And also when is useful to use each one of them?
This is a good question as you will find people have mixed responses about this.
The way I have viewed this and multiple agencies and companies I have worked at also view this is: User profiling is a collection of data from analytics that you use to create a set of profiles based on data and behaviour on your site.
So for example say I have an e-commerce site, I could look at data/analytics that shows me three groups of customers: high spenders, repeat customers and passive customers, within those three profiles I could then analyse what amount of my customers sit within those, their average age, their average spend, etc and come up with something like this (example):
- High Spenders: Average order value £90, average yearly orders 12, average age 45-50.
- Repeat Customers: Average order value £40, average yearly orders 4, average age 35-50.
- Passive Customers: Average order value £50, average yearly orders 1-2, average age 28-32.
This obviously is quite vague information but gives you an idea and a goal for your objectives, which could be getting more Passive customers to be Repeat customers and then maybe doing a marketing campaign or some A/B testing etc to reach this goal.
Where as personas describes a representative (but fictional [Based on some comprehensive site data!]) individual rather than an entire group, a persona would have specific details of age, interests, location, similar brand interests, etc. This is great for writing things like tone of voice and defining your customer to your business or site.
So their is a big difference between the two but both can co-exist and complement each other in your user research if you understand the difference.
A good source I found that also gives you some insight in to this and that I feel explains this best is here: http://www.tandemseven.com/blog/benefits-of-using-personas-and-user-profiles-together-8/