In Excel, I very often use the filter dropdown menu to select which data points I want to show.
If anyone here hasn’t used it, you can basically click the filter button at the top of any column and it will give you a dropdown menu pre-filled with every unique data point that currently exists in that column. There is then a checkbox next to each unique data point, and selecting / deselecting it will immediately show or hide records that have that data point in that column. This makes it very easy to quickly filter a column for multiple criteria.
This also works for multiple columns at a time - e.g. I can show only records that have (Apple OR Orange in Column A) AND (Skyscraper OR House OR Bus Station in Column B), just by having only those data points ticked in each column’s filter dropdown.
So far, this seems a lot more clunky in Airtable for a few reasons:
I have to manually type in what I want my filter to be, which is inherently slower than simply clicking a checkbox, and means I have to remember or look up every one of the unique data points in that column (instead of having a pre-filled list)
The filter menu does not even really support filtering for multiple columns well, as you effectively only get one AND / OR divergence before the system doesn’t let you make more. (This is because there currently isn’t a way to distinguish between [Orange in Column A AND Banana in Column B] OR Apple in Column C and Orange in Column A AND [Banana in Column B OR Apple in Column C], for instance.)
Groups do seem to give more options on how many data points I filter down to and in what combination, but they won’t show only those data points, which still involves a bit of searching.
There may well be a formula or two that I could write to help me out here, but the fact remains that manually writing a formula to do this is always going to be a lot more tedious and error prone than simply using a dropdown list.
If I have missed some simple way of doing this already, please let me know! Otherwise this feature would be a huge time saver for me.|
EDIT: I should mention the workaround I currently have in case anyone is having the same problem!
Say I am trying to filter Column Fruit by multiple data points (and also want to have other filters in the table, so the current filter UI would be clunky to just have AND, AND, AND)… What I can do is link Fruit to a new table (Table 2), which will summarise all the different types of data currently in Fruit.
I can then create a new column in Table 2 called Quality, and “class” each of those data points by how tasty they are. e.g. Bananas, Apples, and Cranberries might have an entry in Quality saying “very tasty”, while every other fruit might have “not so tasty”.
Now when I go back to my original table, I can create a simple lookup column next to Fruit that will tell me whether that fruit is tasty or not, and I can then filter by that column. So all the actual filtering goes on in the second table, and in my original table, I just have a single column which essentially tells me if it would pass the multi-faceted filter or not.
The problem is that this is sitll tedious to create in the first place and modify if I want to change the filter, so would still appreciate a different approach.