Since the Department for Communities and Local Government issued a consultation on City Development Companies (CDCs) in December 2006, cities and city-regions across England have expressed strong interest. The intention, as announced by Ruth Kelly is for cities to use new, specialist agencies to more clearly define responsibility for economic development, more easily pursue longer-term objectives, and draw together the energies and resources of a wide range of public and private sector partners. The overall expected impact is to increase business confidence and speed up decision making, leading to better and higher levels of investment overall. This all sounds promising - but how could CDCs best be used?

Special purpose vehicles are nothing new in England - and CDCs don't even require new legislation. Yet the recent push to develop CDCs should be welcomed. The Government's proposals renew awareness of special purpose vehicles and 'development agency' models - which have often been successful in the past. With strong Whitehall backing, the potential for CDCs to overcome the limitations of local authority and regional boundaries is very real.

A number of local authorities in England are already actively pursuing the CDC model, including Plymouth City Council and Hull City Council, amongst others. Newcastle and Gateshead City councils are actively considering a collaborative CDC. There are also recent examples of CDC-type agencies being set up - including Creative Sheffield (est. 2005) and The Manchester Enterprises Group (est. 1999).

And if we go further back into our institutional history, we'll find a useful track record to learn from - including agencies such as Enterprise Boards (and their successors); Enterprise Trusts; Enterprise Agencies; Business Links; Training and Enterprise Councils; and even New Town Development Corporations. If we look overseas, we find even more examples.

So what can we learn from past experience? The rest of this article offers a few points of guidance to consider before setting up your CDC.

Your CDC: A 10-point plan for English cities

1. Firstly, and most importantly - concentrate on what you want your CDC to achieve, rather than on what structures you want to create. Structures and reorganisations in themselves are not the answer. You need clear focus on the remit and rationale for a CDC - form should follow function.

2. Ask questions. What can a CDC achieve that existing institutional structures cannot? Your CDC should focus on economic goals that existing agencies fail to deliver. For example, the CDC could be a focus for multi-area or multi-agency collaboration, and can better engage the private sector.

3. Keep the mission specific. Organisations with vague or insufficiently defined roles and goals fail. It's as simple as that.

4. Focus your CDC on a specific role or area. There's a danger that CDCs will become overloaded with multiple functions. And there's a knock-on danger of pushing expectations far too high. Remember that CDCs are new, and may be risky. Why not start with a narrow range of functions or a small area first to see if the approach is successful?

5. Ensure sufficient 'arms length' operational independence. CDCs need this to avoid being political footballs, and to achieve the most effective methods of working. Set them priorities and tasks, but leave them work out how best to deliver them. Ideally, CDCs will become entrepreneurial and work effectively with the private sector.

6. Appoint leaders that are pragmatic, commercially astute, and skilled political operators. CDC leaders need to make tough decisions, based on reasonable rationales and excellent evidence. They also need to be very well networked, building political capital with all kinds of stakeholders. And they need to know their market and build core competencies too.

7. Create Governance structures that build organisational capacity. Democratic legitimacy is useful, but so is having board members that add practical value and expertise to the CDC.

8. Beware of giving your CDC too much policy and strategic leverage. It is better that local authorities and stakeholders set out policy aims and priorities, and the CDC is given significant autonomy to decide how best to deliver these.

9. Evaluation and monitoring should be built in. You are only really going to know if the CDC has added value if you establish proper performance monitoring and evaluation procedures from the outset.

10. Allow your CDC to retain surpluses and build assets for non-profit use. When public revenue support expires or is cut, an asset base can provide an income stream, and allow further diversification and growth. Successful use of this has been made in Greater London Enterprise and Glasgow's Govan Initiative.

Lastly - before they get better, things may get worse. It takes a while for new organisations to 'bed in', internally organise themselves, staff up, and forge the right relationships. And it is likely that your CDC will face some very challenging tasks that are new and for which there is no instruction manual. You should be starting to seriously evaluate success after two- to- three years, not after six months.