Insights
Software companies
256 guides · page 4 of 6
Jira, Zendesk or GitHub: which records should a software company license first?
Which data to license first at a software company? Score Jira, Zendesk and GitHub on rights clarity, privacy load, linkage and export effort, then start.
Knowledge base cleanup before AI: fix, archive or delete old articles?
Knowledge base cleanup for AI starts with one call per article: fix, archive or delete. Use the decision table, and archive old versions: history has value.
Legal practice management software vendors: what is licensable and what is not
Legal software vendor data licensing: why client matter data is off-limits, which vendor-owned records may qualify, and what to check before any AI license.
License-back terms after a software carve-out, including AI training rights
License back agreement in a carve-out: the terms that carry risk, from field of use to sublicensing, plus a sample AI training clause to review with counsel.
Licensing code vs open-sourcing it: which is better for a retired product?
License vs open source for a retired product: compare control, revenue, community, rights cleanup and what you can never undo once the repo is public.
Licensing data that contains your customers' end-customer information
Licensing data with end-customer information involves three parties' rights. See the rights chain, what to exclude or de-identify, and who must sign off.
Logistics, TMS and WMS software vendors: licensing shipment and exception data
TMS and WMS software vendor data licensing: which shipment and exception records you own, which belong to shippers, carriers or 3PLs, and who approves each.
Macros and canned responses: do they lower support data value?
Canned responses add little AI training value on their own, but macro choice, agent edits and follow-ups do. Learn to flag macros instead of deleting tickets.
Maintenance revenue vs subscription revenue: how acquirers value legacy software
How acquirers value maintenance revenue vs subscription revenue in legacy software, which lines count as recurring, and the records that prove renewals.
Manufacturing ERP, MES and QMS vendors: whose quality data is it?
MES and QMS software vendor data rights: who owns NCRs, CAPAs and inspection records, what a vendor can license, and which designs to exclude.
Marketing automation vendors: subscriber data vs your own platform records
Marketing automation platform data ownership: subscriber lists and content belong to customers, while deliverability and support records can be yours.
MCP and your business data: is connecting AI tools the same as licensing?
Model Context Protocol links AI tools to your business data live, for your own use. Licensing hands a defined copy to an outside developer under contract.
Memorization and regurgitation clauses for licensed source code
A code license memorization clause limits how much licensed source code a model can reproduce. Compare output filters, no-verbatim warranties and checks.
Merging Salesforce orgs after an acquisition: keep the old history
Salesforce org merge after an acquisition: an archive-before-merge checklist and the history that usually fails to migrate, from field history to old emails.
Migrating between GitHub, GitLab and Bitbucket: keeping pull request review history
Migrate GitHub, GitLab or Bitbucket pull request history without losing review comments, approvals and links. What importers keep, and how to archive first.
Migrating helpdesks: what to do with tickets you will not migrate
A helpdesk migration leaves old tickets behind. How to decide what to migrate, archive or license, and how to export full history before the old plan ends.
Moving from Jira to Linear: what history you lose and how to keep it
Jira to Linear migration history: which fields, changelogs and links survive an import, and the archive rule that keeps the rest usable after cutover.
Moving from Slack to Teams (or back): what happens to message history?
Slack to Teams migration history rarely moves intact in either direction. Compare migrating, exporting to an archive and keeping the old tool read-only.
Moving on-premise customers to your cloud product: a vendor's data checklist
Migrating on-premise customers to cloud makes you the custodian of their data. Check the DPA, hosting region, exit terms and which records stay yours.
NAIC AI model bulletin: what insurance software vendors must give carriers
The NAIC AI model bulletin binds insurers, who pass duties to vendors. See the documents carriers request, from model inventories to training data provenance.
No-AI-training clauses in customer contracts: what B2B companies are seeing
No-AI-training clauses now appear in MSAs, DPAs and AI riders. See the common patterns and what each one leaves open for licensing your records.
Non-dilutive ways to raise cash for an established company
Non-dilutive funding options compared: term debt, revenue-based financing, receivables, tax incentives, asset sales and data licensing, by cost and speed.
Notion and Confluence: separating confidential HR and finance pages before a data review
Exclude confidential Confluence and Notion pages before a data review: triage each space by permissions, labels and owners to find HR and finance pages.
Old codebases in COBOL, Delphi or VB6: do AI developers want them?
Legacy COBOL, Delphi or VB6 code can have AI training value when it is documented, tested and paired with migration history. How to judge your codebase.
Old SVN, Perforce or TFVC history: is it worth converting?
Is old SVN, Perforce or TFVC history worth converting to Git? When it pays off, steps that keep authors, dates and messages, and checks to run first.
On-premise software vendors: you may hold less data than you think
On-premise software vendors rarely hold customer databases. What you hold is support, engineering and implementation records, which can still be licensable.
Open source code in your repos: what to check before licensing it for AI
Open source license terms follow third-party code into an AI training package. Learn to inventory permissive, copyleft and vendored code and what to exclude.
Open-source code inside your repositories: what must be excluded
Open-source code in proprietary repos must be excluded from a data license or handled on its own terms. See the screen list and how exclusion reaches history.
Open-source license audit before licensing your codebase
An open source license audit for your codebase before a data license: scan, classify permissive, copyleft and unknown code, exclude it and document it.
Opsgenie end of support: how to keep your alert and incident history
Opsgenie end of support migration: which alert and incident history to export, how to verify the export, and why the records keep value after the switch.
Opt-in vs opt-out for AI training in B2B SaaS contracts
Opt-in vs opt-out for AI training in SaaS contracts, compared on customer trust, legal exposure and how much customer data each model lets you license.
Processor or controller? Why SaaS companies usually can't license customer personal data
SaaS processor or controller status shapes AI training rights. See why customer personal data is usually off limits and which records you actually control.
Product requirement documents and roadmaps: licensable or too sensitive?
PRD and roadmap confidentiality decides what a SaaS company can license: shipped specs are often fine, cancelled ideas need review, future plans stay out.
Product usage analytics: can SaaS telemetry be licensed?
Product analytics data licensing turns on usage data clauses. See how aggregate, customer-level and user-level SaaS telemetry differ, and what to check.
Property management software vendors: what records you can license
Property management software data ownership, sorted: vendor records, landlord data and tenant personal data, and which of them a proptech vendor can license.
Proprietary algorithms in your code: exclude or include?
Proprietary algorithm code and trade secrets in a code license: a decision rule and three-tier scope that keeps core IP out and the surrounding workflow in.
PSA and RMM software vendors: licensing MSP ticket data
PSA and RMM software data licensing involves three parties. See what a vendor can license alone, what needs MSP opt-in and what end clients control.
Records under legal hold: can they be licensed?
Legal hold and data licensing: why originals must be preserved untouched, when counsel may approve licensed copies, and how to scope holds out of a package.
Redacting customer company names and deal details from B2B records
Redact customer company names from B2B data with an entity list, consistent pseudonyms and domain patterns, so records stay linked and customers stay private.
Rejected pull requests: why abandoned code changes are valuable
Rejected pull requests data shows AI teams what a reviewer turned down and why. Learn which closed, unmerged PRs to keep and how to scope them for a license.
Retiring a product sold on perpetual licenses: what you still owe
Perpetual license end of life obligations: what customers keep, when maintenance can end, and which notices to send before you retire a software product.
Revenue options for an end-of-growth software product compared
How to monetize a legacy software product: extended support, migration, sale, managed sunset and licensing its records, compared on revenue, effort and risk.
Runbooks as AI training data: what IT and operations teams hold
Runbooks become AI training data when linked to the incidents where they were used. What IT and ops teams hold, and what to remove before licensing.
SaaS shutdown: what you owe customers about their data
A SaaS shutdown customer data deletion checklist: notice dates, export windows, deletion certificates and the records that stay yours after customers leave.
Salesforce field history retention: preserve history before it disappears
Salesforce field history retention is limited. Learn which history objects to export, how to keep IDs linked and why the change trail holds the value.
Selling a cybersecurity software company: telemetry rights buyers check
Cybersecurity company acquisition due diligence checklist: training rights in customer terms, data residency, model lineage and the telemetry red flags.
Selling a failed startup's assets: code, data, domains and IP
Sell failed startup assets the right way: which to sell outright, which records to license instead, and the rights checks for code, data, domains and IP.
Selling a legacy product to a legacy-software specialist acquirer
How to sell a legacy software product to a specialist acquirer: who these buyers are, what they pay for, and which records to hand over versus keep.