How I clean a messy contact list with AI (a real example)

A client sent me a "contact list" like this:

张三 13800138000 北京市朝阳区建国路88号
李四 139 0013 9000 上海市浦东新区
张三 13800138000 北京朝阳区建国路88号
王五 电话未知 广州市天河区
李四 13900139000 上海浦东

Duplicate rows, inconsistent phone formats, and one entry with a missing phone number.

I pasted it into Snapnote (the tool I built) in data cleaning mode. Here is what came back:

Name	Phone	Address
张三	138-0013-8000	北京市朝阳区建国路88号
李四	139-0013-9000	上海市浦东新区
王五	(unknown)	广州市天河区

Plus two sections I did not expect to care about, but now do:

What changed

• 1 duplicate removed
• phone numbers normalized to 1xx-xxxx-xxxx
• two 李四 records merged (different phone formats, same person)

Needs human review

• 王五 has no phone number
• 李四 had two address variants

The important part for me: it did not invent the missing phone. Missing stayed missing, and it told me so.

That is the whole idea — messy in, clean out, plus an honest report of what it could not fix.

Try it: https://getsnapnote.com