Create a Repeatable Demand Engine

Task

Launch the initial SaaS demand-generation funnel.

Summary

Build a consistent mix of content, search, community, events, paid acquisition, or outbound activity.

Deep Research Scoping Report for an Ambiguous or Missing Prompt

Executive summary

The stated assumption that the attachment is missing does not match the materials available in the session. The operative Markdown prompt is highly specific: it requests a publication-ready, research-backed article on launching an initial software-as-a-service demand-generation funnel, with marketing-qualified pipeline as the principal measure and a weekly target to be set from company evidence rather than presented as a universal benchmark. It also specifies audience, editorial voice, source hierarchy, article structure, citation standards, examples, diagrams, length, and a strict source-exclusion rule.

Two supporting presentations add implementation detail. One describes a phased process built around stakeholder selection, baseline metrics, diagnosis, prioritization, strategy development, and executive approval. The other supplies campaign-workflow patterns connecting content, landing pages, email nurturing, webinars, marketing automation, sales qualification, customer-relationship-management records, opportunities, assets, owners, platforms, and budgets. These presentations are useful as internal planning aids, but the final public work would still need independent verification through authoritative public sources. In particular, the Markdown prompt expressly prohibits reliance on one named organization represented in the supplied material, creating a mandatory source-isolation and final-text audit requirement.

The strongest domain-agnostic interpretation is that any serious research assignment should be converted into a research charter before evidence collection begins. That charter should define the decision to be supported, primary and secondary questions, scope boundaries, stakeholders, deliverables, acceptance criteria, source hierarchy, methods, timeline, budget assumptions, data-handling rules, and unresolved issues. ISO’s project-management guidance is deliberately applicable across organizations, project types, life cycles, sizes, costs, and delivery approaches, supporting the use of one adaptable control framework rather than unrelated processes for each domain.

A rigorous medium-scope engagement would normally require approximately 100–150 person-hours over four to six weeks, assuming the work includes attachment inspection, a documented search, source verification, one or two public examples, analytical synthesis, drafting, visual development, citation checking, and editorial quality assurance. This is a planning estimate rather than an industry benchmark. It should be replaced by a bottom-up estimate once scope, source access, geography, review depth, and stakeholder availability are known. The U.S. Government Accountability Office recommends grounding estimates in purpose, scope, schedule, a work-breakdown structure, assumptions, data, methods, risk analysis, documentation, presentation, and later updates with actual costs.

The recommended immediate actions are:

1. Freeze a one-page research charter and record all unresolved questions.
2. Build a source-and-claim matrix before drafting.
3. Conduct the research through parallel evidence streams: official sources, open research, public company or institutional examples, and implementation guidance.
4. Produce a traceable evidence pack alongside the main report or article.
5. Separate verified facts, calculations, interpretations, and assumptions.
6. Conduct independent citation, source-prohibition, confidentiality, and numerical-consistency checks before publication.

The central success criterion is not merely delivery of polished prose. Completion should leave an informed decision-maker able to understand what was found, how reliable it is, what remains uncertain, which action is recommended, what it would cost, and what evidence should trigger reconsideration.

Reconstructed brief and requirement extraction

The supplied Markdown file is best understood as both a content assignment and a quality-control specification. Its content objective is narrow, but its procedural and editorial constraints are extensive. The presentations function as supporting operational context rather than as the final evidence base.

Extracted requirements

The exact task record, source hierarchy, editorial requirements, deliverable, measure, and target are all explicit in the supplied Markdown prompt. The campaign-workflow presentation adds a concrete operational model in which awareness activity leads to a landing page or event, then to download or registration, nurture, marketing qualification, sales qualification, opportunity creation, and either continued nurturing, disqualification, opt-out, or journey completion. It also pairs each workflow with an asset register covering teams, technology platforms, and budget.

The diagnostic presentation contributes a useful general project sequence: select a steering group and working team, gather baseline metrics, conduct a diagnostic, prioritize weak areas, build a roadmap, create a strategy, and obtain executive approval. Its externally sourced benchmarks and performance claims should not be carried into the public article without re-verification, both because some are dated and because the operative prompt establishes a stricter public-source policy.

Stakeholder map

No assignment should be treated as ready merely because these role titles have been listed. For a smaller organization, one person may fill several roles; the important control is that each decision, source of evidence, review, and approval has an identifiable owner.

Success criteria

A credible final result should meet five tests.

Decision utility. The work should identify the decision it supports and distinguish between evidence that changes the decision and background that merely adds context. CDC’s current evaluation framework similarly emphasizes relevance and utility, rigor, independence and objectivity, transparency, and ethics, with a sequence from context assessment through evidence gathering, conclusions, and action.

Traceability. Each material assertion should map to a source, calculation, interview, dataset, or clearly labelled assumption. For evidence reviews, Cochrane recommends planning and documenting sources, searches, record management, selection, and eligibility decisions.

Reproducibility. Another qualified analyst should be able to understand the query set, inclusion rules, transformations, calculations, and reasons sources were accepted or rejected. PRISMA 2020 provides a 27-item reporting checklist and flow-diagram templates for systematic reviews, although its full requirements should be applied only when the project is genuinely a systematic review.

Operational usability. The recommendation should identify who acts, what they produce, when they act, which systems are involved, how outcomes are measured, and what happens when results fall short. The supplied workflow presentation demonstrates the value of representing both the customer path and the asset-owner-platform-budget structure.

Explicit uncertainty. The report should separate observed evidence from estimates, interpretations, and hypotheses. Cost and schedule estimates should identify their assumptions and uncertainty rather than present a single number as certain; GAO treats comprehensive, documented, accurate, and credible estimates as distinct characteristics of quality.

Domain-agnostic research design

The same basic workflow can support a technology-product assessment, academic study, policy brief, or market analysis, but the evidence standards and outputs should change by domain.

flowchart LR
    A[Clarify the decision] --> B[Create research charter]
    B --> C[Inspect supplied materials]
    C --> D[Build questions and source hierarchy]
    D --> E[Collect official and primary evidence]
    E --> F[Evaluate quality and conflicts]
    F --> G[Synthesize findings and options]
    G --> H[Validate with stakeholders]
    H --> I[Produce deliverables]
    I --> J[Audit citations, assumptions, risks, and compliance]
    J --> K{Decision ready?}
    K -->|Yes| L[Approve and act]
    K -->|No| M[Collect targeted missing evidence]
    M --> F

Text description: the process starts with the decision, converts it into a controlled research charter, collects and evaluates evidence, validates the interpretation, and cycles back for targeted evidence when the result is not decision-ready.

Research charter

The charter should contain:

ISO 21502 supports adapting project practices to predictive, incremental, iterative, adaptive, or hybrid delivery rather than forcing every project into a single life cycle. The research charter should therefore be strict about controls but flexible about sequencing.

Evidence collection model

A strong evidence base normally uses four layers:

These layers are not interchangeable. A company’s official documentation may establish what its product does, but it is rarely sufficient to establish comparative superiority. An academic study may estimate an effect in one setting, but not prove that the same effect will occur in a different market. Internal conversion data may be the best evidence for a company-specific weekly pipeline target, but weak tracking definitions can make that data misleading.

Research options

For systematic evidence work, Cochrane distinguishes the steps of determining scope and questions, setting inclusion criteria, searching and selecting studies, collecting data, assessing bias, preparing synthesis, and interpreting results. For academic transparency, the Open Science Framework defines preregistration as posting a time-stamped, read-only study plan before data collection or analysis.

Research plans across representative domains

Comparative plan

The hour ranges are planning estimates, not universal standards. They assume that the topic is reasonably bounded, source access is mostly public, and no large-scale original data collection is required.

Technology-product research plan

The product plan should begin with a decision such as: “Should the organization build, buy, pilot, or defer this capability for this user group and use case?” The work should not begin with a feature inventory.

Methods. Conduct approximately 8–15 interviews across intended users, administrators, buyers, support staff, and affected risk owners. Observe current work where feasible. Convert findings into a problem statement, workflow, requirements, exclusions, acceptance tests, and risk register. Build the smallest prototype or technical test capable of resolving the highest-value uncertainty. Test usability, reliability, integration, privacy, security, accessibility, and operating cost separately rather than blending them into one subjective score.

For an artificial-intelligence product, NIST’s AI Risk Management Framework is use-case agnostic and organizes implementation around the functions Govern, Map, Measure, and Manage. Its development used public consultation, workshops, and contributions across industry, academia, civil society, and government. The framework is voluntary, so it should inform rather than replace applicable laws, sector requirements, and internal controls.

Data sources. Prioritize standards and regulator materials, official product and application-programming-interface documentation, vulnerability or incident databases, original benchmark papers, user research, internal support records, product analytics, and cost records.

Deliverables. A decision report, requirements document, tested prototype or benchmark notebook, risk register, test dataset with data dictionary, and executive slide deck.

Success criterion. The team should know which problem is being solved, for whom, under what conditions, with which measurable acceptance tests and residual risks.

Academic-study research plan

The academic plan should translate the topic into a structured question, define the population and variables, establish what would count as confirmatory versus exploratory analysis, and identify ethical and data-management requirements before collection begins.

Methods. Conduct a structured or systematic literature review; define eligibility criteria; develop a protocol; select or validate instruments; perform a sample-size or precision analysis; obtain institutional ethics approval when human participants are involved; preregister the study where suitable; create a statistical-analysis plan; and define data-quality, missing-data, exclusion, and correction procedures.

PRISMA 2020 supplies reporting checklists and flow diagrams for systematic reviews, while Cochrane provides detailed guidance on planning, searching, selecting studies, collecting data, assessing bias, and synthesis. These resources govern review conduct and reporting; they should not be misapplied as universal standards for every empirical design.

NIH’s Data Management and Sharing Policy requires covered research to plan and budget for data management, submit a plan, and comply with the approved plan. Common plan elements include data types, software or code, standards, preservation and access, reuse considerations, and oversight. Although NIH rules apply specifically to covered NIH research, their planning structure is useful more broadly.

Data sources. Use bibliographic databases, original papers, trial registries, discipline repositories, official measurement manuals, government health or social datasets, and cited-data repositories.

Deliverables. Protocol, preregistration, ethics package, search strategy, screening log, extraction form, analysis code, documented dataset, manuscript, and reproducibility appendix.

Success criterion. A knowledgeable reviewer should be able to distinguish planned from exploratory work, evaluate bias and uncertainty, and reproduce the analysis subject to legitimate privacy restrictions.

Policy-brief research plan

The policy plan should define the public problem, affected populations, current legal and institutional baseline, plausible options, distributional consequences, implementation capacity, fiscal implications, and evaluation design.

Methods. Review current legislation, regulations, court decisions, administrative responsibilities, existing programs, and prior evaluations. Develop a problem tree or logic model. Compare policy options using effectiveness, cost, feasibility, administrative burden, equity, legal authority, stakeholder acceptability, implementation time, and reversibility. Consult affected groups rather than relying solely on institutional stakeholders.

CDC’s policy analytical framework moves from problem identification to identifying, assessing, and prioritizing options and then developing an adoption strategy. OECD’s public-policy evaluation toolkit emphasizes institutional capacity, continuous learning, and the practical use of evaluation evidence, while its entrepreneurship-policy framework highlights clear objectives, common metrics, and credible comparison groups.

Data sources. Use statutes, regulations, official gazettes, government budget documents, administrative data, national statistics, audit reports, court records, public consultations, and open impact evaluations.

Deliverables. A concise policy brief, technical evidence annex, legal and jurisdictional map, option-comparison table, implementation plan, fiscal model, stakeholder record, and monitoring-and-evaluation framework.

Success criterion. Decision-makers should understand not only which option appears strongest, but also the legal authority, implementation burden, beneficiaries, potential losers, cost uncertainty, and evidence needed to evaluate the policy after adoption.

Market-analysis research plan

The market plan should resist the common error of declaring a large “total addressable market” without specifying customer, product, geography, purchase unit, price, adoption constraints, and substitutes.

Methods. Define the relevant market and adjacent alternatives. Build both a top-down model using official industry and demographic data and a bottom-up model using identifiable accounts, units, usage, contract values, and realistic penetration. Interview customers and former customers. Review competitors’ filings and maintained product materials. Test assumptions through sensitivity and scenario analysis.

The U.S. Census Bureau’s Business Builder combines demographic, socioeconomic, and business information and supports geographic comparisons, downloadable reports, and market analysis. The SEC provides free public access to millions of filings through EDGAR, including annual and quarterly reports and searchable financial-statement data. BEA’s industry accounts provide output, value-added, employment, and input-output information that can help test industry-level assumptions. BLS sources can support occupational, wage, employment, and projection assumptions.

Data sources. National statistics, industry accounts, trade data, regulatory registrations, procurement databases, public-company filings, pricing pages, product documentation, customer interviews, and internal sales data.

Deliverables. Market-definition memo, customer segmentation, competitive map, top-down and bottom-up models, pricing comparison, scenario analysis, source dataset, and executive deck.

Success criterion. The model should expose its assumptions and show which variables most affect the conclusion. It should not conceal uncertainty behind a single large market number.

Recommended actions, deliverables, timeline, and budget

Priority sequence

Deliverable packages

Budget scenarios

The following budgets are transparent planning scenarios rather than market-rate claims. They use illustrative blended labor assumptions and should be recalculated after the work-breakdown structure is approved.

The budget should identify whether editing, visualization, legal review, participant incentives, transcription, proprietary databases, translation, travel, and taxes are included. GAO’s cost guidance recommends documenting purpose, technical baseline, work breakdown, assumptions, data, methodology, sensitivity, risk, presentation, and later updates with actual costs.

Illustrative medium-scope timeline

gantt
    title Illustrative structured research engagement
    dateFormat  YYYY-MM-DD
    axisFormat  %b %d

    section Definition
    Research charter and scope freeze       :a1, 2026-08-04, 4d
    Attachment and stakeholder audit        :a2, 2026-08-04, 6d

    section Evidence
    Search protocol and source matrix        :b1, after a1, 4d
    Official and primary-source research     :b2, after b1, 12d
    Academic and independent research        :b3, after b1, 12d
    Interviews or targeted data requests     :b4, after a2, 15d

    section Analysis
    Evidence assessment and synthesis        :c1, after b2, 7d
    Models, tables, and workflow              :c2, after b3, 7d

    section Production
    Full draft                               :d1, after c1, 7d
    Stakeholder review                       :d2, after d1, 4d
    Revision and executive deck              :d3, after d2, 5d
    Citation and compliance audit            :d4, after d3, 3d

Text description: the six-week example overlaps official-source research, academic research, and targeted outreach, then consolidates the evidence before drafting, review, revision, and final audit.

The schedule assumes prompt access, responsive stakeholders, mostly English-language sources, and no lengthy ethics, procurement, legal-discovery, or public-records process. Public-records requests and restricted-data agreements can extend the schedule substantially because response time depends on record complexity, exemptions, agency backlog, legal review, and negotiation. FOIA.gov advises requesters to search existing public materials first, identify the agency holding the records, and describe requested records reasonably and in writing; agencies are not required to create new records, perform analysis, or answer research questions.

Risks, unknowns, assumptions, and mitigation

Governing assumptions

The timeline and budget above assume that:

• The decision can be framed without a major discovery phase.
• Most sources are accessible in English and without proprietary subscriptions.
• The research does not require a large representative survey or controlled trial.
• Stakeholders can review materials within two to three working days.
• The output requires no jurisdiction-specific legal opinion.
• Supplied data are machine-readable or can be converted without extensive manual transcription.
• One lead researcher can coordinate specialists without substantial procurement delay.
• The primary output receives no more than two substantive review rounds.

Changing any of these assumptions should trigger a revised estimate, not merely greater pressure on the original schedule. GAO’s schedule guidance notes that an integrated schedule should show activities leading to major events and support analysis of how changes affect completion.

Recommended controls

The minimum control set should include a research charter, versioned search log, source matrix, definitions register, assumptions register, risk register, decision log, claim ledger, calculation workbook or code repository, and final quality-assurance checklist.

For academic or data-intensive work, a data-management plan should state the data type, software and code, standards, preservation, access, reuse restrictions, and oversight. These fields parallel the structure used in NIH’s current data-sharing guidance.

For public policy work, evaluation planning should be connected to implementation rather than postponed until after rollout. CDC’s framework includes context assessment, program description, focused questions and design, credible evidence, supported conclusions, and action on findings.

For technology or AI work, risk review should cover the lifecycle and involve multiple affected functions. NIST emphasizes that identifying and managing AI risks requires perspectives and actors across the lifecycle rather than a purely technical review.

Likely primary sources and outreach templates

Primary-source bibliography

The following sources form a strong domain-agnostic starting library. They should be supplemented by topic-, jurisdiction-, and industry-specific materials.

Project definition, schedules, and estimates

• International Organization for Standardization, ISO 21502:2020 — Project, Programme and Portfolio Management: Guidance on Project Management. The standard is designed for diverse organizations, project purposes, delivery approaches, sizes, costs, and durations.
• U.S. Government Accountability Office, Cost Estimating and Assessment Guide: Best Practices for Developing and Managing Program Costs. Useful for work-breakdown structures, assumptions, risk, sensitivity, documentation, and estimate updates.
• U.S. Government Accountability Office, Schedule Assessment Guide: Best Practices for Project Schedules. Useful for dependencies, critical paths, schedule quality, and change analysis.

Evidence reviews and academic studies

• Page and colleagues, PRISMA 2020 Statement. Provides a reporting checklist and flow diagrams for systematic reviews.
• Cochrane, Handbook for Systematic Reviews of Interventions. Covers scope, inclusion criteria, searching, selection, collection, bias, synthesis, and interpretation.
• Open Science Framework, Registrations and Preregistrations. Defines time-stamped study-plan registration and provides operational guidance.
• National Institutes of Health, Data Management and Sharing Policy. Provides requirements and templates for prospective data-management planning in covered research.

Technology and product risk

• National Institute of Standards and Technology, Artificial Intelligence Risk Management Framework. Provides a voluntary, use-case-agnostic framework for governing, mapping, measuring, and managing AI risks.
• U.S. Government Accountability Office, Agile Assessment Guide. Useful when incremental development, continuous evaluation, and customer feedback are part of the technical delivery model.
• U.S. Government Accountability Office, Technology Readiness Assessment Guide. Useful for distinguishing an experimental capability from one mature enough for operational use.

Policy analysis and evaluation

• Centers for Disease Control and Prevention, Policy Analytical Framework. Provides a sequence for identifying problems, describing and assessing options, prioritizing options, and planning adoption.
• Centers for Disease Control and Prevention, Program Evaluation Framework. Establishes evaluation standards and a practical cycle from context through action.
• OECD, Implementation Toolkit for the Recommendation on Public Policy Evaluation. Supports institutional evaluation capacity and evidence-informed policy learning.

Market and company analysis

• U.S. Securities and Exchange Commission, EDGAR Search Filings and Data Resources. Provides free access to public-company filings, full-text searches, structured financial data, and application-programming interfaces.
• U.S. Census Bureau, Census Business Builder. Provides demographic, socioeconomic, geographic, and business data for market analysis.
• U.S. Bureau of Economic Analysis, Industry Economic Accounts. Provides industry output, value added, employment, productivity, and input-output data.
• U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics and Employment Projections. Supports labor availability, occupation, wage, and industry-employment analysis.

Public records

• U.S. Department of Justice, FOIA.gov. Explains how to search proactively disclosed materials, identify the correct agency, describe records, specify format, and submit a written request.

Expert-interview request

Subject: Request for a short research interview on [topic]

Hello [Name],

I am conducting research to support a decision about [specific decision] for [organization or project].

Your experience with [specific area] would help us understand:

• [Question or uncertainty]
• [Question or uncertainty]
• [Question or uncertainty]

Would you be available for a [30–45]-minute interview during [date range]?

The discussion will be used for [internal analysis/public report/academic study]. We will [attribute comments by name / seek approval before attribution / treat comments as background]. Participation is voluntary, and you may decline to answer any question.

I can provide the discussion guide in advance.

Thank you,
[Name]
[Role and organization]
[Contact information]

Private data request

Subject: Request for data on [subject, population, and period]

Hello [Data owner],

We are researching [decision or question] and would like to request the following existing data:

• Records or variables: [precise list]
• Population or unit: [customers, transactions, organizations, locations]
• Date range: [start and end]
• Preferred format: [CSV, XLSX, JSON, or database extract]
• Required metadata: field definitions, units, missing-value codes, revision history, and source-system description
• Identifying information: [none required / requested fields]
• Intended use: [analysis and decision supported]
• Retention: [period]
• Access: [named individuals or team]
• Publication: [aggregate findings only / no external publication / terms to be agreed]

Please advise whether the data exist, which restrictions apply, what approvals or agreement are required, and whether there are extraction or licensing costs.

We are available to narrow the request to reduce burden or privacy risk.

Regards,
[Name and contact details]

Academic author or replication-data request

Subject: Request for materials related to “[paper title]”

Dear [Author],

I am reviewing your paper, “[Title]”, as part of research on [topic].

Could you share or direct me to any publicly available:

• Analysis code
• Data or synthetic data
• Survey instrument or interview guide
• Codebook or data dictionary
• Supplementary tables
• Preregistration or protocol
• Documentation of exclusions, transformations, or robustness checks

The materials would be used to verify our interpretation and assess applicability to [context]. We will cite the paper and repository according to your preferred citation.

I understand that participant privacy, licensing, or institutional restrictions may prevent sharing some materials. A description of those constraints or a controlled-access route would also be helpful.

Sincerely,
[Name, role, and organization]

Public-records request

Before submitting a public-records request, search the agency website and disclosure library, because the records may already be public. FOIA.gov states that a request must reasonably describe existing records and that agencies need not create records, conduct research, analyze data, or answer questions.

Subject: Freedom of Information Act request for [records]

To the FOIA Officer:

Under the Freedom of Information Act, I request copies of the following agency records:

Records sought:
[Describe the document type, subject, office, system, correspondence participants, contract, report, dataset, or decision record.]

Date range:
[Start date] through [end date].

Likely custodians or office:
[Office, division, program, or named role, when known.]

Search terms or identifiers:
[Contract number, project name, case number, organization, key phrases.]

Preferred format:
Electronic, in the original machine-readable format where available, including associated data dictionaries or field definitions.

Please provide all reasonably segregable non-exempt portions of responsive records. If the estimated fees will exceed [$ amount], please notify me before processing. If any part of the request is unclear or unduly burdensome, please contact me so that I may narrow it.

[Optional fee-waiver or expedited-processing justification, where applicable.]

Sincerely,
[Name]
[Address or contact information]

Final acceptance checklist

The research should not be accepted until the sponsor can answer “yes” to each of the following:

• Is the supported decision explicit?
• Are scope, exclusions, stakeholders, and definitions approved?
• Can every material claim be traced to evidence or a labelled assumption?
• Were primary and official sources prioritized?
• Were current claims checked against current sources and dated?
• Were contradictory findings and material limitations addressed?
• Are calculations reproducible?
• Are the timeline and budget tied to a work-breakdown structure?
• Are legal, ethical, privacy, and confidentiality risks addressed?
• Are deliverables usable by the people responsible for acting?
• Were all prohibited sources, phrases, unsupported claims, and confidential details removed?
• Does the result state what remains unknown and what evidence would change the recommendation?

A report that passes these tests provides more than information. It creates a controlled record of the question, evidence, assumptions, analysis, recommendation, and remaining uncertainty—enough for an accountable decision and a later review of whether that decision proved sound.