Importing spreadsheets

Lynqu reads the spreadsheets teams already keep, from a CRM export to a contract tracker with a title block and a totals row. It proposes where each column goes and shows you every record before anything is written. For 24 hours afterwards you can undo the whole import.

Start from Import on the Leads page, or Import contacts or Import companies on the Companies page. Managers and admins can import; employees do not see the button.

What it reads

• Excel (.xlsx, .xlsm), OpenDocument (.ods), CSV and TSV, and an HTML table saved as .xls. You can also paste a table copied from a spreadsheet.
• Up to 10 MB per file and 20,000 rows per sheet.
• CSV files in UTF-8, UTF-16 or Windows-1252, separated by commas, semicolons, tabs or pipes, with either decimal convention (1.234,56 or 1,234.56).
• Every visible sheet in a workbook. A sheet that only holds lists of allowed values is recognized as a list sheet: its lists become the options of any new list field, and the sheet itself is not imported.

Lynqu finds the header row even when a title or notes sit above it, and leaves out blank rows, totals rows, footnotes and repeated headers. If it guessed the table wrong, correct the header row or the first and last data row in the Structure step.

Formulas are never run. The value your spreadsheet last calculated is the value imported.

Mapping columns

Each column gets a destination, in this order:

1. A Lynqu field. Name, email, phone, company, website, industry, company size, deal value, currency, stage, close date, owner, and the deal terms below.
2. One of your custom fields, when nothing built in means the same thing.
3. A new custom field, only when the column has no home at all. The import proposes the type (text, number, date, list, yes or no, money or link), and an admin can accept, rename or drop it. Fields are created when you confirm the import, never before.

A column Lynqu works out from another one is left out on purpose and says why: a discount amount when the percentage is mapped, a “months in pipeline” count when the created date is mapped, or a start month that is combined with its year column. A column with no header is shown as unnamed so you can map it or leave it out.

A currency written in the cells (€ 90.000) becomes the deal currency, so it never needs a column of its own.

Phone numbers and dates

Phone numbers written without a country code (06 30 123 4567) are read as numbers from your organization’s country and saved in international form (+36 30 123 4567), so they match the same person however they were typed. A value that is not a phone number we can read is kept exactly as written and listed as a problem.

A date like 03/04/2025 can be 3 April or 4 March. When a column shows its order on its own (a day above 12 anywhere in it), Lynqu follows it. Otherwise it uses your organization’s date format.

Admins set the country and the date format in Settings → General. Each sheet in the Map step can change both for one import, for a file that came from a partner abroad, for example.

Deals and their terms

Deal columns can carry the commercial terms Lynqu stores on every deal: gross value, discount (percent, amount or free months), term, start date, guaranteed term, close reason and competitor. When gross and discount are both present the deal value is calculated from them.

Pipelines and stages in the file are matched to yours when the name is close to one you have. One with no match is offered as new and created when you confirm, unless you point it at an existing one; if you cannot create pipelines, you map it to an existing one. A won or lost deal lands in your won or lost stage. A won deal counts on its won date; when the file has none, you choose its start date or today.

Customers, sites and people

A row is read as a person, a company, a deal, or all three. A customer with several sites (plants, branches, stores) becomes one company with a child company for each site, one lead for the customer, and one deal per site on that lead. A row with a company and a deal but no person puts the deal on one lead for that company.

Review before anything is written

The Review step counts what is new, what matches a record you already have, and what needs a decision:

• Matches use the email, or the phone together with the name, for leads (the email only for contacts), the website and then the name for companies, and the lead, company, deal name and type for deals. A matching record only gains the values it was missing; nothing you already filled in is overwritten.
• Ambiguous rows (two companies with the same name, for example) wait for you to link one, import the row as new, or skip it.
• A row that matches a record outside your view is skipped, and that record is not shown to you.
• Problems (an invalid email, a date that could not be read, a value with no mapping) are listed per row. When the import finishes, Download rows with problems gives you a CSV of them to fix in your file.

Warnings appear when deals would count as won this period, or when deal values are in a different currency from your organization’s.

New leads from an import arrive quietly: no follow-ups, sequences, automation rules, webhooks or AI employee work start for them. A lead you already had that gains a value or a tag is an ordinary update, so rules and webhooks that watch for updates still run. The people new leads are assigned to get one inbox message, and you get one when the import finishes.

Undo

Undo import is available for 24 hours after an import finishes. It removes what the import created and restores the fields it filled in, in reverse order. Anything your team has worked on since stays: a lead with a note, a task, a call or a new deal is kept, and so is a company or custom field something else now uses.

Using AI on hard files

On Business + AI and Enterprise, when AI is switched on for your organization, Improve with AI (or Try with AI when Lynqu could not find the table) asks a model about the columns the rules could not settle.

• It sees sheet names, column headers, column types and up to three sample values per column. Samples are sent only for numbers, dates and columns that repeat a short list of values (a status or a type). Free text, names, emails and phone numbers are replaced with placeholders.
• It can only choose a Lynqu field or one of your existing custom fields. New fields are never invented by the model.
• Its answers are marked AI in the mapping, and you can change any of them.
• Each organization has 30 AI reads a day. Reading the same file again uses the saved answer.

When a file cannot be read

Lynqu names the reason (a password-protected workbook, no header row, no column that looks like a name, email, phone or company, an unsupported file type) and offers a template. The template has the columns Lynqu reads for that page, your custom fields, and an Allowed values sheet listing your pipelines, stages, member emails and list options. A filled-in copy maps on its own.

What Lynqu keeps

The uploaded file is deleted 24 hours after the import finishes, or after 7 days if it was never imported. The per-row record that makes undo possible is deleted after 30 days.

Importing team members

The Bulk Upload import on the Employees page uses the same reader, so a roster with a title above the header or saved as .xlsx reads the same way. See Team.