Well, there is not an "easy" way of creating the sub-territories. If you are basing your territories off of postal codes you can simulate sub-territories by building the right queries. You will need to layer your data.

Let's say you want to see a count of clients by sub-territory within a larger territory as well as see the client locations by address.

First import your main territory (by zip/postal code). Then have a second query that groups ZIPs or postal codes into your sub-territories. Import this second query as a shaded area dataset. Plot the data by zip, not by territory. Your third query could plot client location by address. Plot this as a pushpin dataset and use a symbol like a large or small square. Build a fourth query that sums your client count by zipcode or assign one zip in each sub-territory as the "display" zip. Use a count of client id to get the total number of clients by zip or by your display zip. Import this query as a shaded circle dataset.

Downside to all of this is the limitation on the unique values that can be imported at once. I have had to split this into multiple queries and datasets as well as add some text boxes and symbols of my own.

Sometimes you just have to get creative with how you push your datasets around. Let me know if you want to see a small sample of how the output and datasets might look.