To OCR a table to Excel, you can use Lido. Lido runs OCR on scanned PDFs and images, detects the table's rows and columns, and exports it to Excel with every value in the right cell.
In this tutorial, we will walk you through how to OCR a table to Excel using Lido. Simply follow the steps below.
Lido's PDF Table Extractor is the best way to OCR a table to Excel because it finds the table structure first, then reads each cell, so columns and rows stay aligned instead of running together.
Before you start, make sure you have your scanned PDF or image saved on your computer. Phone photos of printed tables work too.
If you don't have a Lido account yet, you can create one for free here.
Once you're logged in, create a New File.
Click on Tools in the top menu and select the PDF Table Extractor. This tool runs OCR on your file and keeps every table's rows and columns in place.
Drag and drop your file into Lido, or click to upload. Lido runs OCR automatically, so there's no need to convert the scan or image first. The OCR step is what reads the text inside a PDF image or photo.
For the purpose of this tutorial, we are using a sample scanned invoice. Drag the selection box over the table you want in Excel, such as the line items table. If the table continues across several pages, check Apply selection to all pages.
Once you have your selection, click Extract Data.
Review the extracted table to make sure it includes everything you need. You can go back to adjust your selection. When you're satisfied, click Insert at Active Cell to add it to your spreadsheet.
You can edit and format the cells however you need. Adjust column widths, format numbers as currency, or reorganize the data to suit your workflow.
Click Export to Excel at the top to download your table as an .xlsx file, ready to sort, filter, and calculate.
Standard OCR reads text left to right, so it often scrambles tables. Here are the most common problems and how Lido handles them:
Borderless tables: Without grid lines, basic OCR merges or splits columns. Lido uses the alignment of text across rows to find the correct column boundaries.
Merged cells: Headers that span several columns break the grid in most tools. Lido recognizes merged cells and keeps the data under the right headers.
Multi-line cells: Text that wraps inside a cell often turns into extra rows. Lido keeps wrapped text in a single cell.
Multi-page tables: Most tools treat each page as a separate table. With Apply selection to all pages, Lido extracts the table from every page in one step.
Misread numbers: Basic OCR can confuse 0 with O or 1 with l. Lido reads characters in context, so numbers stay numbers. For more on what affects results, see our guide to OCR accuracy.
These methods can move data from scans into Excel for simple tables, but each has limits:
Free online OCR tools: Sites like OnlineOCR.net and i2OCR read text from uploaded images. They're quick for one simple, bordered table, but they often lose column structure and upload your file to a third-party server.
Adobe Acrobat Pro: Export PDF runs OCR and exports to Excel. It handles bordered tables well, but it requires a paid subscription, struggles with borderless tables, and splits multi-page tables by page.
Python libraries: Developers can combine Tesseract with table detection libraries to build a custom pipeline. It's flexible, but it takes significant coding and tuning for each table layout.
Lido works in every scenario: scanned PDFs, phone photos, borderless tables, merged cells, and multi-page tables. It's also SOC 2 Type II and HIPAA compliant, so sensitive documents stay secure. For more options, see our guides to the best image to table converters and the best PDF to Excel converters.
Now that you know how to OCR a table to Excel, you can turn scanned tables into data you can sort, filter, and calculate in minutes.
Open Lido's PDF Table Extractor, upload the scanned PDF or image, and drag the selection box over the table. Lido runs OCR, keeps the rows and columns in place, and lets you export the table to Excel.
Lido is the best OCR for tables because it detects the table structure before reading each cell. Borderless tables, merged cells, and multi-page tables come out aligned instead of scrambled.
Basic OCR tools often merge or split columns in borderless tables. Lido finds column boundaries from how the text lines up across rows, so borderless tables convert cleanly.
Accuracy depends on the tool and the scan. Basic OCR works on simple bordered tables but struggles with complex layouts. Lido reads text in context and achieves 99%+ field-level accuracy, and scanning at 300 DPI or higher gives the best results with any tool.
Yes. Free online OCR tools like OnlineOCR.net handle simple, single-page tables but often lose column structure. Lido offers 50 free pages so you can test it on your own scanned tables.
%20(1).svg)
.png){kind=small}
