When multiple Word, docx, or doc documents contain content that is similar in pattern but not identical, manually opening each file to delete them is very inefficient. This article uses HeSoft Doc Batch Tool as an example to demonstrate how to use the "Find and Replace keywords in Word" function with regular expressions to fuzzy match uppercase abbreviations, numbers, and other content, and leave the replacement content blank, thereby achieving batch deletion of keywords. It is suitable for processing template documents, course materials, contract files, and batch cleaning of sensitive fields.
When organizing a large number of Word documents, a common scenario arises: the content to be deleted is not a fixed word, but a type of text with a pattern. For example, documents may contain three-letter uppercase English abbreviations, numerical codes, lesson numbers, template identifiers, etc. Opening docx files one by one to find and delete them is not only time-consuming but also prone to omissions.
This article describes how to use the office software " HeSoft Doc Batch Tool " to batch delete keywords from multiple Word, docx, and doc documents using regular expressions. The core idea is: use formulas to find text fuzzily, matching a category of content; leave the replacement keyword list empty, allowing the software to directly delete the matched content.
Applicable Scenarios
This method is suitable for Word batch cleanup tasks where "the content follows a pattern, but the specific text is not completely identical," for example:
- Batch deleting uppercase abbreviations in multiple Word documents, such as three-letter English codes like BOT, ELA;
- Batch deleting numbers in docx documents, such as lesson counts, serial numbers, codes, monetary amounts;
- Cleaning up fixed-format markers, codes, and placeholders in template documents;
- Performing unified find-and-replace or deletion across a batch of Word files without opening them individually.
Compared to ordinary "exact text search," regular expressions are more suitable for fuzzy matching. For instance, [A-Z]{3} can match three consecutive uppercase letters, and \d+ can match one or more consecutive digits.
Preview of Effect: Before and After Processing
Before Processing: Multiple docx files need unified cleanup
In the example, there are 6 Word documents to be processed, all in docx format. Manual processing would require opening 1.docx through 6.docx individually and then searching for and deleting the content in each.

Opening one of the documents reveals content that needs cleaning in the body text, such as BOT in the title, ELA in the main text, and the number 60. This content is not necessarily identical in every document, but it follows certain patterns: three-letter uppercase words, consecutive digits.

After Processing: Matched keywords are batch deleted
After processing is complete, the content matching the regular expression rules has been removed from the Word documents. In the example, the three-letter uppercase abbreviations and numbers have been cleaned up, while the main document structure remains intact.

The advantage of this method is that you do not need to list every specific keyword in advance. As long as the keywords match the set rules, they can be found and deleted in batches.
Operation Steps: Batch Fuzzy Deletion of Word Keywords Using Regular Expressions
Step 1: Enter the Word tool and select "Find and Replace Keywords in Word"
Open HeSoft Doc Batch Tool , and select Word Tools on the left. In the feature list, find and click "Find and Replace Keywords in Word". This function is for batch finding, replacing, or deleting keywords in Word file content.

After selecting this function, the software will enter a batch processing wizard page. Follow the subsequent steps to add files, set search rules, configure the save location, and start processing.
Step 2: Add the Word, docx files to be processed
After entering the function page, in Step 1 "Select records to process," click Add Files to add the Word documents needing processing to the list. If the files are concentrated in a single folder, you can also use Import files from folder.

Once added, the list will display the file name, path, extension, creation time, and modification time. The example shows 6 docx files imported, indicating they will all participate in the subsequent batch find-and-replace task. After confirming the files are correct, click Next at the bottom.
Step 3: Select "Use formula for fuzzy text search"
In Step 2 "Set processing options," locate the "Search method." To use regular expressions for fuzzy matching, you should select Use formula for fuzzy text search, rather than ordinary exact text search.

This step is critical. After selecting formula fuzzy search, the software will match Word content according to the entered expression rules, rather than looking for a specific fixed term.
Step 4: Fill in the regular expressions for content to be deleted
In the Keyword list to find on the left, enter the matching rules. The example includes two rules:
- [A-Z]{3}: Matches three consecutive uppercase English letters, e.g., BOT, ELA;
- \d+: Matches one or more consecutive digits, e.g., 60, 100, 2026.
On the right is the Replacement keyword list. The prompt in the interface says "leaving blank means delete," so if the goal is to batch delete the matched content, leave the replacement list on the right empty.
In other words, the processing logic for this example is: find all text matching [A-Z]{3} and \d+, and then replace it with empty content, thereby achieving deletion.
Step 5: Proceed to the next step, set the save location, and start processing
After completing the rule settings, click Next. Follow the wizard interface to continue to the "Set save location" and "Start processing" steps. It is recommended to save the processed files to a new directory, making it easy to distinguish them from the original Word files and convenient for verifying the results later.
Once processing starts, the software will sequentially perform the find-and-replace operation on the multiple added Word documents. After completion, check the output file to see that the keywords matching the rules have been batch deleted.
Common Questions and Considerations
1. Why does an empty replacement keyword list mean deletion?
In find-and-replace logic, "replacing with empty" is equivalent to deleting the matched content. The prompt in the right area of the screenshot also says "leaving blank means delete," so as long as search rules are filled in on the left and the replacement content is left empty on the right, batch deletion can be achieved.
2. Could the regular expression accidentally delete content?
It is possible. Regular expressions match a type of rule, not a single fixed word. For example, \d+ will match all consecutive digits; if the document contains numbers like serial numbers, dates, page numbers, or amounts, they may also be deleted. It is recommended to test with a small number of files before formal processing.
3. What content is [A-Z]{3} suitable for deleting?
It is suitable for deleting three consecutive uppercase English letters, such as BOT, ELA, ABC. If the document contains other three-letter uppercase abbreviations that also fit the rule, they will be deleted as well. To delete only a few specific words, consider using more precise rules or ordinary exact search.
4. Is it necessary to check "Ignore letter case"?
If you only want to match uppercase letters, it is generally not recommended to check "ignore case." Checking it might expand the match scope, causing lowercase letter combinations to be recognized as well. Choose carefully based on the document content during actual operation.
5. Should I back up the original files before processing?
Backup is recommended. A characteristic of batch processing is that it affects multiple files at once. Especially with regular expression deletion operations, improper rule settings can cause unnecessary content loss. Saving the results to a new folder is a safer practice.
Summary
Using HeSoft Doc Batch Tool , the deletion tasks that originally required opening Word, docx, and doc files one by one can be transformed into a single batch processing task. Through the "Find and Replace Keywords in Word" function and selecting "Use formula for fuzzy text search," you can use regular expressions to match a category of keywords; leave the replacement content empty to batch delete these contents.
If you often need to clean numbers, abbreviations, template markers, or sensitive fields from a large number of Word documents, you can prepare the matching rules first, test with a few files, and execute in batches once confirmed correct. This reduces repetitive work and significantly improves document organization efficiency.