OPM3 may help but is also very complex. Rather than start with this or Unified Team Management or any other methodology, I'd suggest starting with some simple questions, such as:
What are the business objectives of the PMO? Are they still the right ones?
How is project success being judged? Who is judging project success?
What appear to be the main issues? What different results are desired?
What projects succeeded, and why do the people who managed them think they succeeded? This might be your best source of knowledge and ideas on how to proceed - ask for the opinions of those who have had success.
The number one reason for project failure, according to my colleague Frank Winters in his series of Gantthead articles on project failure, is poorly trained or inexperienced project managers. I agree with this judgement, so suggest you try and assess the competence of the project managers. I would venture to state that if the company has great project managers, then almost no matter how poorly run the PMO, most projects would be deemed a success.<\b>
