready to be used as the source for your sequences and campaigns.

Click "Score" and each record receives a score from 0 to 100 based on how well they match what you sell: related categories, priority province, having a website and email, recent activity. Sequences enrol those with the highest scores first.

CRM segment builder with filters by tag, province and category, and a live counter
A segment is evaluated every time: today it's 212 records, tomorrow it could be more.

6. The first sequence

In Campañas, you create a sequence with three steps: an introductory email, a four-day wait, and a second email for those who haven't replied, plus a call as a task. You enrol the segment; those who cannot enter (no email, "do not contact", on suppression list) are listed with the reason. The engine generates the drafts, you approve them, and they are sent from your NAiOS Mail account, with an hourly limit, within working hours, and with a two-week warm-up period. As soon as someone replies, their sequence stops, and they become a lead in the CRM.

What's done by the end

  • Thousands of companies and contacts in the CRM, with no duplicates, complete with source and date.
  • Custom fields ready to filter without you having to touch a thing.
  • A declared legal basis of consent per record, which determines which channel is opened.
  • A live segment, a score, and an active sequence.

The CRM and Campañas are available on all plans. If your file is different, the mapping adapts: try it with a sample of twenty rows, check the summary, and then upload the entire file. And if you'd like us to do it with you, get in touch.

Hashtags to share:

#NAiOS #IA #PracticalGuides #CRM #AIAgents

Compartir:

Related articles

NAiOS GIS: el mapa que te mandaron, abierto en un minuto y en 3D
Naios Functions

NAiOS GIS: the map they sent you, open in a minute and in 3D

A shapefile zip, a Google Earth KMZ, a GPX or an Excel file with coordinates: until today, you needed a desktop program and a technician to open them. GIS, the new module included in all NAiOS plans, converts them into layers of a project, calculates hectares and kilometres, shows them on a map in plain language and in 3D over the real terrain, and lets the chat read them as attachments. It was born from a support ticket and was on all platforms in a day.

11 de octubre de 2026
Read more
Soporte en 2026: un departamento entero, no un buzón
Naios Functions

Support in 2026: an entire department, not a mailbox

A mailbox stores messages; a department knows the status of each request, who is handling it, and who can see it. Support, included in any NAiOS instance, unifies form, email, Talk and chat into a single inbox, sets in advance who sees what, connects your platform with the NAiOS hub and lets AI do the triage while a person decides.

5 de octubre de 2026
Read more
NAiOS, explicado hoy: el chat con IA que ya incluye lo que tu empresa paga por separado
Naios Functions

NAiOS, explained today: the AI chat that already includes what your business pays for separately

What NAiOS is today: a multi-model AI chat and, with the same account, the native suites that a company pays for separately (office tools, email, meetings, customers, invoicing, accounting, people, training...). Table of which subscription covers each module, nine use cases by sector, the agents layer with a person in front, what NAiOS is not and how it is paid for.

4 de octubre de 2026
Read more
Habilidades: instrucciones que el chat aplica solo
Naios Functions

Skills: instructions that the chat applies on its own

New NAiOS module: 52 installable skills in ten categories, in open SKILL.md format and in six languages. They apply on their own when the request fits or with a slash in the chat, companies publish their own for the whole team and agents wear them. And how NAiOS Labs built it in one day: request, concept, harness, build, test and plating.

3 de octubre de 2026
Read more
El agente multiplica lo que ya sabes, y también lo que no sabes
Digital Transformation

The agent multiplies what you already know, and also what you don't know

One person with a coding agent closed 47 commits in a day. The figure is not the important thing: what matters is that writing code has stopped being the bottleneck and that the agent amplifies, with the same good appearance, both the judgement and the gaps of the person directing it. What to ask for, what it can touch, how far it goes, how we do it and why it applies equally to agents that do not write code.

2 de octubre de 2026
Read more

Did you enjoy this article?

Discover more content on our blog.

View all posts