STRATIGRAPHIC BENEFIT SCORE Standard v0.1 for pilot testing November 7, 2026 | Founder: Steve Oliver Mayr Status: Proposed evaluation method. Not a validated rating system. No public grades authorized by this document. WHAT IS THE STRATIGRAPHIC BENEFIT SCORE The Stratigraphic Benefit Score, or SBS, is a proposed way to measure how much good a data-center project brings to the people who live and work nearby—and whether it keeps its promises. WHY WE USE THE WORD STRATIGRAPHIC Stratigraphic means relating to layers. The word comes from studying layers of rock and earth. We use it as a way to explain that a strong community also has many layers: safe workers, good jobs, clean water, dependable power and opportunities for local businesses. SBS examines these layers together to ask: Does this project help build a stronger community for the future? The name describes our approach. It does not mean this is a geological assessment. WHO DEVELOPED IT Developed by Stratigraphic Design + Development. Built on evidence. Focused on people. SBS is being developed and tested. No company ratings have been issued. Business relationships must be disclosed, and paid work cannot purchase a better rating. PURPOSE AND PUBLIC DUTY The Stratigraphic Benefit Score exists to help communities evaluate whether data-center development creates lasting, verifiable benefits for workers, families and future generations. We will publish our evidence and limitations, correct errors, and apply the same rules regardless of commercial relationships. Investment size cannot offset serious unresolved safety or resource concerns. Scores are decision support, not engineering certification, legal compliance determinations or guarantees. EVALUATION UNITS Evaluate the specific campus, phase, location and period. Identify developer, owner, operator, tenant, general contractor and relevant subcontractors separately. Attribute conduct only where evidence establishes responsibility. A project result is not automatically a company-wide finding. Maintain separate committed and delivered scorecards. Commitments require executed, measurable obligations. Delivered outcomes require outcome evidence. Announcements appear as unscored proposals. Compare like stages; do not place an unbuilt proposal in an operating-performance league table. Keep the Community Site Value Profile separate from SBS. A site's commercial attractiveness does not measure benefits to residents. WEIGHTS AND SUBMETRICS Seven categories total 100 points. Each category has five equally weighted submetrics. Weights are provisional until the pilot is complete. 1. Lasting jobs and industrial development — 25 points A. Net additional permanent employment, with local resident participation separately reported. B. Wages, benefits and sustained employment, including job-years. C. Documented new or expanded U.S. manufacturing and durable domestic supplier commitments. D. Additional employers enabled by shared infrastructure or documented recruitment partnerships. E. Training completion, job placement, retention and advancement. Report construction hours and job-years separately from permanent positions. Distinguish local, regional and national outcomes. Do not count transfers between U.S. locations as net new national jobs. Attribute a contribution rather than claiming sole causation where multiple parties enabled an outcome. Do not duplicate one job across employers or categories. 2. Water and environmental stewardship — 15 points A. Water consumption and withdrawal relative to sustainable design-specific supply, including drought. B. Air emissions and verified mitigation. C. Noise, lighting and neighborhood exposure. D. Land, habitat and site-specific impact mitigation. E. Monitoring, actual performance and corrective action. Report absolute use as well as intensity. Technical terminology alone earns no credit. Do not assume zero water use from a closed-loop label. 3. Power and shared infrastructure — 15 points A. Documented power deliverability and reliability arrangements. B. Allocation of upgrade and operating costs without unsupported burdens on other customers. C. Protection against stranded infrastructure and early departure. D. Additional usable infrastructure capacity accessible to other employers or residents. E. Measured infrastructure performance and maintenance responsibility. Do not award shared-benefit points for infrastructure exclusively serving the campus. SBS does not substitute for utility engineering studies. 4. Net public financial benefit — 15 points A. Net fiscal value by affected public entity after incentives and costs. B. Exposure to public capital, service costs, guarantees and contingent liabilities. C. Comparison with no-project and credible alternative land uses. D. Distribution and timing of benefits relative to burdens. E. Downside resilience, enforceable performance terms and recovery mechanisms. Use stated time horizons and discount rates, with sensitivity analysis. Do not subtract abatements twice when receipts are already net of abatements. Keep wages and procurement separate from tax receipts. Avoid benefit-cost ratios where denominators are too small to be informative. 5. Worker health and worksite safety — 15 points A. Fatalities, serious injuries and illnesses, with exposure hours and appropriate comparisons. B. Prevention, qualified supervision and control of critical hazards. C. Subcontractor coverage and oversight across the site. D. Worker reporting, retaliation protections and evidence of reporting completeness. E. Independent verification, corrective action and recurrence prevention. Zero reported injuries is not proof of safe work. Describe reporting scope and denominators; do not compare raw incident counts across unequal project sizes. Small samples and incomplete hours can preclude a rate comparison. Serious incidents require context and investigation outcomes; a fatality is not by itself a finding of legal fault. Worker reports are leads, not established misconduct. 6. Construction quality and lifecycle responsibility — 5 points A. Independent commissioning and specification verification. B. Documented defects, correction and repeat failures. C. Maintenance, warranties and supported service-life planning. D. Equipment replacement and continued financial responsibility. E. Closure, remediation and end-of-life arrangements. Price and country of manufacture are not proxies for quality. Do not double count the same commissioning result as both quality and worker safety without distinct evidence for each outcome. 7. Transparency, accountability and community participation — 10 points A. Accessible public terms, metrics, source documents and updates. B. Resident participation before consequential decisions, including affected neighborhoods. C. Measurable obligations, deadlines and independently reviewable remedies. D. Accessible local procurement and documented enterprise participation. E. Complaint resolution, public reporting and response to verified concerns. This combines the earlier 5-point transparency category and 5-point participation category without changing the overall allocation. Maintain both subthemes visibly. RATING ANCHORS For every submetric use 0–5; unknown is blank. These common anchors are a framework, not sufficient numerical thresholds for immediate public scoring. 0: Documented serious adverse performance or material failure of the defined benchmark. 1: Documented major shortfall with limited effective response. 2: Partial fulfillment; material gaps remain. 3: Meets a predeclared, project-appropriate acceptable benchmark. 4: Verifiably exceeds the benchmark with a durable, material benefit. 5: Verifiably achieves the predeclared exceptional benchmark and demonstrates durability. Before scoring, write a metric-specific anchor sheet: metric definition, unit, geography, timeframe, comparator, numeric thresholds where appropriate, required evidence, attribution rule and treatment of contradictions. Freeze it before reviewing scores. Do not retroactively change anchors to favor a company. Legal minimum compliance alone does not establish exceptional performance. A documented adverse outcome is not automatically assigned to an entity without examining its responsibility. Where specialists are needed, withhold the assessment until qualified review is available. CALCULATION AND MISSING DATA Each submetric carries one-fifth of its category weight. Weighted points = submetric weight × rating / 5. Known points = sum of assessed weighted points. Coverage = sum of assessed submetric weights / 100. Report the uncertainty interval as known points through known points plus unassessed weight. It is a mathematical range, not a probability interval or an allegation of poor performance. Do not reweight known items to conceal missing evidence. Headline point scores require at least 80% coverage, all critical gates assessed, and no material unresolved contradiction. This provisional threshold must be tested. Even above the threshold, show missing weight and the interval alongside known points; never present an observed-only average as a complete score. No letter grades, badges, best-company claims or recommendation thresholds until calibration. No score can be purchased. CRITICAL REVIEW GATES Assess essential safety, water and power feasibility, material public liabilities, and evidence integrity before a favorable recommendation. A substantiated unresolved critical hazard or essential feasibility gap produces a review hold regardless of points. Record the exact basis, responsible entity, uncertainty and evidence needed to resolve it. A hold is not a court finding or declaration that a project is legally prohibited. Unknown critical information means insufficient evidence for recommendation, not guilt. EVIDENCE LEDGER For each claim retain: claim ID; project and phase; responsible entity; metric and units; baseline; geography; reporting period; source title and URL or document identifier; document date; retrieval date; evidence class; exact supporting passage or location; assessor; limitations; contradictions; response; score impact; next review date. Evidence classes: official or executed record; independently verified measurement; company-reported result; credible attributed reporting; unverified lead. Source authority and outcome certainty are separate: an official filing may contain estimates. Company reports alone may support a disclosed provisional assessment but cannot establish exceptional delivered outcomes without corroboration. Avoid publishing personal worker information. Unknown, unavailable, disputed and not applicable must remain distinct. During the pilot, do not remove a submetric as not applicable and redistribute weight without a published method revision. INDEPENDENCE AND CONFLICTS Disclose ownership and commercial relationships among SBS, DataCentersTX, Stratigraphic Design + Development and related businesses. Paid consulting may support better projects but cannot purchase score changes, suppress findings or guarantee outcomes. No fees contingent on ratings or favorable approvals. Separate consulting and evaluation responsibilities; any rating involving a client requires independent external review with the conflict shown on the profile. If independent review cannot be secured, do not issue a public grade for that client. Advertising must be labeled and cannot affect selection, weights, findings or corrections. Companies may provide documents and respond to factual findings; they cannot negotiate points. Government incentive eligibility is determined by public authorities and applicable law, not SBS. COMPANY AGGREGATION Do not issue aggregate company ratings in v0.1. First disclose project-level findings and sample-selection limits. Before aggregation, predefine sampling, minimum coverage across projects, stage matching, recency, role attribution, treatment of acquisitions and weighting. Show ranges and critical holds so one large project cannot conceal repeated failures. A small convenience sample is not a representative company track record. REVIEW AND CORRECTIONS An assessor documents each preliminary finding; a second qualified independent reviewer scores the same evidence without seeing the first score. Reconcile differences with written reasons. Technical disputes go to a suitably qualified reviewer, not a marketing decision. Offer the affected company 15 business days to supply a factual response before initial publication; identify this as a proposed editorial policy. Missing response does not prove a finding. Publish material responses and explain remaining disagreements. Correct verified errors promptly with a dated change log. Appeals receive a reviewer who did not make the original determination. Set review dates based on stage and evidence freshness; material events trigger earlier review. PILOT WORK PLAN Step 1: Complete all 35 metric-specific anchor sheets and obtain safety, infrastructure and public-finance review. Until this is done, no numeric ratings. Step 2: Select three to five projects across at least three unrelated developer/operator groups, with documented selection reasons. Include operating and construction examples but compare them within stages. Include Abilene as a candidate only after identifying the exact campus and entity roles; allegations from workers remain unverified leads. Step 3: Collect agreements, official project records, utility evidence, resource studies, employment data, available safety records, commissioning documentation and public-finance schedules. Record gaps instead of filling them with assumptions. Step 4: Run independent paired reviews. Publish disagreement rates and resolutions. Test 70%, 80% and 90% evidence thresholds and weight changes of plus or minus 20%, renormalized to 100. Report whether conclusions change materially. Avoid selecting thresholds merely to generate rankings. Step 5: Seek community and company factual feedback; revise the method with a public version log. Step 6: Release the method, evidence coverage and pilot findings. Decide whether the evidence supports numeric scores. Draft the op-ed around supported findings and clearly labeled forecasts. Do not claim validation from a small pilot alone. IMMEDIATE STATUS Completed: founding commitments, category structure, common scoring anchors, evidence rules, calculation, conflict policy and pilot sequence. Outstanding: metric-specific thresholds, qualified external reviewers, actual project evidence, calibration, company responses and publication decisions. No companies scored. No government incentive promised. No outreach, submission, website change or publication performed under this document.