Research · Measurement · Software
No hyperbole, armchair theory, or inflated costs.
Graystone Industries
Graystone Industries is the independent research, measurement, and software practice of Eric Gladstone, Ph.D.
Eric Gladstone holds a doctorate in Organizational Behavior from Cornell University and has held senior scientist roles at Meta and Roku. He currently serves as Head of Behavioral Science at a public affairs firm. Academic appointments include the University of Kentucky, the Stockholm School of Economics, and the Institute for Analytical Sociology.
Selected Clients
U.S. Department of Defense·Army Research Laboratory·NIH·CDC·SAGE Foundation·Kentucky State Government·Meta·Spotify·Roku·Novartis·Boehringer Ingelheim·Sealy·Gray Construction·Backcountry / CSC Generation·SYPartners·Victorious·Lexandria·Global Citizen Year·ACEs Connection·Hoop.la·Parity.org·Michigan Rotary·Yourprojectx
Services
The best summary is slightly obtuse but accurate: Graystone answers problems that involve humans, at any scale, through measurement. Measurement produces answers to questions and appropriate prescriptions.
Graystone does research and measurement on human problems — why people do what they do, at any scale — and builds software around what it finds. The domain spans every level of human organization: individual behavior, group dynamics, organizational design, market behavior, and policy. The method depends on the question. On the research side, this includes behavioral research and measurement, survey design and psychometrics, decision science and predictive modeling, people and organizational analysis, policy and communications research, and computational modeling and simulation, among others. Work has been fielded at scales ranging from targeted studies of a few hundred respondents to platform-scale experiments involving tens of millions of users.
The service categories are not dissimilar from what the large shops offer. The differences are in cost, speed, and depth of training. Large consultancies staff client engagements thinly — a few people, typically with business or applied psychology backgrounds, working within a proprietary framework. The scientific foundation is wildly shallower than the branding suggests. What Graystone offers is the same scope, faster, at a fraction of the cost, and with substantially stronger methodological grounding behind it. Eric Gladstone remains an active publisher in peer-reviewed journals — methodological approaches are always contemporary.
On the software side, Graystone builds production tools for commercial, organizational, and government use — custom applications, dashboards, monitoring and intelligence systems, scoring platforms, and simulation software. This is a standalone line of work and, at times, an appendage to research. Products can be licensed, deployed, acquired, or operated at scale. Where research and software meet, a finding can be operationalized into something an organization uses every day. Where they don't, Graystone builds because a client needs something built.
Approach
The model reflects how Graystone operates — processes that developed in parallel with the evolution of AI, grounded in measurement and technical fluency built across academia and private industry. In contrast to many shops, there is no hyperbole, no armchair theorizing, no vague frameworks dressed up as methodology. Everything is tailored to the client, empirically grounded, and built to scale.
The model is lean by design. To be precise about what AI-assisted means here: Graystone does not use GPT web interfaces or similar tools. Graystone builds deep, autonomous agents and systems that substantially increase its analytical capacity, digestion, scale, and output — and is more hands-on for it, not less. Projects move fast and overhead stays low.
The same depth of engagement that accelerates Graystone’s own work also makes it possible to evaluate others’. Organizations increasingly produce research, measurement systems, and analytical outputs with AI assistance. Catching what those systems get wrong requires the domain expertise to know what right looks like.
This allows for fair pricing. On price, speed, and quality of output, Graystone is quite confident it can outperform any comparable shop. Organizations should not be paying what the market currently charges for rigorous behavioral research and custom software. The overhead structures, partner billing rates, and framework licensing fees that drive those invoices do not reflect the actual cost of doing the work well. Eric Gladstone has bid on this work, hired for it, and delivered it across a range of contexts. When positioned correctly, the value is self-evident.
Capabilities
Research & Measurement
Most organizations know what their data says. Fewer know what their people are actually doing — and why. Graystone designs and executes behavioral research that produces defensible answers to questions about human conduct at any scale: individual, team, organization, market, or policy.
↗Research & Measurement
The only way to know whether an intervention caused an outcome — rather than coincided with it — is a well-designed experiment. Graystone designs and fields randomized controlled experiments and quasi-experimental studies at scales from a few hundred respondents to millions of users. The output is causal inference, not correlation.
↗Research & Measurement
Some questions can’t be answered with a single measurement. Tracking the same people, teams, or populations over time reveals how behavior, attitudes, and performance change in response to interventions, events, and organizational shifts — and whether the changes last.
↗Research & Measurement
Full-cycle survey research: instrument design, scale development and validation, sampling design, experimental paradigms, and quantitative analysis. The standard applied is academic validity, not the surface-level consistency checks that characterize most commercial survey work. Applied at scales from targeted samples of a few thousand to national files covering hundreds of millions.
↗Research & Measurement
How decisions are made, what predicts them, and how to change them. Behavioral segmentation, propensity scoring, predictive classification, and similarity modeling — applied to identify responsive populations, predict behavior, and prioritize resources across commercial, organizational, and policy contexts.
↗Research & Measurement
Every organization has two structures: the one on the org chart, and the one that actually governs how work gets done. Behavioral analysis of communication patterns, information flow, and collaboration structures identifies where the two diverge — and what that divergence is costing.
↗Research & Measurement
Organizations routinely assume their people systems — promotion, compensation, resource allocation — work as designed. Experimental research tests whether that assumption is correct: who actually advances, what actually drives pay, and where structural dynamics are producing outcomes the formal system wasn’t designed to create.
↗Research & Measurement
Survey-based culture scores describe how employees feel about culture. They don’t describe how culture actually operates. Graystone measures culture through behavioral indicators — how information moves, how decisions are made, how exceptions to stated norms are handled — producing a picture of what the culture actually rewards, independent of what it claims to value.
↗Research & Measurement
Who actually shapes decisions in your organization? Where does information stall before it reaches the people who need it? Network analysis identifies the real connectors and bottlenecks using structural data — and designs interventions that work at the network level, not the individual level.
↗Research & Measurement
Organized manipulation — coordinated review fraud, competitor disinformation, coordinated inauthentic behavior — leaves structural signatures that content-level monitoring misses. Graystone identifies these patterns using network-theoretic models developed through applied work at platform scale, applicable to any organization in a contested information environment.
↗Research & Measurement
Most organizations invest heavily in communications and have no rigorous way to know whether any of it is working. Experimental and quasi-experimental designs measure whether messages are changing understanding, attitudes, and behavior — internally and externally — rather than just generating activity metrics.
↗Research & Measurement
Did the reorganization improve coordination? Did the training change behavior? Did the campaign move opinion? Rigorous evaluation uses counterfactual inference to establish what would have happened without the intervention — producing an answer an organization can actually act on, not just a favorable comparison.
↗Research & Measurement
What customers tell you in surveys and what they actually do are different datasets. Behavioral research combines observation, experimentation, and quantitative modeling to produce an accurate picture of what drives customer behavior — what changes it, what predicts it, and where the standard assumptions are wrong.
↗Research & Measurement
Organizations introducing something new need to know not just whether people like it, but whether and why they use it — and what the behavioral barriers to adoption actually are. Behavioral research applied to new products, features, and interfaces before and after launch, at scales from targeted studies to platform-level measurement.
↗Research & Measurement
Sometimes the data already exists. Government records, employment files, platform logs, health data, financial records — large-scale administrative datasets contain answers that primary research alone can’t access. Graystone extracts those answers at the scale those datasets require, using the same methods applied to primary data.
↗Research & Measurement
Research designed to inform policy decisions and public communications strategy — public opinion measurement, message testing, audience segmentation for policy contexts, and behavioral modeling of how policies are adopted and how communications land across different population segments.
↗Research & Measurement
When the signal is in the text — community forums, internal communications, open-ended surveys, news coverage, social media — Graystone builds the pipelines that extract it at scale. Semantic, tonal, and affective analysis of large corpora, producing structured insight from data that resists manual analysis.
↗Software & Systems
Custom web applications, analytical tools, and data pipelines — built to spec against a functional need, or directly downstream of a research engagement. The output is deployed software an organization uses, not a prototype or a recommendation.
↗Software & Systems
Real-time systems that surface what’s happening before standard reporting channels catch it — user problems, reputational threats, emerging narratives, competitive signals, or coordinated adversarial activity. Built to ingest and analyze data at scale from social media, community platforms, news sources, and internal systems.
↗Software & Systems
Analytics environments that give non-technical users direct access to behavioral data and research outputs — with integrated statistical testing and interpretation built in. An organization that can answer its own questions between engagements is a different kind of organization.
↗Software & Systems
A continuously updated picture of how an organization is actually functioning — communication patterns, information flow, collaboration health, cultural indicators — delivered in a format executive leadership can read and act on. Persistent behavioral signal, not a periodic survey.
↗Software & Systems
Research outputs and datasets made accessible through natural language. Rather than receiving a static report, users query findings, explore data, and generate scenario analyses without analytical training. Built using LLM integration designed for the specific data environment and user population.
↗Software & Systems
When a research engagement produces a validated model or measurement instrument, Graystone builds the production tool that operationalizes it — accessible to non-technical users, integrated with existing infrastructure, and designed for repeated use at scale rather than one-time delivery.
↗Software & Systems
Unified person-centered data systems integrating consumer databases, organizational records, platform data, and proprietary sources with persistent identity resolution. Enables behavioral targeting, audience modeling, and contact enrichment — owned by the client, built for multi-cycle reuse.
↗Software & Systems
Agent-based and computational simulation of population-level behavioral dynamics — for scenario testing, policy design, and planning when field experiments are impractical or the system is too complex for standard analytical solutions.
↗Software & Systems
The engineering phase that follows a completed research engagement: translating a validated framework, scoring model, or measurement system into a deployed production tool that an organization uses every day.
↗Advisory
Before a research engagement begins, the most important work is defining the actual question. Graystone helps organizations articulate what they are trying to know, identify the right unit of analysis, and design a research architecture that will produce a defensible answer — before committing to a methodology or a vendor.
↗Advisory
A structured engagement to identify what is actually producing a performance, culture, or operational problem — as distinct from the presenting symptoms. Behavioral data analysis, network mapping, and structured inquiry, delivered as a lightweight advisory engagement or as a scoping engagement that precedes a full study.
↗Advisory
Advisory on what a communications program is actually designed to accomplish, who it needs to reach, what behavioral change it is asking for, and how to measure whether it’s working. Applicable to internal change communications, external campaigns, advocacy strategy, and crisis communications planning.
↗Advisory
Training and development of internal research capability — experimental design, measurement methodology, quantitative methods, and analytical interpretation — for research teams, people analytics functions, and leaders who need to commission, evaluate, and act on behavioral research. Delivered as workshops, curriculum design, or embedded advisory.
↗Advisory
Independent review of research designs, analytical methods, and statistical findings applied to work in progress or completed work product. Useful before a major decision is made on the basis of research results, or when existing findings are producing inconsistent or counterintuitive outcomes.
↗Advisory
Critical evaluation of research or analytical work produced by a third party — whether findings are valid, whether methods fit the questions, whether claims are supported by the data, and whether the organization is interpreting the work correctly. Useful when significant decisions or investments are on the line.
↗Advisory
Strategic advisory for organizations operating in contested information environments — assessing vulnerability to coordinated disinformation, adversarial influence campaigns, and organized manipulation of public or internal perception. Developed through research conducted with the U.S. Department of Defense and Army Research Laboratory on misinformation dynamics and adversarial network behavior.
↗Advisory
Organizations are using AI to design surveys, build measurement systems, analyze data, and produce findings — often without the domain expertise to evaluate whether the outputs are valid. Graystone reviews AI-generated research and analytical work for methodological soundness, measurement validity, and interpretive accuracy. The standard is the same as for any research: whether the conclusions are supported by the methods, and whether the methods fit the questions.
↗Contact