Emails & conversations
Questions, agreed responses and decisions, where their use can be authorized.
Emails. Spreadsheets. Project files.
The records you already have capture real work in context. Aegis helps you license that knowledge for AI training and evaluation while retaining ownership.
Explore the licensing process · Online inquiries are temporarily unavailable
A practical focus on rights, informed by our founder’s experience as an attorney.
Usefulness, permission and buyer fit determine whether a collection can be licensed.
You do not need to be a data company. A useful collection may be a byproduct of running your business.
Questions, agreed responses and decisions, where their use can be authorized.
Operational histories, working tables and the context that makes them understandable.
Plans, revisions, procedures and documented results that show how real work unfolds.
Material you created or can license, with the relevant content and participant permissions.
These are examples to discuss, not a request to upload files. Customer restrictions, confidentiality, personal information and third-party rights still need review.
Start by identifying your business workflow and one coherent collection. These guides explain the questions an assessment would address.
Aegis assesses the collection, reviews permissions, and handles agreed preparation and buyer outreach.
You approve the terms. If a deal completes and buyer funds are received, your share follows the written agreement.
Free initial assessment. No upfront supplier fee for agreed work. Scope, revenue share and payment terms are set out in your written agreement.
The opportunity starts with the rights you actually hold. An assessment does not authorize buyer training or transfer ownership.
Identify confidential, personal or restricted material. We review exclusions and permissions before preparing an agreed sample.
The agreement defines the included data, permitted uses, buyer obligations and your share of proceeds. It also addresses which licensed rights continue after supply ends.
Learn how business records are assessed, how permissions are reviewed and how licensing terms are agreed.
Licensing value depends on buyer requirements, usable scope, permissions and commercial terms. We assess those factors together and explain the basis for the opportunity.
Begin with a short business description. Aegis handles agreed preparation and licensing work. Your team explains the records, identifies restrictions and arranges approved access.
Customer information requires authority for the intended use and a review of the relevant obligations. We examine customer agreements, confidentiality and privacy before agreeing the included material. Removing names alone does not establish anonymity or permission.
The initial assessment is free, with no upfront supplier fee for agreed preparation and outreach. Scope, responsibilities, revenue share, payment timing and any deductions are agreed in writing before proceeding.
Explore the path from business records to an agreed licensing opportunity.
Informational website · Online inquiries are temporarily unavailable
What buyers need, what you may already have, and what to ask before agreeing.
Read guide →ArticleHow to keep ownership while granting a defined use under written terms.
Read article →ArticleWhy everyday records may have a second commercial use.
Read the original article →Christopher D. Forrest
Founder · Attorney · Business founder
Aegis Data Partners is led by attorney Chris Forrest, drawing on approximately 20 years of legal experience and experience building businesses. That background shapes a practical approach to commercial value, the permissions behind it and the terms you approve.
Meet Chris and read the approach →Explore four illustrative dataset profiles, or discuss a tailored requirement. Availability, rights and scope remain to be established.
How licensing value is determined, what buyers need, and how to get started without becoming a data expert.
These publications describe the intended licensing process. Online inquiries are temporarily unavailable.
Your business creates records because it needs to operate: plans, spreadsheets, job histories, reports, procedures, photos and footage. Those records sometimes contain something another organization cannot easily recreate: examples of how real work unfolds.
Consider a construction issue documented from the initial question through an approved response and final inspection, or a production run that connects job settings, rejected units and corrective actions. The value is in the context: the task, the evidence and the documented result.
The first task is to identify one coherent collection. There is no need to export your entire archive or send customer information to get started.
A license grants defined permission to use material. With Aegis, your company retains ownership and authorizes specific uses through a written agreement. Evaluation access and model training are defined separately.
Important terms include who may use the material, for what purpose, for how long, whether the permission is exclusive, and what happens to source copies and trained models after the agreement ends. A right to end future supply is different from a promise to remove information from an already trained model.
There is documented interest in proprietary data. OpenAI’s 2023 partnership announcement described interest in nonpublic collections across several formats [1]. Shutterstock’s 2023 announcement described a license covering image, video and music libraries with associated metadata [2]. Neither announcement establishes a price for an ordinary business’s records.
Aegis prepares business data for AI training and evaluation. We define the task it supports, organize the relevant records and evidence, and pursue buyers whose requirements match the collection.
Look for material that answers a recognizable question. Can someone connect a task to the evidence, a decision and an outcome? Are the files readable? Are dates and units consistent? Does the collection show meaningful variation rather than hundreds of copies of the same form?
Useful records are often scattered across software and folders. Aegis organizes and prepares the agreed collection. Your team explains the records and identifies exclusions. We assess the usable material and its context, rather than relying on a universal minimum file count.
A file on your server can include customer material, a vendor’s work, a contractor’s creation or personal information. Our review examines its origin, relevant agreements, privacy commitments and intended use. We document the authorized scope, exclusions and any permissions that need resolution.
The FTC has emphasized that AI companies must honor privacy and confidentiality promises [3]. For an owner, the practical takeaway is to explain existing obligations before sharing material. Removing names is not a substitute for reviewing permission, confidentiality and identification risks.
Start by describing your business and its records. We assess the collection and agree the sample and permissions for further review. The written supplier agreement defines the material, Aegis’s authority, permitted buyer uses and the revenue arrangement.
There is no supplier assessment fee or upfront preparation charge for agreed work. Aegis handles the preparation and buyer outreach and earns an agreed share of licensing proceeds. Your team provides the context and approved access needed to prepare the collection.
We prepare a scoped data product, approach suitable buyers and negotiate within the authority you grant. Buyer access follows the agreed approval process. The agreement specifies payment timing, receipt of buyer funds, reporting and any permitted deductions. No licensing deal means no supplier fee under this model.
The product determines the relevant commercial unit: usable cases, distinct clips, original tasks or a defined collection. Pricing reflects the useful material and the rights being licensed, rather than a universal rate per file, token or hour of footage.
Price reflects buyer requirements, usefulness, permitted uses, quality, scale, alternatives and commercial terms. The supplier agreement sets your share. Payment follows the completed license and its written provisions; recurring arrangements specify their own refresh and payment terms.
Write a short description of one collection: the work it documents, approximate volume, date range, where it is kept and any known restrictions. Do not assemble a technical package or upload sensitive files. A free assessment should help determine whether a more detailed conversation is worthwhile.
Primary sources support the numbered factual statements. Product design, operating practice and commercial analysis reflect Aegis’s approach.
Sources reviewed October 3, 2026. Historical sources describe the agreements and guidance at their publication dates. Aegis assesses each collection using its own scope, permissions and commercial terms.
Explore the process and readiness questions. Online inquiries are temporarily unavailable.
Read the readiness checklistProject questions, approved changes and inspection records can tell a story that individual files cannot.
Construction produces a detailed record of decisions under real constraints. We connect requests for information with relevant drawings, authorized responses, changes and documented inspections to assess coherent cases for training and evaluation.
Source context matters. A later drawing is not automatically the controlling instruction. Document-reasoning evaluation requires an answer supported by the permitted evidence and the authority recorded in the case.
For example, finish-selection records can support a task that asks an AI system to identify the authorized finish and cite the evidence. Human review confirms the answer and records missing information or ambiguity. This example explains the case structure; the collection and permissions are scoped with the supplier.
Drawings and reports may involve architects, engineers, owners and subcontractors with different rights and confidentiality obligations. Site photos may reveal people, addresses, security layouts or sensitive facilities. Active disputes and privileged material warrant separate consideration. Having a project file does not automatically authorize reuse of all its contents.
How many completed projects have linked records? Can your team identify authorized decisions? Where are documents kept? What contract restrictions apply? Could sensitive projects or third-party material be excluded while leaving useful cases?
Aegis handles agreed preparation and licensing work without an upfront supplier fee. Your team explains the records, identifies restrictions and arranges agreed access. We define the collection around its evidence, permissions and the buyer’s task.
Explore the process and readiness questions. Online inquiries are temporarily unavailable.
Read the industry guideThe opportunity may lie in how jobs were produced and corrected, rather than in the customer content you handled.
A printing or mail-processing company may have years of job specifications, machine settings, inspection results, production logs and rework records. Those records can document a practical workflow. Mailing lists, customer artwork and campaign content are a different category and should not be presumed available for licensing.
Units deserve special attention. Sheets, impressions, finished pieces and mailings are not interchangeable. A record becomes more useful when another person can understand its definitions without relying on undocumented shop knowledge.
In this worked example, a run begins with 12,900 sheets and ends with 12,000 accepted sheets and 900 rejects, with no other material movements. The task asks for the accepted yield, approximately 93.02%, and the evidence supporting a reported correction. The arithmetic is straightforward; interpreting what the records establish is the richer task.
An operator’s note that a setting changed is different from evidence that the change caused an improvement. The dataset preserves that distinction so an evaluation can assess evidence-based reasoning. The figures above are an illustrative worked example.
Customer contracts, confidential campaigns, mailing addresses, financial details and licensed artwork require review before reuse. Machine and software vendors can also hold relevant rights. We identify what can be included, what must be excluded and how those decisions affect the operational collection.
Can logs be linked to a job and outcome? Are units and codes documented? How much history is available? Do your systems support an authorized export? What obligations apply to customers and vendors?
You do not need to clean the archive first. Aegis assesses the collection, agrees the permitted scope and handles preparation and buyer outreach under the no-upfront-fee supplier model.
Explore the process and readiness questions. Online inquiries are temporarily unavailable.
Read the industry guideYour operational knowledge, expert network and business relationships may each open a different door.
Staffing firms create records of role demand, scheduling, assignment operations and service delivery. We begin with a description of those workflows, without requesting candidate files. Review covers identification risks, including small-group aggregates, before the licensing scope is agreed.
Recruiting data frequently contains personal and employment information. Resumes, background checks, identity documents, medical details and individual screening histories are not a default source collection for Aegis. The first conversation should describe categories and restrictions, not transfer records.
Your client relationships create an additional route: introductions to businesses with useful operational histories. We speak directly with an authorized decision-maker about that business’s collection. An introduction does not authorize a staffing firm to license or disclose its customer’s data.
Referral compensation is agreed separately and disclosed where appropriate. Introductions require no confidential customer material and do not grant licensing authority.
Experienced professionals create original exercises, answer rubrics and business scenarios for defined buyer requirements. Examples include a finance reconciliation exercise or expert review of an illustrative construction question.
Original expert work has its own scope, compensation, confidentiality and intellectual-property terms. It is distinct from licensing an existing archive. Contributor agreements address authorship, employer obligations and the exclusion of confidential or third-party material.
For example, a staffing firm can describe scheduling workflows, introduce consenting operators and identify experienced managers for original task work. We assess each route separately and agree its permissions, responsibilities and commercial terms.
Which industries do you serve? Which operational records are genuinely controlled by your firm? What personal or client information do they contain? Could you make an introduction without sharing confidential details? Are experts interested in original work under new agreements?
Select the route that fits your relationships. A business introduction may be more useful than trying to turn sensitive candidate records into a product.
Explore the process and readiness questions. Online inquiries are temporarily unavailable.
Read the industry guideYou can keep ownership while granting a defined use. The important work is deciding exactly what that permission covers.
“License your data” sounds simple until you ask what the buyer is allowed to do. Can it inspect a sample? Train a model? Retain source files? Give an affiliate access? Use the material in another product? These are different permissions, and a useful agreement should address them explicitly.
The license identifies its subject clearly. The product description specifies the date range, included formats, excluded categories and release identifier. Those details establish the boundary of the collection being licensed.
Buyer evaluation starts with a scoped sample and separately agreed permitted uses. Evaluation permission and training permission are distinct; the written terms govern each.
The agreement also addresses sample retention, reporting of results and access by other parties. Those terms are settled with the supplier and buyer before access is granted.
Retaining ownership does not make every license consequence disappear. A nonexclusive arrangement may permit licensing to other buyers; an exclusive arrangement may restrict it. A long-lived training license may continue after the supplier stops providing new data.
Similarly, deleting a delivered source copy does not necessarily remove what a trained model learned. An owner should understand those practical limits before agreeing to a use, rather than rely on a vague promise of being able to take everything back.
A supplier should be able to follow the calculation from buyer payment to supplier proceeds. The agreement should define the revenue base, Aegis’s share, any expressly permitted adjustments, collection timing and reporting. “You get a percentage” is incomplete if no one explains a percentage of what.
Owning a business does not automatically establish rights to every element in its files. Customer content, vendor documentation, employee or contractor contributions and privacy obligations require review. A contract with Aegis cannot manufacture a right that the supplier does not have.
Our model is straightforward: a free assessment, an agreed scope, work managed by Aegis and shared proceeds from a completed licensing deal. The written terms define the responsibilities, permissions and payment arrangements.
Explore the process and readiness questions. Online inquiries are temporarily unavailable.
Read the readiness checklistA large archive is a starting point. A buyer’s need, usable material and the permission being sold determine the opportunity.
“How much will I get?” is the right question for a business owner to ask. Early in a conversation, it may also be the question that cannot yet be answered responsibly.
There is no universal conversion from files, tokens or hours into licensing revenue. Counting material helps scope a project; it does not establish that anyone wants to purchase it.
A collection’s commercial value begins with the task it supports: following authorized project instructions, reconciling operational records or interpreting footage. We scope the material and evaluation around the buyer’s requirements.
For workflow records, a useful unit may be a complete case. For video, it may be a permitted clip with descriptive information. For original expert work, it may be a task with a checked answer and scoring rules. Total archive size can greatly exceed the amount that survives preparation and review.
The quality of preparation matters, but effort alone does not create a price. DataComp-LM studies dataset design through standardized training and evaluation [1]. That research supports taking data selection seriously; it does not establish commercial rates for Aegis products.
A limited evaluation and a broad training license are different commercial propositions. Exclusivity, permitted users, duration, refreshes and downstream rights can affect negotiations. Broader rights can also impose greater costs on the supplier, including limiting other opportunities.
Buyers compare source collections, generated exercises and commissioned expert work. We describe the distinctive context, evidence and coverage in each collection so the buyer can evaluate its fit. Claims of uniqueness require a concrete account of what distinguishes the material.
The buyer’s license payment and the supplier’s share are distinct amounts, defined in writing. We assess usable scope, rights and buyer requirements before discussing a collection-specific earnings estimate.
For now, the useful first question is whether your records justify an assessment. If they do, a scoped sample and buyer conversation can produce better evidence than a speculative earnings calculator.
Primary sources support the numbered factual statements. Product design, operating practice and commercial analysis reflect Aegis’s approach.
Sources reviewed October 3, 2026. Historical sources describe the agreements and guidance at their publication dates. Aegis assesses each collection using its own scope, permissions and commercial terms.
Explore the process and readiness questions. Online inquiries are temporarily unavailable.
Read the readiness checklistPreparation is more than removing names. It is the work of making a collection understandable, appropriately scoped and ready to evaluate.
An archive built to run a business was not necessarily built for someone outside that business to understand. Codes may be undocumented, files duplicated, outcomes missing and permissions spread across several agreements.
Aegis manages preparation with defined assistance from the supplier. We build a coherent data product with context, permissions and a clear evaluation purpose.
Before requesting a sample, we agree what it contains, who authorizes its review and how it is accessed. We identify customer restrictions and sensitive categories at this stage. An assessment does not authorize model training.
We check relationships between records, missing fields, duplicates and the readability of scans and exports. A person familiar with the workflow helps resolve ambiguity and explain the evidence.
Preparation should preserve the difference between source evidence and a later interpretation. If AI assists with extraction or labels, those outputs should be distinguished from original material and checked against an agreed quality standard.
Privacy review covers direct identifiers, contextual identification, confidential information and third-party material. We determine which records require exclusion or treatment. Automated assistance supports the review; human judgment and documented permissions govern release.
The FTC has warned that quietly expanding privacy terms to permit new uses may be unfair or deceptive [1]. The practical implication is that preparation must respect existing commitments instead of treating a revised policy as an automatic cure.
The release package describes the data units, source lineage, permitted uses, limitations and quality checks. For products with answers, reviewers confirm that the supporting evidence justifies them. Buyer evaluation and training permissions are agreed separately.
We approach buyers within the supplier agreement’s authority and release approved material under documented terms. Version identifiers connect delivery, payment, questions and corrections to the same collection.
Not every collection will reach this point. Discovering an unsuitable archive early is a useful outcome: it avoids turning uncertainty into an expensive preparation project.
Primary sources support the numbered factual statements. Product design, operating practice and commercial analysis reflect Aegis’s approach.
Sources reviewed October 3, 2026. Historical sources describe the agreements and guidance at their publication dates. Aegis assesses each collection using its own scope, permissions and commercial terms.
Explore the process and readiness questions. Online inquiries are temporarily unavailable.
Read the readiness checklistDefine the capability, evidence, acceptance checks and permitted use you need. We scope the collection and evaluation around those requirements.
Explore information for AI teams Explore the catalogA practical framework for documenting origin, permissions, privacy boundaries and commercial scope before a dataset is offered to a buyer.
These publications describe the intended licensing process. Online inquiries are temporarily unavailable.
A data product requires a defined collection and a documented permission path. Our rights-review framework establishes what material is included, the authority for its intended use, and the exclusions and obligations governing release. It connects commercial scope to evidence and written approval.
The framework separates four questions: origin, authority, sensitive content and licensed use. Combining them into a single statement that “we own the data” can conceal important gaps.
Begin with the intended capability and a defined collection. Identify the data unit, source types, date range and exclusions. A focused product makes quality review and rights investigation precise and manageable.
For example, a construction evaluation case links an authorized decision to permitted evidence. Scope review identifies exclusions such as security plans or disputed communications, and records the reasons for those decisions.
Maintain a record of where each source came from, who supplied it, when it was received and how it changed during preparation. Distinguish company-created material from customer, vendor and contractor contributions. Record AI-generated labels separately from source facts.
The release manifest links each product unit to its permitted sources and transformation history. It supports investigation and correction while limiting unnecessary sensitive details in buyer-facing material. Source lineage records the history; rights review establishes the licensing authority.
The review should identify the person authorized to act for the supplier and the agreements affecting the material. Relevant questions can involve intellectual property, confidentiality, privacy, employment and vendor terms. Requirements depend on facts, jurisdictions and the intended use; this paper does not provide a complete jurisdiction-specific legal checklist.
A rights register should record the evidence supporting a permission, its scope, unresolved issues and the reviewer. A supplier’s assurance may be relevant, but it should not be treated as proof that every third-party contribution has been cleared. Unresolved categories should be excluded or held until investigated.
The FTC’s January 2024 guidance emphasizes honoring privacy and confidentiality commitments [1]. Its February 2024 guidance warns against surreptitious retroactive changes to privacy terms [2]. These are useful signals for the framework: review the commitments under which information was collected before proposing a different use.
Use a source-specific assessment to identify direct identifiers, contextual identification, confidential business material and prohibited categories. Decide whether to exclude, redact, aggregate or obtain further permission. No single processing method resolves all of these issues. Removing a name does not establish that the remaining record is anonymous or licensable.
Preparation combines automated assistance with human review appropriate to the collection and intended use. Review findings and release decisions are documented. Security, privacy and certification claims require evidence of the relevant controls.
Specify evaluation, training and any other permitted uses distinctly. Address affiliates, contractors, onward transfer, source redistribution, retention and derived outputs. Decide how buyer identity and scope are approved and which actions Aegis may take without obtaining a new supplier approval.
Termination language should distinguish ending future collection from ending existing licenses. It should also distinguish deletion of source copies from the practical consequences of prior model training. Parties should not promise model unlearning or complete reversibility without a defined, supportable mechanism.
The supplier agreement defines licensing proceeds, the agreed share, expressly permitted adjustments, payment triggers and reporting. It states the no-upfront-fee arrangement and the work within scope, so responsibilities and payment obligations are clear from the start.
Nonexclusive licensing is our preferred starting point. The supplier and buyer agree any exclusivity explicitly, with a defined scope, duration and commercial justification.
These documents should make a product reviewable rather than merely assert that it is “clean.” Quality failures, rights gaps and restricted categories should be recorded honestly. A buyer can then evaluate an actual scope instead of relying on broad marketing language.
Proceed when there is a plausible buyer use, understandable source evidence, a supportable permission path and an economic reason to prepare the product. Pause when authority, contractual restrictions, privacy obligations or essential context remain unresolved. Decline a collection when the intended use cannot be supported.
This framework keeps rights review, commercial scope and release decisions connected. We apply it to the collection at hand, maintain the evidence and document the permissions, exclusions and approvals that govern the license.
Primary sources support the numbered factual statements. Product design, operating practice and commercial analysis reflect Aegis’s approach.
Sources reviewed October 3, 2026. Historical sources describe the agreements and guidance at their publication dates. Aegis assesses each collection using its own scope, permissions and commercial terms.
Explore the process and readiness questions. Online inquiries are temporarily unavailable.
Read the readiness checklistDefine the capability, evidence, acceptance checks and permitted use you need. We scope the collection and evaluation around those requirements.
Explore information for AI teams Explore the catalogA worked example showing how a project question becomes an evidence-backed case for AI evaluation.
A fictional contractor encounters conflicting finish references. Its file includes a schedule, an approved change, a response to a project question and an inspection record. The final authorized choice is Finish B. These facts are invented to illustrate the process.
Ask the model to identify the authorized finish and cite the supporting record. The evaluation also checks whether it identifies a conflicting document and distinguishes a revision number from decision authority.
Connect the records using consistent case references. Preserve dates and document relationships. Remove or exclude material outside the agreed scope. Have a qualified reviewer verify the answer, supporting evidence and ambiguity. Record any added labels as preparation outputs.
Review the rights and confidentiality obligations affecting each document, including identification risks, sensitive facilities and protected third-party material. Access to a drawing and permission to license it for model training are separate questions; the review documents the relevant authority.
Acceptance requires an authoritative record, documented permission for the intended use, sufficient distinct cases and a task that matches the buyer’s requirements. The review records the findings and determines the approved scope.
The example shows how a business workflow becomes an evidence-backed evaluation case. For a supplier collection, the release package records the source material, permissions, review findings and commercial scope. The worked example uses fictional project records.
Explore the process and readiness questions. Online inquiries are temporarily unavailable.
Read the readiness checklistDefine the capability, evidence, acceptance checks and permitted use you need. We scope the collection and evaluation around those requirements.
Explore information for AI teams Explore the catalogA clear process from your business description through assessment, preparation and licensing, with your approval at each stage.
An assessment begins with the work a business does, the records it keeps and known restrictions. This informational website explains that process; online inquiries are temporarily unavailable.
The free assessment reviews one collection: its business context, accessibility, licensing authority and fit with buyer requirements. We explain the findings and the next steps.
Before a sample is shared, we agree its scope, access method, confidentiality conditions and permitted review. The assessment does not authorize buyer training. Restricted or unresolved material is held back.
The supplier agreement defines what Aegis prepares and offers, how buyer access is approved and how proceeds are shared. It addresses exclusivity, duration, reporting, termination and existing buyer licenses. Our preferred starting point is a defined, nonexclusive scope.
Aegis organizes approved records, reviews permissions and sensitive information, and checks the resulting product. Your team explains the workflow and arranges agreed access. Agreed assessment, preparation and outreach carry no upfront supplier fee.
Buyer evaluation uses a permitted sample under agreed terms. We define acceptance criteria, price, delivery and training rights separately. Evaluation findings guide the final scope and commercial discussion.
Aegis negotiates within its authorized role and releases material under agreed terms. You receive your share according to the written payment provisions, which specify closing, receipt of buyer funds and payment timing.
New releases, additional buyers and refreshes follow the granted authority and permissions. Recurring arrangements define their own delivery and payment terms. If no license is sold, there is no supplier fee under this model.
Explore the process and readiness questions. Online inquiries are temporarily unavailable.
Read the readiness checklistExploring income should begin with understanding what can be shared and what must stay private.
Describing your business does not transfer ownership or authorize AI training. We agree the scope and review conditions before a sample is shared. You decide whether to proceed at each stage.
Aegis licenses defined uses while the supplier retains ownership. We review the underlying authority, including customer content and third-party material, and document the permitted scope.
We review authorship, governing agreements, confidentiality obligations and identification risks. The review determines treatment or exclusion of personal details, sensitive facilities, trade secrets and privileged communications. Automated assistance supports the process; documented findings and approvals govern release.
The license addresses evaluation, training, permitted users, onward access, retention and other agreed uses. Exclusivity is agreed explicitly. Aegis operates within documented authority and the supplier’s process for approving buyer access.
Future supply and existing buyer rights are different questions. A trained model may not be fully reversible merely because a source file is deleted. These limits should be discussed before a deal rather than hidden behind a broad promise of control.
The release documentation describes the review process, findings and permissions governing the collection. Specific certification, security or commercial claims require supporting evidence. Each collection receives a review of its own circumstances.
Explore the process and readiness questions. Online inquiries are temporarily unavailable.
Read the readiness checklistChristopher D. Forrest · Attorney and business founder · Founder of Aegis Data Partners
Businesses accumulate valuable knowledge through everyday work. Aegis helps owners turn that knowledge into licensing opportunities, handling preparation and commercial work within a clearly agreed scope.
I bring approximately 20 years of legal experience as an attorney, along with experience building and operating a consumer business. That background informs my focus on clear agreements, practical execution and commercial value for both sides.
Successful licensing brings together buyer requirements, documented permission and reliable preparation. We establish the scope, involve technical specialists and outside counsel where the work requires them, and keep responsibilities clear.
Our resource collection is intended to explain the opportunity without obscuring its limits. The articles explore concrete business records, the economics of licensing and the rights questions that determine whether a transaction can proceed.
Our resources explain the business process, the commercial decisions and the permissions involved. They connect practical examples to source-backed facts and clear analysis.
Help owners understand what they have. Handle agreed preparation and licensing work. Share proceeds from a completed deal. Explain costs, permissions and limitations before an owner commits.
Explore the process and readiness questions. Online inquiries are temporarily unavailable.
Read the readiness checklistClear guidance connects documented facts, practical examples and the decisions involved in licensing business data.
Our market statements link to primary sources where practical. We distinguish documented facts from commercial analysis. Worked examples explain the process and identify illustrative records clearly.
Collection-specific pricing reflects the scope and commercial terms of that collection. Buyer relationships, transactions, endorsements, certifications and results require specific supporting evidence.
Christopher D. Forrest is responsible for the positions in the bylined resources. Research assistance supports the writing process; source links, conclusions and legal or technical statements require substantive human review.
Resources carry revision dates, and historical announcements retain their original dates. Material updates and corrections keep the collection current and traceable.
Written agreements define service scope, licensing authority, revenue shares, payment terms and responsibilities. The free assessment and no-upfront-fee arrangement apply to agreed work. This website provides information; any commercial relationship requires a separate written agreement.
Explore the process and readiness questions. Online inquiries are temporarily unavailable.
Read the readiness checklistTen original guides and articles on business data licensing, written to make the next conversation more useful.
These publications describe the intended licensing process. Online inquiries are temporarily unavailable.
The records you keep to get work done may have a second commercial use. The opportunity begins with a question, not a promise.
Read article →Featured original article · 2 min readPreparation is more than removing names. It is the work of making a collection understandable, appropriately scoped and ready to evaluate.
Read article →10 resources
How licensing value is determined, what buyers need, and how to get started without becoming a data expert.
Read resource →The records you keep to get work done may have a second commercial use. The opportunity begins with a question, not a promise.
Read resource →You can keep ownership while granting a defined use. The important work is deciding exactly what that permission covers.
Read resource →Project questions, approved changes and inspection records can tell a story that individual files cannot.
Read resource →The opportunity may lie in how jobs were produced and corrected, rather than in the customer content you handled.
Read resource →Your operational knowledge, expert network and business relationships may each open a different door.
Read resource →A large archive is a starting point. A buyer’s need, usable material and the permission being sold determine the opportunity.
Read resource →Preparation is more than removing names. It is the work of making a collection understandable, appropriately scoped and ready to evaluate.
Read resource →A practical framework for documenting origin, permissions, privacy boundaries and commercial scope before a dataset is offered to a buyer.
Read resource →A worked example showing how a project question becomes an evidence-backed case for AI evaluation.
Read resource →Try another topic or clear the search and category.
Practical guides to licensing, rights review, dataset design and buyer evaluation. Reading times are based on the article text at 200 words per minute.
You do not need to be a technology company. Begin with the records your team already keeps.
Project questions, approved changes and inspection records can tell a story that individual files cannot.
Read industry guideIndustry guideThe opportunity may lie in how jobs were produced and corrected, rather than in the customer content you handled.
Read industry guideIndustry guideYour operational knowledge, expert network and business relationships may each open a different door.
Read industry guideThese are initial categories to investigate, rather than a complete list or a claim of confirmed buyer demand. Production footage, professional expertise and other business records may also warrant a conversation.
Read the readiness checklistRead the owner’s guideFive questions to help you describe your records and identify what a conversation should resolve.
Use these questions as a static checklist for your own review. This website does not collect answers, provide a valuation or certify readiness.
For example: an issue and its resolution, or a production run and inspection.
Codes, units and document relationships are useful context.
You do not need to export anything now.
Consider customer agreements, third-party material and who can approve.
Consider personal information, confidential content and restricted projects.
Explore how a clear model capability guides the records, evidence and permissions needed for a useful evaluation.
Scope, samples, pricing and licensing terms are agreed for your requirement.
What should the model do better?
What evidence supports a correct answer?
How will you judge a useful pilot?
Who may use it, and for what purpose?
Dataset profiles for AI training and evaluation. We scope each collection around the task, available sources and permitted uses.
4 illustrative profiles
Linked project records that preserve a question, its authorized decision and the documented outcome.
Job histories that connect the specification, production events, quality issues and corrective actions.
Original business scenarios with working materials, worked solutions and evaluation criteria.
Scoped footage with scene, action and capture context for a clearly defined perception task.
Try a broader term or clear the filters. These four profiles are starting points; a custom requirement may be worth discussing.
Describe the capability, evidence, intended use and acceptance criteria.
A worked example shows the intended structure without implying a secured dataset.
A finish schedule, approved change, RFI response and inspection.
Which finish was authorized, and which records support the answer?
A task, source citations, answer rubric and documented permission boundaries.
Linked project records that preserve a question, its authorized decision and the documented outcome.
Each profile describes a data structure and its applications. We agree the final collection, source permissions and acceptance criteria around your requirements.
We agree volume, coverage, formats, exclusions, permitted uses, price and delivery before a pilot. Evaluation permission and training rights are defined separately. Source and sample availability are confirmed during scoping.
The schema defines the fields, types and source references used to structure the collection. Final fields follow the agreed task and source material.
| Field | Type | Purpose |
|---|---|---|
project_reference | string | Pseudonymous project key; the mapping remains restricted. |
record_type | enum | Identify the source class, such as RFI, drawing or inspection. |
created_at | datetime | Place the record in its chronology. |
related_record_ids | array<string> | Link supporting records and revisions. |
process_stage | enum | Describe the stage of the workflow. |
decision_summary | string | Summarize an authorized decision with source support. |
outcome | string | null | Record a documented result; preserve unknown outcomes. |
source_reference | string | Trace the entry to a permitted source. |
Review covers context, remaining identifiers, obligations and the intended use. Removing names is one treatment; it does not replace the permissions and privacy review.
Acceptance criteria and required checks are agreed before a pilot. Findings and approvals are documented for the collection before release.
Discuss the task, evidence, intended use and acceptance criteria.
Job histories that connect the specification, production events, quality issues and corrective actions.
Each profile describes a data structure and its applications. We agree the final collection, source permissions and acceptance criteria around your requirements.
We agree volume, coverage, formats, exclusions, permitted uses, price and delivery before a pilot. Evaluation permission and training rights are defined separately. Source and sample availability are confirmed during scoping.
The schema defines the fields, types and source references used to structure the collection. Final fields follow the agreed task and source material.
| Field | Type | Purpose |
|---|---|---|
job_reference | string | Link records for the same job without exposing customer identifiers. |
specification | object | Describe permitted job requirements and unit definitions. |
revision_number | string | Identify the recorded specification revision. |
quantity | number | Capture the measured amount, interpreted with its unit. |
unit | enum | Distinguish sheets, impressions, pieces or other defined units. |
issue_category | enum | null | Classify a documented quality or production issue. |
corrective_action | string | null | Record an action as reported, with evidence where available. |
final_status | enum | Describe the recorded completion or review status. |
source_reference | string | Link a value or event to its permitted source. |
Review covers context, remaining identifiers, obligations and the intended use. Removing names is one treatment; it does not replace the permissions and privacy review.
Acceptance criteria and required checks are agreed before a pilot. Findings and approvals are documented for the collection before release.
Discuss the task, evidence, intended use and acceptance criteria.
Original business scenarios with working materials, worked solutions and evaluation criteria.
Each profile describes a data structure and its applications. We agree the final collection, source permissions and acceptance criteria around your requirements.
Original expert work. This profile covers newly created expert tasks with original working materials and rubrics. Contributor scope, compensation, confidentiality and intellectual-property terms are agreed separately. Review findings are recorded for each task.
We agree volume, coverage, formats, exclusions, permitted uses, price and delivery before a pilot. Evaluation permission and training rights are defined separately. Source and sample availability are confirmed during scoping.
The schema defines the fields, types and source references used to structure the collection. Final fields follow the agreed task and source material.
| Field | Type | Purpose |
|---|---|---|
task_id | string | Uniquely identify an original task. |
scenario_type | enum | Describe the business reasoning skill being exercised. |
prompt | string | State the task and permitted assumptions. |
input_materials | array<asset_reference> | Link newly created working files and context. |
expected_output | object | string | Specify the intended answer or output format. |
solution_rationale | string | Explain the reproducible solution. |
evaluation_criteria | array<object> | Define how the output is scored. |
difficulty | enum | Record an agreed difficulty classification. |
author_rights_status | enum | Track documented contributor permission or an unresolved review. |
review_status | enum | Record the review status and approval decision. |
Review covers context, remaining identifiers, obligations and the intended use. Removing names is one treatment; it does not replace the permissions and privacy review.
Acceptance criteria and required checks are agreed before a pilot. Findings and approvals are documented for the collection before release.
Discuss the task, evidence, intended use and acceptance criteria.
Scoped footage with scene, action and capture context for a clearly defined perception task.
Each profile describes a data structure and its applications. We agree the final collection, source permissions and acceptance criteria around your requirements.
We agree volume, coverage, formats, exclusions, permitted uses, price and delivery before a pilot. Evaluation permission and training rights are defined separately. Source and sample availability are confirmed during scoping.
The schema defines the fields, types and source references used to structure the collection. Final fields follow the agreed task and source material.
| Field | Type | Purpose |
|---|---|---|
clip_id | string | Identify the scoped asset and its source trail. |
duration_seconds | number | Describe the agreed clip length. |
resolution | object | Record pixel width and height. |
frame_rate | number | Support timing interpretation. |
audio_present | boolean | Flag audio for a separate permission review. |
scene_type | enum | string | Describe the scene under an agreed taxonomy. |
action_labels | array<object> | Associate actions with defined time intervals. |
capture_context | object | Retain useful, permitted context about the recording. |
rights_status | enum | Track documented permission or unresolved review. |
Review covers context, remaining identifiers, obligations and the intended use. Removing names is one treatment; it does not replace the permissions and privacy review.
Acceptance criteria and required checks are agreed before a pilot. Findings and approvals are documented for the collection before release.
Discuss the task, evidence, intended use and acceptance criteria.
A construction finish conflict shows how linked business records form an evidence-backed evaluation case. The project records in this walkthrough are fictional.
Four records describe the same issue. A later date alone is not proof of authority.
The original schedule calls for Finish A in the workshop.
Fictional recordThe authorized change replaces Finish A with Finish B.
The response points to CHANGE-007 as the governing decision.
The inspection records Finish B installed in the workshop.
Use only the supplied records. Identify the controlling approval and distinguish the decision from the observation.
Finish B was authorized by CHANGE-007, which RFI-014 cites. Finish B was recorded as installed in INSPECT-003.
The original Finish A schedule does not override the approved change. The inspection supports installation, not approval authority.
This is the intended packaging, not a claim that real sources have been cleared or checked.
Supplier releases document source permissions, exclusion rules and review results. Evaluation access and training rights are defined separately.