GFO 25-802 Q&A Analysis - MUST READ

What the California Energy Commission’s Latest GFO 25-802 Q&A Really Signals for Bidders

The California Energy Commission’s newly released Questions and Answers for RFP-25-802, focused on enhancing behind-the-meter PV and storage adoption modeling in California, is worth close attention from firms active in energy modeling, distributed energy resources, electrification planning, and public-sector contracting. On the surface, the document answers routine bidder questions. In practice, it offers much more: a roadmap for how bidders should think about teaming, compliance, methodology, and execution risk before they decide how aggressively to pursue this opportunity.

What stands out immediately is that this does not appear to be a narrowly defined “plug and play” technical contract. The CEC gives bidders meaningful room to shape their own approach, but with that flexibility comes greater responsibility to define assumptions, justify methods, and prove operational readiness. For some teams, that flexibility is a competitive opening. For others, it may expose gaps in staffing, subcontractor readiness, data strategy, or project management discipline.

One of the most important signals in the Q&A is that team composition will matter. The CEC confirmed that small business and DVBE participation can strengthen the competitiveness of a proposal, and it clarified that DVBE participation does not need to be confined to a narrow portion of the work but may be distributed across tasks and subcontractors. It also stated that letters of support from academic institutions, utilities, and research partners may be considered as part of qualifications scoring. Taken together, those answers suggest the agency is open to evaluating a proposal not just on technical design, but on the overall strength, credibility, and relevance of the delivery team.

That matters because many firms still approach opportunities like this as if technical methodology is the only real differentiator. In reality, the CEC’s answers point toward a broader scoring logic: does the team have the right technical capability, the right supporting relationships, the right participation structure, and the right administrative standing to execute in California? That is a more demanding standard, but it is also a more realistic one for a contract involving state-level modeling, utility-related data, and policy-relevant outputs.

Another major issue is subcontractor eligibility and compliance. The CEC made clear that California Secretary of State registration requirements extend to subcontractors, including smaller subcontractors, and that this expectation does not go away simply because a sub has a limited scope or a lower dollar value. Sole proprietors must also be properly registered at the local level and provide proof of that status. International firms, including Canadian companies, may compete as primes if they are validly registered with the California Secretary of State.

For bidders, this is not a minor footnote. It means that prime contractors need to vet eligibility and administrative readiness across the entire team before submission, not after selection. Many strong technical teams lose time, weaken narratives, or create avoidable contract risk because they assume subcontractor paperwork can be handled later. This Q&A is a reminder that for California opportunities, compliance readiness is part of competitiveness.

There is also a notable degree of post-award flexibility on subcontracting, although it comes with conditions. The CEC states that most subcontractor additions can be made without a formal amendment, depending on the contract type and the terms negotiated at contract execution. That is helpful for firms that want to keep some flexibility in reserve, especially where specialized data science, DER economics, software engineering, or utility engagement capacity may need to be added later. At the same time, “can be flexible later” is not the same as “do not plan now.” Teams still need a credible initial structure, particularly if they want to maximize participation incentives and present a coherent staffing plan.

From a technical perspective, the Q&A may be even more interesting. The CEC confirms that it will provide interval meter data to the selected contractor. It also indicates that for storage attachment-rate work, staff can provide recent historical attachment-rate data, and that available technical documentation and prior materials that may support onboarding can also be shared. In addition, the California-specific adaptation of dGen will be provided after contract execution, and staff expect to deliver it prior to the kickoff meeting, although some supporting datasets may be delayed if NDAs are required before access is granted.

That combination of answers tells bidders something very important: the agency is prepared to support contractor onboarding, but not all uncertainty is removed at award. Some core inputs come from the CEC, some relevant ownership data must be drawn from publicly available CPUC DGStats datasets, and some supporting data flows may be delayed by confidentiality requirements. In practical terms, successful bidders will need to show that they can sequence work intelligently, begin productively even if some data arrives later than hoped, and manage the early phase of the project with discipline.

The CEC’s answers also leave considerable room for methodological creativity. The agency states that it has no preferred proxy dataset or methodology for service territories where interval meter data is unavailable. It says there are no predefined performance benchmarks or accuracy targets for the updated dGen model because those are expected to emerge through coordination with the selected contractor. It further confirms that for the market dynamics task, it is open to replacing the Bass diffusion framework entirely if the contractor proposes a technically sound alternative.

That is a significant opening. It suggests the CEC is not looking only for a vendor that can execute a predetermined template. It may also be looking for a partner capable of advancing the model in a way that better reflects California market conditions, adoption behavior, cost structures, and ownership choices. But openness is not the same as lack of standards. Where the agency does not prescribe a method, the contractor must do more work to establish confidence. The burden shifts to the bidder to explain why a chosen framework is robust, how it will be calibrated, how it will be validated, and how decision-makers should interpret the outputs.

In some ways, that is one of the most important strategic implications of the Q&A. Bidders are being invited to bring judgment, not just capacity. They can propose different approaches, but they will likely need to make those choices legible to evaluators who care about usability, defensibility, and long-term maintainability. A proposal that is technically sophisticated but poorly explained may be less competitive than one that is slightly less ambitious but clearly structured and easy to trust.

The Q&A also points to a few operational details that should not be overlooked. The updated dGen work must remain Python-based, the contractor should use standard .py files rather than notebooks, and code should be configured with Windows operating systems in mind because that is the CEC’s environment. The CEC also notes that it will coordinate with the contractor on repository practices such as commits, pull requests, and branching, while keeping the overall GitHub repository structure unchanged unless a change is essential. Those details may sound routine, but they matter. They signal that the state wants maintainable, transferable, production-ready work rather than a black-box analytical product that only the original development team can operate.

There is a broader lesson here for consulting firms, labs, and technical contractors. Agencies are increasingly expecting software and analytical deliverables to be auditable, reproducible, and usable by internal staff after handoff. That means bidders should not treat documentation, code structure, onboarding materials, or environment compatibility as back-office items to be finalized later. In opportunities like this, they are central to the value proposition.

The communication model implied by the Q&A is also worth noting. The CEC expects check-ins as needed and says that based on prior experience, weekly coordination is likely. This suggests the selected contractor will not be operating in a vacuum. Instead, the work is likely to be collaborative, iterative, and shaped in real time through interaction with staff. Bidders who are used to long periods of independent technical work followed by occasional reporting may need to adapt their project management approach.

That has practical consequences for staffing. The strongest team may not simply be the one with the most advanced modelers. It may be the one that combines model development expertise with enough project leadership, documentation capacity, and client coordination skill to keep technical progress aligned with agency expectations week after week. In state-facing technical work, communication is often a performance variable, not an administrative side task.

The Q&A also confirms that the behind-the-meter adoption scope includes both residential and non-residential sectors. That matters because it expands the behavioral, economic, and data complexity of the modeling challenge. Residential and non-residential adoption dynamics are not identical, and bidders will need to think carefully about segmentation, calibration, cost treatment, and ownership structures across customer classes. The answer on third-party ownership data is especially relevant here: the CEC does not hold historical ownership data for TPO versus customer-owned installations, but it identifies the CPUC DGStats interconnected project datasets as a public resource for that work.

This is another subtle but meaningful signal. The agency is willing to point contractors toward data sources, but it is not eliminating the need for contractor-led synthesis. That means proposals should probably do more than list tasks. They should show how the team will bridge state-provided inputs, public datasets, and methodological decisions into a coherent modeling update.

So what should clients and bidders take from all this?

First, this opportunity appears well suited for teams that can combine technical depth with strong California contracting readiness. Second, the evaluation may favor organizations that think intentionally about team composition, support letters, and participation strategy rather than treating those as compliance attachments. Third, the technical scope leaves room for innovation, but bidders will need to use that freedom carefully and explain their choices with confidence and clarity. Finally, the implementation details suggest that delivery discipline, reproducibility, and communication cadence will be just as important as analytical sophistication.

For us, this Q&A reinforces a point we often make to clients: on public-sector technical RFPs, the strongest proposals do not merely answer the scope. They reduce evaluator anxiety. They show that the team understands not only the modeling challenge, but also the operational, administrative, and collaborative realities of getting the work done in a government environment.

This Q&A gives bidders a clearer picture of the rules, but it also raises bigger strategic questions about what public agencies really want from technical partners today. Is the winning team the one with the most advanced model, or the one that can best combine technical rigor, administrative readiness, and collaborative delivery? We would be interested to hear how others in the energy, DER, and public-sector consulting space are reading these signals