Merged results from three map sources
Lists built from Google, Bing and Apple Maps overlap. Here the same pottery appears twice with a trailing space and a differently capitalised email, and one print shop was saved twice from the same source.
With no column ticked, only the exact repeat of Marlow Print Works goes, because the two pottery rows differ in the source column. Ticking just email removes both repeats: comparison is case-insensitive, so [email protected] matches [email protected]. Ticking business and town gives the same result here, which is a good sign the key is right.
business,town,email,source Kiln & Co Pottery,Frome,[email protected],google maps Kiln & Co Pottery ,Frome,[email protected],bing maps Marlow Print Works,Bath,[email protected],google maps Oak Lane Joinery,Frome,,search Marlow Print Works,Bath,[email protected],google maps Pip & Parsnip Deli,Wells,[email protected],apple maps
No columns ticked (whole row): 1 duplicate removed email ticked: 2 duplicates removed business + town ticked: 2 duplicates removed Kept: Kiln & Co Pottery, Marlow Print Works, Oak Lane Joinery, Pip & Parsnip Deli
De-duplicating a file
Choose the CSV
Pick a .csv file of up to 25 MB. It is read by your browser; there is no paste box, so export from your spreadsheet first if the data is not already a file.
Confirm the header
Leave The first row is a header ticked if row one holds column names. It is then kept at the top and never compared.
Pick the matching columns
Tick the columns that identify a duplicate. With none ticked, two rows must match in every column. Each ticked value is trimmed and lower-cased before comparing.
Check the count and download
The panel shows rows read and duplicates removed. Download the clean file saves a copy named after your original with -deduplicated added.
Which columns to tick
| You want one row per | Tick | Watch out for |
|---|---|---|
| Email address | Rows with a blank email are treated as one value, so only the first blank row survives | |
| Business at a location | business name and postcode (or town) | Chains: every branch has the same name, so include the postcode |
| Website | domain or website | http and https or www variants are different text; tidy them first |
| Phone line | phone | Different formats of one number do not match; format phones first |
| Exact record | nothing | Any difference, even a source label, keeps both rows |
Exact matching and its limits
This tool compares text, not meaning. It catches the common, cheap duplicates: the same row exported twice, the same email in two lists, the same name and postcode saved from two sources. It will not see that Kiln and Co Pottery Ltd and Kiln & Co are the same firm, or that 01632 960410 and +44 1632 960410 are one number.
For those, normalise first: format phone numbers to one style, reduce websites to their bare domain, and run company names through a fuzzy matcher that groups near-identical spellings for you to review. Then de-duplicate on the cleaned columns. Removing duplicates before outreach is not just tidiness; it stops one business getting the same email twice and keeps your reply and bounce rates honest.
Questions people ask
Which duplicate does it keep?
The first one in the file. Later rows that match on the columns you ticked are dropped, and the kept rows stay in their original order. If one copy has better data, such as a phone number the other lacks, sort the file so the most complete rows come first before you de-duplicate.
Is the comparison case sensitive?
No. Each value in the columns you tick is trimmed of leading and trailing spaces and converted to lower case before it is compared, so [email protected] and [email protected] count as the same. Spaces or punctuation inside the value still matter, so Smith & Co and Smith and Co are different.
How do I remove duplicates based on one column only?
Tick just that column, for example email or website, and leave the others unticked. Any row whose value in that column has already appeared is removed, even if the rest of the row is different. Be careful with columns that have blanks, because every blank counts as the same value.
What is the largest file it can handle?
Files up to 25 MB are accepted, which covers most lead exports from a spreadsheet or CRM. Everything happens in your browser's memory, so very wide files with many columns take longer. For bigger exports, split the file first or remove columns you do not need.
Where your data goes
Your CSV is opened with the browser's file reader and compared on this page. The rows are never uploaded, and the clean file is built and saved from your own device.
Checked against the tool on