Below is a list that explains exactly what each criteria operator means and how it works. 

Text inputs are not case sensitive. "Gold" and "gold" would give the same outcome. 

Operator                                              Use Case                                                                   Text Input Example

Adjust by daysAdd or subtract daysN/A
all the productsConsider all productsN/A
any of the productsConsider any one or more than one productN/A
begins withBegins with the exact string.

Eg. You want to tag a SKU that begins with "PRD".
This should not be used for multiple inputs. 
"string"
containsMatches the exact string at any position

Eg. You want to tag an order that contains a product that has "Necklace" in the product title. If you have a "Silver Necklace" in the order, it would be tagged.
This should not be used for multiple inputs.
"string"
does not begin withShouldn't begin with the exact string

Eg. You only want to tag orders where the product title doesn't begin with "Silver".
This should not be used for multiple inputs. 
"string"
does not containShouldn't match the exact string at any position

Eg. You don't want to tag anything when the order contains a product that has "Chain" in the product title.
This should not be used for multiple inputs.
"string"
does not includeShouldn't exactly match any of the comma-separated values

Eg. You want to tag orders that don't include SKUs "SKU1, SKU2, SKU3". Inverted Commas are not needed either. Multiple inputs can be given. 
"string1, string2, string 3"
excludeSelected values aren't considered

Eg. You want to tag orders that don't include two specific products.
single/ multiple items can be selected
greater thanShould be more than the value

Eg. You only want to tag the order when the total order cost is greater than (>) 100.
value
greater than or equal to Should be more than or equal to the value

Eg. You only want to tag the order when the total order cost is greater than or equal to (>=) 100.
value 
ignore items not tracked by ShopifyDon't consider products that aren't tracked by Shopify N/A
in all*Should match in all values

Eg. You want to tag orders where each product in the order should be in every selected collections.
single/ multiple items can be selected
includesMatch any of the comma-separated values

Eg. You want to tag orders that include SKUs "SKU4, SKU5, SKU6".  Inverted Commas are not needed either. Multiple inputs can be given.
"string1, string2, string 3"
Inventory state: available Products inventory state is availableN/A
Inventory state: on handProducts inventory state is on handN/A
isMatch exact / selected  value  

Eg. You want to tag orders where the customer's name is "John Doe"
This should not be used for multiple inputs.
"string" 
is emptyMatch null valuetextbox is disabled
is equal toShould be equal to the value

Eg. You only want to tag the order when the total order cost is equal to (=) 100.
value
is notShouldn't match exact / selected  value

Eg. You want to tag orders where the customer's name is not "Jane Doe"
This should not be used for multiple inputs.   
"string"
is not emptyMatch not null valuetextbox is disabled
less thanShould be less than the value

Eg. You only want to tag the order when the total order cost is less than (<) 50.
value
less than or equal toShould be less than or equal to the value

Eg. You only want to tag the order when the total order cost is less than or equal to (<=) 50.
value
not in all*Shouldn't match in all values

Eg. You want to tag orders where atleast one of the products in the order is not present in atleast one of selected collections.
single/ multiple items can be selected
only count physical productsOnly consider physical products N/A
only count unfulfilled productsOnly consider products which unfulfilled in orderN/A
PrefixAdd text before the tag

Eg. Adding "Payment_type" before the dynamic tag that adds the type of payment used in the order. 
string
SuffixAdd text after the tag

Eg. Adding "Payment_type" after the dynamic tag that adds the type of payment used in the order. 
string 


 

To further clarify the operators "in all" & "not in all" here are some cases. Suppose there are two products and two collections.

case 1: one product is in one collection & other product is in other collection.

case 2: both products are in one collection only.

case 3: both products are in both of the collections. (in all)

case 4: both products aren't in any of the collections (not in all)