Mail merge from .xlsx to .docx is removing special characters / symbols?

Question

I am on Windows 11 Enterprise and Microsoft Office 365 Enterprise.

I am doing a mail merge from Excel to Word, and it is replacing all my special characters with a question mark.

The steps I use are:

1. I export a report from software called LibCal which houses all the requests for patron programs for a large library system.
2. It exports automatically into .csv (the information looks correct at this point). The default font is Aptos Narrow 11.
3. I then save as .xlsx, rename the file, format, make corrections, etc. to get ready for my mail merge into Word (the information still looks correct at this point). The default font is Aptos Narrow 11.
4. I have a pre-made mail merge template with fields already set up to pull from the report. The default font is Aptos (Body) 10.
5. I run the mail merge in Word as letters, edit individual documents, view as an outline, create separate documents, then grab those and check the file names, etc. (the text is wrong at this point).
6. I use the documents to copy/paste into e-mail to all the stakeholders, put in a request to marketing for all the posters and flyers in their system (copy / paste into their form and upload a copy of the Word document), and then eventually put the information into the public facing calendar.

I set all this up because the person before me was typing all this manually, which is very inefficient, wasting enormous amounts of time and leads to errors from typos.

Now I found out that my special characters did not come across correctly, so mail merge is creating typos. I have already processed over 8,000 events for this year, so I really do not want to have to proofread every document that might have special characters because Word is altering information in the mail merge.

None of these systems talk to each other unfortunately, so I must do these steps, but the information must be right because potentially hundreds of thousands of people can view this calendar.

I know there must be a setting to tell Word to take the information as is, but I do not know what to do to make that happen.

I searched for this issue and already went into Word > File > Options > Advanced > General > and checked “Confirm file format conversion on open.”

I do not know what else to do to make this transfer correctly.

Here are a few examples from yesterday:

• Día de los Niños > D?a de los Ni?os
• Memory Café > Memory Caf?
• Pokémon Trading Card Game > Pok?mon Trading Card Game

This process functioned correctly throughout November and December, and I am not aware of any configuration changes on my end that would have affected its behavior. The example shown was generated using the same workflow during that period, and at that time the system successfully preserved several special characters and symbols during export.

• ¡Celebra el Día de Muertos en la biblioteca! Explora la tecnología creativa a través de proyectos prácticos para todas las edades. ¡Ven con su familia y amigos para explorar y crear juntos!
• ¡Viaja por las páginas de un buen libro! Acompáñanos cada mes a leer un libro en español y disfrutar de una conversación bilingüe. ¡Ya sea tu primer idioma o uno que estes aprendiendo no pierdas la oportunidad de visitar destinos nuevos o conocidos a través del español! Libros y snacks incluídos.

 Unfortunately, in processing 50 programs recently (January), I did not even realize Word did this and I sent emails and uploaded documents with corrupted text. I did not even think I would have to double check this type of thing. I do not know what I might have done to cause this to happen.

Any help would be appreciated.

Answer 1

Dear @Jennifer C,
Thanks for the update.

There are some steps can try:

1/ Try a different connection method

When Word asks how to open your Excel file, instead of choosing “OLE DB,” try selecting “MS Excel Worksheets via DDE (*.xls).” This method uses Excel directly to send the data to Word and might keep the special characters correct.

2/ Make a new mail merge template

Even if your current template is only a few months old, it might have saved some settings that are now causing problems. Try creating a new Word document, set up the mail merge again, and reinsert the fields. This can help fix hidden issues.

If these don’t work and the problem started after a recent update, it’s possible that a recent Office update may have caused this problem. You can roll back to an earlier version of Office that worked fine while waiting new update which may fix this issue.

To revert Office version, please follow the step form this article: How to revert to an earlier version of Office

I truly appreciate your patience and understanding. If you have any further questions or need further clarification, please feel free to reach out. I'm looking forward to hearing from you. 

Thank you for your cooperation.

Answer 2

Dear @Jennifer C,

Thank you for posting your question in the Microsoft Q&A forum. I understand you are running a mail merge from Excel (.xlsx) into Word on Windows 11 Enterprise with Microsoft 365, and some special characters (like á, é, í, ñ, ó, ü, Pokémon, Café) are being replaced with question marks.

To better assistance, could you please confirm following information:

• When exporting from LibCal, does the file come out as CSV first, or do you get a clean XLSX file?
• If CSV is involved, do you know whether it is encoded as UTF‑8 or something else?
• When you open the Excel file, do the special characters display correctly before mail merge?
• In Word, are you connecting the data source using OLE DB or the default file conversion dialog?
• Did Word recently start showing the message “Confirm file format conversion on open”?
• Do you use the same pre‑made merge template, or did this issue begin after creating or modifying a new version?
• Are multiple people running the mail merge, and do they all see the same character issue?

This issue usually causes by Word reading the file with the wrong text encoding during the merge. This can cause accented characters to appear as “?” when Word interprets text as ANSI instead of UTF‑8.

Therefore, I recommend some steps below:

1/ Avoid using CSV if possible
CSV is very sensitive to encoding changes. If LibCal can export XLSX directly, use that file for the merge instead of CSV. (CSV encoding issues are a known cause of special‑character corruption in mail merge.)

2/ If you must use CSV, save it as UTF‑8 before converting to XLSX

• Open the CSV in Excel
• Save As > CSV UTF‑8
• Reopen > Save as XLSX

This helps Word correctly detect Unicode characters.

3/ Use OLE DB to connect Word > Excel
OLE DB tends to preserve Unicode more reliably than text/dde-based connections during mail merge.
When Word asks how to open the file, choose: OLE DB Database Files

This method handles Unicode more reliably.

4/ Test with a new, clean mail merge template
Sometimes older templates keep outdated encoding connections. Creating a fresh Word document, reconnecting the data source, and re‑adding merge fields often stabilizes the character mapping.

Please understand that our initial response does not always resolve the issue immediately. However, with your help and more detailed information, we can work together to find a solution.

I truly appreciate your patience and understanding. If you have any further questions or need further clarification, please feel free to reach out. I'm looking forward to hearing from you. 

Thank you for your cooperation.

---

If the answer is helpful, please click "Accept Answer" and kindly upvote it. If you have extra questions about this answer, please click "Comment". 

Note: Please follow the steps in our documentation to enable e-mail notifications if you want to receive the related email notification for this thread.