One trend observed is that the so called "Identity Managers", e.g. the provisioning products, are constantly growing in functionality - and complexity. This isn't surprising. There is strong competition between vendors and thus many vendors try to add all the functions which are offered by other vendors. The customers as well expect very complete products. But there are two things which should let us think about this strategy:

  1. The increasing complexity: Thus it really make sense to create more and more complex products?
  2. The still existing weaknesses: In many areas there are better solutions available as separate products than are implemented in most or all provisioning products. Have a look at business role management, GRC (Governance, Risk Management, Compliance) functionality, or workflows.
Besides this, there is not just one user group which has to deal with identity management. There are departmental managers which have to do some attestation and to invoke workflows. There are the persons which act as interface between IT and the rest of the organization which, for example, have to deal with the translation of business roles into system roles. There are technical administrators of the connected systems. With other words: There are several levels within the organization which have to be adressed - and there are several technical layers.

I personally don't believe that more and more complex provisioning products are the best answer for the customer's requirements. In contrast, a modular approach with defined interfaces and defined responsibilities would suit much better in most cases, especially in the larger companies. For smaller companies, a one-stop-solution might be appropriate. But in that case it has to be one which is pre-configured and easy to use, something which isn't delivered today.

My expectation is that the market will change, with vendors who offer modular solutions (or just some modules) in a service-oriented architecture and others, who focus on the midsize market with integrated products. But todays approach to put more and more functionality (business role management, auditing,...) into a technical product will fail. Like yesterdays "Enterprise Systems Management Frameworks" have failed.