Managing Intellectual Property in MarTech and AdTech
MarTech and AdTech companies need to manage the technology and data behind targeting, bidding, creative generation, conversion measurement and budget allocation, not just the screens on which advertisements appear. Patents, trade secrets, trademarks, copyright and contracts serve different roles.
Start with disclosure and reverse-engineering risks
| Asset | Main protection | Management considerations |
|---|---|---|
| Bidding, targeting and measurement technology | Patents or trade secrets | Assess whether processing can be observed externally and copied after disclosure. |
| Training data, weights and experimental results | Trade secrets and contracts | Set access permissions, export logs, confidentiality markings, retention periods and employee-exit procedures. |
| Service names, feature names and logos | Trademarks | Specify goods and services appropriate to launch countries and the services provided. |
| Dashboards, reports and advertising creative | Copyright, design rights and contracts | Record creators, source files, modification rights, delivery terms and permitted reuse. |
| APIs, SDKs and external data | Licences and usage agreements | Check conditions for collection, combination, model training, third-party disclosure and handling when the service ends. |
Describe data processing, not merely advertising operations
Do not describe automated bidding only as a way to spend a budget efficiently. Specify how campaign objectives, remaining budget, time of day, conversion delays, impression opportunities and user signals are received, how a bid is calculated, and how results update the next bid.
Potential patent subjects include invalid-click detection, attribution, customer segmentation, creative-to-placement matching, frequency control, budget allocation and performance prediction. Consistent with Korea's official business-method invention guidance (in Korean), distinguish business rules from the technical configuration of the computer system.
For AI features, describe input data, preprocessing, outputs and resulting system actions rather than only the model name. Implementation differences—such as training a model for each advertiser, adapting a shared model or handling missing and delayed conversions—provide the basis for claims.
External advertising platforms are dependencies
Platforms such as Google Performance Max, Meta Advantage+ and Naver Ads continue to change automation features and available interfaces. Descriptions of a platform's functions should not be treated as your company's own technology.
Focus protection on processing your company implements and controls: normalizing platform data, converting metrics across media into comparable values, or compensating for API errors and data delays. Describe alternative inputs and exception handling that remain useful when APIs are discontinued, fields change or tracking is restricted, rather than relying on platform names or policies.
Check data rights in contracts
Advertiser CRM data, platform performance data, website and app events, and third-party data have different sources and usage conditions. Possession alone does not permit reuse for model training or another client's analysis.
Contracts should address who collects the data, processing purposes, model-training use, ownership of derived statistics, subcontracting, retention, deletion or return, incident response and post-termination use. Review privacy obligations and IP ownership clauses separately, while managing them together through a data-flow map.
Record open-source, API and generative-AI use
Maintain names and versions of external libraries, models, datasets, APIs and SDKs in development records. At deployment, check licence notices, source-disclosure obligations, commercial-use restrictions and redistribution conditions.
When AI generates advertising creative, clarify rights in the inputs, responsibility for reviewing outputs, possible inclusion of third-party marks or works, and ownership between the advertiser and producer. Keep prompts and generation histories linked to output-review records.
Pre-launch management checklist
- Document inputs, processing, outputs and external integrations for each feature.
- Distinguish externally observable technology from confidential operational know-how.
- Identify inventions for patent review before proposals, demonstrations or public product releases.
- Check service names and key feature names for trademark conflicts in each launch country.
- Align data use among advertisers, platforms and analytics providers with their contracts.
- Regularly assess how platform-policy and API changes affect existing rights and product behaviour.
