There are some things to consider when it comes to the columns when designing SharePoint 2010 solutions. Firstly, you want to minimize the number of columns that are to be filled in. The more columns that are offered the less likely it will be that a user will centime the metadata. They will only do so if there is a high value offered. Select the columns that need to be used for the system processing and navigating. There are some fields that should be required and others that will only be options. Required fields need to be used when they are part of the navigation process. This includes using the Content Query Web Parts for a given field. They need to be used for administrative purposes including the retention of actions on a date field that need to be specified by the user. The query for a single value column should be faster than multiple value columns. That is why you should try to make them all single value columns when possible. The total of the columns on a list can cause row wrapping. This will reduce the overall performance so you should minimize the number of columns to avoid using it as much as possible.
You may use columns to specify the type of metadata for an item. You can mark them as being hidden, required, or optional. When you use hidden columns for tasks that should be automated then users aren’t able to edit them. You should only use required columns when they are necessary. The fewer of them that you have the user fill out the easier it will be for them to navigate.
It is important to consider the tradeoff here when it comes to user tax versus metadata. The number of metadata columns increases. Then it is less likely that the users are going to fill in metadata due to the additional work to figure out which of the columns apply so they can give them a specific value. When you use a large number of required columns the user has to adopt a process that can be difficult and time consuming for them.
By keeping the process open and collaborated though you can use the value of the content to create something that is able to grow when content increases. Then it is more likely that the users will take the time to fill in the appropriate fields. This is very true for infrequent operations. The design phrase for you should consider what metadata is needed to perform the requirements of the operation.
This allows for the content to be retrieved. It also allows you to evaluate how long it’s going to take users to fill in the metadata. Then they can evaluate the effect for the user. End users don’t adopt the system because the overhead for creating content is very high. This can be difficult to reconstruct the system later on. This is due to the fact that the metadata may be used in a variety of ways.
The one consideration when it comes to the choice of columns is the type. It can be single field or multi-value. The queries that take place on the managed metadata field are going to be efficient when you compare them to the choice columns. You may wish to consider how to manage the metadata files instead of the choice fields.
Columns allow for the managed metadata and groups to support multiple values. Queries for multi value columns aren’t as efficient as they are for single value columns though. The columns and content types are commonly the central components for the classification and retrieval of the content in a large list.
The list of columns and content types should have been prepared during the planning stages. The number of columns and the content types that have been added can affect the overall performance in many ways. The number of columns for a specific type added to a single list will cause row wrapping. It is best to avoid this if you can.