Building an RFP for Business Continuity Services
27 August 1999
Simon Mingay
 
We outline what needs to go into an RFP when assessing potential providers of BC and disaster recovery services.

 Strategy & Tactics/Trends & Direction
Note Number:  TU-08-8887
Related Terms:  Sourcing Strategies; Business Continuation Services; Business Continuity
Download:  PDF 

Building an RFP for Business Continuity Services

We outline what needs to go into an RFP when assessing potential providers of BC and disaster recovery services.

Bottom Line

Key Issue
What process should an enterprise follow to select the best ESPs for its specific needs in Europe?

We outline a framework for developing RFPs for BC services, which are becoming increasingly complex.

Mission and scope: The RFP is used to detail the solution an enterprise wants to implement, and to evaluate vendor proposals in a systematic and consistent manner (see Figure 1). The RFP is not used to gain education on potential solutions or vendors; this is achieved through an RFI. The RFP is used to help qualify the ESP and the ESP’s proposal. RFPs should be detailed, and therefore are long. While the RFP is part of the legal document trail, nothing is final until a contract is negotiated and signed.


Figure 1
The Acquisition Process
%img

Source: GartnerGroup

Control and distribution: There should be extensive internal reviews and approvals of RFPs (see Note 1). Copies of all RFPs should be maintained in a central location for reference. Version control is important as a formal means of document tracking, and ensures that users and vendors are working from the same document. Distribution should be selective and based on market research and an RFI process. Only vendors that can provide a quality solution should be included. The number of vendors should be limited to three.


Note 1
Review
The review should involve representatives from groups with responsibility for each component of the solution, as well as procurement. Some organizations may also need to involve financial controllers, legal council and internal auditors to comply with internal policy.

Position in the acquisition process: The RFP should be sent out only after the BC plan contains sufficient detail to enable the detailed specification of requirements (see Note 2), or, at the very least, when there is a clear idea about what needs recovering and when. Vendor-proposal evaluation and additional investigation (e.g., calling references and performing financial due diligence) will determine which ESP’s solution will be selected. This information is then put into the recommendation. Below is a basic RFP checklist:


Note 2
Just a List of Equipment
The IS group is frequently under time pressure. The result is that it simply passes the list of equipment it has to the vendor and invites a proposal. This usually results in the enterprise spending four to 10 times more than it needs as a result of covering too much equipment in too tight a time schedule.

  • Executive summary (one-page maximum)
  • Introduction: 1) Background on the corporation and/or business division. 2) Scope of services being requested, e.g., consulting, hot-site services. 3) Background on the BC process including current status, existing roles and teams. 4) Statement on the confidentiality of information.
  • Overview: 1) Statement of mission/vision. 2) Statement of business objectives. 3) Statement of scope in terms of which business processes, business units, applications, packages, geographies and technology platforms are being covered. 4) Role of the BC vendor.
  • Project schedule: 1) Vendor RFP question deadline. 2) Vendor analysis meeting (optional). 3) Proposal due date. Vendors need four weeks to respond well to anything other than simple configurations. Less time results in poorer, less-innovative and probably more-costly solutions. 4) Vendor demonstration day or recovery site visit (if appropriate). 5) Contract negotiation. 6) Final decision. 7) Proposed implementation start.
  • Business requirements: 1) Detailed business requirements including RTOs and RPOs for each business process (see Note 3). 2) Detailed technical requirements, describing the applications, platforms and communications requirements for each business process in terms of the normal operational configuration. A description of dependencies. 3) A recovery time line outlining how resource requirements change during at least the first 14 days (see Note 4). 4) Invite the vendor to propose a validation program, but include a best guess at test-scheduling requirements (see Note 5). 5) Implementation, training, start-up and maintenance requirements. 6) Project management and service-level-reporting requirements. 7) Performance criteria for success of solution. 8) Indication of performance penalties.

Note 3
RTO and RPO
RTO is the time window within which the process or system must have its minimal functionality. RPO is the point in time up to which consistent data can be recovered.


Note 4
Recovery Time Line
Typically some 20 percent to 40 percent of capacity is required within the first 48 hours, with the rest coming online in subsequent weeks. Demanding full capacity within the first 24 or 48 hours reduces the flexibility of the vendor to create a more suitable response and increases cost substantially. A recovery time line describes how capacity requirements change over time.


Note 5
Test Scheduling
Test slots are the resource bottleneck for BC vendors. The RFP should lay out the requirements for testing, including first-year requirements, which platforms need to be tested together, and for how much time. The RFP should specify the deadline for having a working recovery process.

  • Proposal requirements: 1) Contents of proposal, including contractual terms and conditions, references and names of key ESP personnel (recommended where consulting work will be included). 2) Proposal format and number of copies (insist that the vendor follows format). 3) Length of contract (see Note 6). 4) Guidelines on how to present costs; this should include a requirement that pricing is broken down (see Note 7). 5) Requirements for "proof of concept." 6) Documents to establish vendor financial stability. 7) Who to contact with questions and where to send proposals. 8) Selection criteria (see Note 8).

Note 6
Length of Contract
Contracts should be kept to three years. Vendors make most of their money toward the end of the contract as their cost of operation per MIPS, per gigabyte and so on decreases year on year. For the client enterprise to realize this saving it needs to keep the contract to three years and then enter another tendering process.


Note 7
Price Breakdown
Prices should be broken down by each category of service (workspace, network services) and for the technology services by platform group (e.g., AS400s, HP9000s). Where reasonable to do so, the RFP should insist on defining costs of adding capacity, e.g., DASD capacity, seats for workspace recovery, mainframe MIPS. All charges should be explicitly stated, e.g., invocation, occupancy, communication, additional test time, annual increases.


Note 8
Selection Criteria
The factors most important to the enterprise should be indicated clearly. Price needs to be included, but may not be the most important factor.

  • Technical appendices: 1) Overview schematics of the current technical environment. Include planned changes. 2) Lists of existing hardware needing coverage by the proposal.
Bottom Line

A well-planned and crafted RFP will result in a much more efficient and competitive procurement process, which in turn will result in saving time and money. Cutting corners upfront results in time wasted downstream.


Acronym Key
BC     Business continuity
DASD     Direct-access storage device
ESP     External services provider
MIPS     Millions of instructions per second
RFI     Request for information
RFP     Request for proposal
RPO     Recovery point objective
RTO     Recovery time objective

This research is part of a set of related research pieces. See AV-14-5338 for an overview.