IDEAS | BLOG

How to Write a Good RFP

At Forum One we have thought a lot about RFP processes through the years. In addition to having participated in a several hundred RFPs, we have also helped many clients write RFPs and have also produced RFP guidelines for several large organizations.

RFPs, if conducted well, are useful in identifying vendors with appropriate skills and are (somewhat) useful in guaranteeing reasonable price and quality. On the other hand RFPs are very expensive. As a firm we spend up to $10,000 in time and expense for a full RFP process. Total costs for client time plus all vendors can reach six figures. RFPs which are poorly run can also result in poor outcomes and bad sentiments.

(It is important to remember that there are ways to meet the objectives of an RFP without actually doing an RFP — such as through contests, conducting small, sole-source trial projects or periodically using incumbents.)

What Makes a Good RFP?

There are many good, standard articles on “how to write an RFP” (found through a quick review of Google search results). Most propose an outline which includes the following:

  • Summary/Key Information
  • Background
  • Objectives
  • Requirements
  • Vendor Details
  • References
  • Budget
  • Schedule
  • Evaluation Criteria
  • Contact Information

This approach is typical and useful. From our standpoint, however, a few key important qualities are often neglected:

  • Objectives: The biggest single problem we see with RFPs is that they are short on overall objectives, but long on tactical details and requests. This often forces us to answer the wrong question (e.g. how to implement RSS feeds) yet not provide creative solutions for more global ambitions. A page or two describing objectives is far more useful than pages of requirements tables.
  • Budget: Many RFPs avoid providing any budget guidance. This leads to proposals that are not useful. We periodically respond to RFPs with our best ideas, are not offered budget guidance, and then are told our proposal is not in the right price range. Providing budget guidance has the theoretical risk of inflating otherwise less expensive proposals, but our experience is that we (and others) squeeze as much as possible into a proposed budget window. Budgets can always be negotiated subsequently.
  • Incumbents: Firms answering an RFP need to know if there are other groups with existing relationships with the client. Even if there are incumbents, firms may still decide to bid — but they should understand the playing field. It is important to avoid the perception that an RFP is simply used to justify a decision that has already been made.

What Makes a Good RFP Process?

There are a number of things to consider including when developing or answering a call:

  • Qualification stage: Because RFPs are expensive, we prefer processes in which firms (any firm or invited firms) are asked to send a brief letter or proposal for a qualification step. At that point 3-4 firms are invited to submit full proposals. Having a large number of firms submit long proposals wastes everyone’s time.
  • Face to face: When possible it is useful to have a face to face meeting with finalists. This helps level the playing field.
  • Dialogue: It is useful to have a dialogue with applicants throughout the process. Before proposals are submitted it is appropriate to have either a conference call or written questions that are shared. After proposal submission it is fine to negotiate deliverables and price, not assuming the proposal needs to be the final word.
  • Feedback: It is important upon completion of the RFP to provide feedback to both winners and losers. It is also important to commit to the process: we have participated in several exhaustive RFPs in which at the end the client decided not to do a program at all.

All of these objectives can be met in a relatively brief document. Many of the outstanding RFPs we received are five to ten pages. If you include these ideas in your next RFP, you’ll have happy applicants and set the stage for a successful project.