Stelian ROMANProject Manager| MicroSafetyCarlingford, New South Wales, Australia
Agile thrives on fast feedback loops and rapid iteration. Teams use analytics, user tracking, and telemetry to inform decisions and deliver value quickly. But with these tools comes a critical responsibility: respecting the privacy of users whose data powers this feedback.
How does your team balance the need for rapid feedback with a commitment to user privacy?
Luis BrancoCEO| Business Insight, Consultores de Gestão, LdªCarcavelos, Lisboa, Portugal
An important and highly relevant discussion.
One of the strongest points in the article is the recognition that fast feedback loops and ethical responsibility cannot be treated as separate concerns in Agile environments.
I would add one critical layer: privacy is no longer only a compliance issue. In increasingly data-driven and AI-enabled organizations, it becomes part of the decision architecture of the product and the organization itself.
The real challenge is not simply collecting data quickly, but ensuring that learning speed does not outpace transparency, informed consent, trust, and ethical understanding of how user data is being used.
Otherwise, organizations may optimize iteration and insight generation while progressively eroding the very trust that sustainable Agile delivery depends on.
Strong and very timely contribution. Saving Changes...
Darline GiraudDriving AI adoption across multilateral institutions | Cross-functional program| IAEADigne Les Bains, France
I agree with this perspective, especially the point that privacy is becoming part of the organization’s decision architecture rather than just a compliance concern. I would add that if Agile delivery teams are having to determine these boundaries on their own, there may be a deeper foundational issue: the absence of clear data governance. Delivery teams should not have to improvise privacy standards sprint by sprint. There should already be organizational guardrails around data collection, consent, retention, access, anonymization, telemetry, and acceptable use. Agile teams can then move quickly, but within a framework that protects users and supports ethical decision-making. In that sense, the challenge is not only how teams balance speed and privacy, but whether the organization has created the governance structure that makes that balance possible in the first place. Fast feedback loops are valuable, but they need to operate within clear data policies and accountability mechanisms. Saving Changes...
Deepak MalhotraLeader - Customer Success| CiscoHyderabad, Telangana, India
I believe rapid feedback and user privacy should not be seen as opposing goals. In a mature Agile environment, both should work together. Fast feedback helps teams improve the product, but privacy ensures that the trust behind that feedback is protected.
In my view, the balance starts with collecting only what is genuinely needed. Teams should avoid gathering data just because it is available. Every data point should have a clear purpose linked to product improvement, user experience, quality, or business value.
A simple real-world example is a mobile banking or payment app. The product team may want to understand where users are dropping off during onboarding or while completing a transaction. Instead of capturing personal information such as names, account numbers, transaction details, or screenshots, the team can use anonymized event data such as “user exited at OTP screen” or “payment failed at confirmation step.” This gives the team useful feedback to improve the experience, while still protecting the user’s sensitive information.
Some practical steps that help are anonymizing or aggregating user data wherever possible, getting proper consent, limiting access to sensitive information, defining clear retention periods, and being transparent about how feedback or telemetry is used. For qualitative feedback, teams should also be careful not to expose personal details unnecessarily during sprint reviews, retrospectives, or stakeholder discussions.
I also agree with the point already shared that this should not be left only to individual Agile teams to decide sprint by sprint. Organizations need clear data governance, privacy guidelines, and accountability mechanisms. Once those guardrails are in place, teams can still move quickly, but within a responsible and trusted framework.
For me, the right approach is simple: use feedback to improve the product, but respect the user behind the data. Agile should deliver value quickly, but not at the cost of trust, consent, or ethical responsibility. Saving Changes...
Balance comes from anonymizing data at collection and using aggregated insights, not individual tracking. Privacy impact reviews happen before any new analytics tool. Speed matters, but trust matters more. On a different note, check out https://www.cameoglass.net/ for impressive glass craftsmanship.
The balance comes from anonymizing data at collection point and using aggregated insights instead of individual tracking. Feedback loops rely on behavioral patterns, not personal identifiers. Regular privacy impact reviews happen before implementing any new analytics tool. Speed matters, but trust matters more. Privacy-first design actually accelerates decision-making by removing red tape later. Saving Changes...
Stelian ROMANProject Manager| MicroSafetyCarlingford, New South Wales, Australia
Luis Branco I agree: privacy is no longer limited to compliance. I believe the focus on tools to 'protect' privacy is wrong. We should focus on the impact rather than fencing information. I had a situation where we couldn't use the driver's license to identify a patient and link their cancer records, but every shop in Australia will take a copy of your driver's license when you rent or buy on credit. Note: In Australia, there is no ID card. When you walk through CBD, your image will be stored on hundreds of servers. Even the basic FB AI Agent can identify a person but some organisations spend millions of dollars 'protecting' information that is available.
I agree that many discussions about privacy become overly focused on “protecting information” as if data could realistically remain isolated in modern digital environments.
In practice, individuals are already continuously exposed through platforms, devices, cameras, telemetry, identity systems, behavioural tracking, and increasingly powerful AI inference capabilities.
But I believe that makes governance even more important, not less.
Because the core issue is no longer only access to information.
It is the legitimacy, proportionality, transparency, and accountability surrounding how data is interpreted, combined, retained, monetized, and operationalized in decisions.
Two organizations may technically possess the same data and still behave very differently from an ethical standpoint.
That is why trust increasingly depends not only on privacy controls, but on whether users believe organizations are exercising responsible judgment with the informational power they possess.
In many ways, the ethical challenge is shifting from “Can organizations collect data?” to:
“How should organizations responsibly govern the use of what they already know?”
An increasingly important conversation for AI-enabled organizations.
Saving Changes...
Stelian ROMANProject Manager| MicroSafetyCarlingford, New South Wales, Australia
Darline Giraud , thank you. In a large IT organisation, privacy, from the IT point of view, is, or should be, part of either data/information governance or enterprise architecture. The challenge is protecting end users' privacy when the Scrum team wants to provide feedback. For example, a team is building a mobile app, and from the technical point of view, they will know the user's location. Can they store that information or use it to contact the user? Saving Changes...
Stelian ROMANProject Manager| MicroSafetyCarlingford, New South Wales, Australia
Deepak Malhotra, very good point: "collecting only what is genuinely needed". I don't believe that it is that easy... I still can't understand why a job ad is asking for your cultural background, sexual preferences or full home address. Saving Changes...
Stelian ROMANProject Manager| MicroSafetyCarlingford, New South Wales, Australia
Henry Nobles, as Luis Branco mentioned. It is not about the tool; it is the principle, the team and organisation culture, personal value and at the end of the day, ethics. If we want to use the information for whatever purpose, we should inform the owner when the information is collected. Why should we spend money and time anonymising information that we have the consent to share? Saving Changes...
Luis BrancoCEO| Business Insight, Consultores de Gestão, LdªCarcavelos, Lisboa, Portugal
May 23, 2026 1:12 AM
Replying to Stelian ROMAN
...
Luis Branco I agree: privacy is no longer limited to compliance. I believe the focus on tools to 'protect' privacy is wrong. We should focus on the impact rather than fencing information. I had a situation where we couldn't use the driver's license to identify a patient and link their cancer records, but every shop in Australia will take a copy of your driver's license when you rent or buy on credit. Note: In Australia, there is no ID card. When you walk through CBD, your image will be stored on hundreds of servers. Even the basic FB AI Agent can identify a person but some organisations spend millions of dollars 'protecting' information that is available.
I agree that many discussions about privacy become overly focused on “protecting information” as if data could realistically remain isolated in modern digital environments.
In practice, individuals are already continuously exposed through platforms, devices, cameras, telemetry, identity systems, behavioural tracking, and increasingly powerful AI inference capabilities.
But I believe that makes governance even more important, not less.
Because the core issue is no longer only access to information.
It is the legitimacy, proportionality, transparency, and accountability surrounding how data is interpreted, combined, retained, monetized, and operationalized in decisions.
Two organizations may technically possess the same data and still behave very differently from an ethical standpoint.
That is why trust increasingly depends not only on privacy controls, but on whether users believe organizations are exercising responsible judgment with the informational power they possess.
In many ways, the ethical challenge is shifting from “Can organizations collect data?” to:
“How should organizations responsibly govern the use of what they already know?”
An increasingly important conversation for AI-enabled organizations.
