Forum Discussion
Finding Matches in columns
Hi,
I am trying to locate the names that match in two different columns. I have tried to locate the matches muiltiple ways but always got nothing. I know there is matches beacuse I found some though Excel. I need to know which names are in both columns. Below is a few of the rows and the set up of my data. I tried adding a conditional column that stated "If (Account Name) equals (Account Name -) then Match" but didnt get any results. PLEASE HELP! I have been trying for a while now.
Working with comparing text is often tricky and you need to figure out what contraints you want to apply to your search.
For Example, a computer will consider "Bob" to be different to "bob".
It will consider "Bob " to be different to "Bob"
It will consider "Bob Smith" to be different to "Bob".
Since you know your data, what search contraints are you expecting to use? Making use of TRIM and UPPER inside your formula will allow you to get around the first 2 examples in my list. The SEARCH function will also help with the third, but is only useful if you know which column is going to be found consistantly inside the other column.
You should also expect that you will never get 100% accuracy, especially if this text is created by users typing into fields in your source system. Typos will always fall through the cracks. You will also likely find scenarios where you get false positivies because a short name is happened to coincedently exist in a much larger string of text. I.e. "Bob" inside "Michael Bobsled"
2 Replies
Working with comparing text is often tricky and you need to figure out what contraints you want to apply to your search.
For Example, a computer will consider "Bob" to be different to "bob".
It will consider "Bob " to be different to "Bob"
It will consider "Bob Smith" to be different to "Bob".
Since you know your data, what search contraints are you expecting to use? Making use of TRIM and UPPER inside your formula will allow you to get around the first 2 examples in my list. The SEARCH function will also help with the third, but is only useful if you know which column is going to be found consistantly inside the other column.
You should also expect that you will never get 100% accuracy, especially if this text is created by users typing into fields in your source system. Typos will always fall through the cracks. You will also likely find scenarios where you get false positivies because a short name is happened to coincedently exist in a much larger string of text. I.e. "Bob" inside "Michael Bobsled"
- Ashish_Mathur
Super User
Hi,
Share a dataset which can be pasted into MS Excel and for a few sample rows, please also show the expected result.
