Delivery risk
Velocity is up. Trust is down. Teams comply in meetings and reject the system in practice.
- Green dashboards, fragile delivery
- Ceremony load crushing the day job
- Best people quietly leaving
For CEOs, COOs, CHROs, and transformation sponsors
Your rollout can look green while your people are breaking.
If your rollout looks green and your people do not, you are in the right place. I built the Human Agile model and Index after years inside enterprise agile work. Below you will find the budget gap, what Jira-style transparency does to team freedom, and what our survey keeps revealing. No firm behind this yet. Just me. Background on anjishbhondwe.com.
What keeps leaders awake
These are not maturity scores. These are the things that keep showing up when the programme office thinks it is winning.
People risk
HR sees stress the programme office does not measure. Retros become tracking, not learning.
- Burnout treated as individual failure
- Honest feedback punished
- Training that never sticks
Scale risk
The next rollout copies what looked successful on slides, not what actually worked with people in the room.
- Framework bought before humans consulted
- CoEs optimizing reporting, not enablement
- Dark Agile spreading across business units
The tabernacle of transformation
Most enterprises I work with are not running one change programme anymore. Agile delivery and AI adoption arrive together, often before the last wave has settled. I call that combined pressure the transformation tabernacle because agile without AI is incomplete and AI without agile discipline is incompetent.
Boards pay for these programmes and regulators ask hard questions, but in the room the load sits with deliverers. Industry scorecards can still look fine while humans absorb the strain. I built this site to document that gap before the next scaling decision.
The socio-technical tension
Every transformation runs two systems. The technical one is loud: tools, velocity, copilots, compliance dashboards. The social one is human: trust, safety, belonging, honest feedback. When only the first gets funded, people pay. That is the design failure behind much of the scrutiny you are seeing now.
Interviews name the harm in different language on each side, from Dark Agile, forced agile, and fake agile on delivery to Dark AI, forced AI, and fake AI on adaptation. The pattern repeats. Tool governance is where I see sponsors lose the plot most often.
When transparency becomes control
Jira, portfolio views, and funding dashboards gave sponsors a live read on every sprint and spend line. That can be fair. It turns forced agile when paying for the programme is treated like owning the team.
Funding a rollout is not the same as owning the humans inside it.
Legitimate sponsor role
- Fund outcomes and guardrails
- Review risk at human gate checkpoints
- Ask whether value and safety are both emerging
- Hold leaders accountable for system design
Overreach (Dark Agile pattern)
- Monitor micro-spend and velocity continuously
- Use dashboards to override team judgment
- Treat retros as tracking, not learning
- Punish honest signal as programme disloyalty
Before these tools, sponsors saw milestones and spend at programme level. Now leadership often sees every epic and burndown in real time, while teams get no equal say on whether the programme is safe to speak in or sane to carry. Visibility went up for sponsors. It did not go up for trust.
I am not against accountability. I am against execution funding becoming control over daily team life. The human gate exists because programmes skip that conversation once the board goes green.
Are your dashboards for learning, or for judgment? The Human Agile Index™ asks that before the next scaling wave.
Human impact under the tabernacle
Anonymous field interviews across telecom, financial services, retail, and tech. Not industry benchmarks. What people said the combined transformation did to them.
Strong teams shutting down while dashboards stay green is not rare. AI mandates stacking on unfinished agile debt is becoming normal. The charts below map where attention goes and what that costs people. Full definitions sit under each chart.
Agile face of the tabernacle
- Dark Agile dashboard theater
- Forced agile budget-dictated rollout
- Fake agile ceremony without belief
From programme reviews I sat in. Mirrors the 37-point people gap on the home page.
Share of interviewees who named each impact. Not a census of all IT workers.
Burnout or sustained overload
Chronic exhaustion, sprints stacked on unfinished work, crying in bathrooms while dashboards stay green.
Motivation loss
Comply in meetings, reject the system in practice. Ceremony without belief.
Continuity risk
Attrition intent, knowledge walking out, best people managed out after honest retros.
Top performers, parallel strain
Teams leadership flagged as high velocity who reported shutdown under forced rollout waves.
AI face of the tabernacle
AI landed before agile strain healed. People described mandates without training, efficiency targets without safety, and fear that honest feedback would be read as resistance.
- Dark AI board-slide theater
- Forced AI mandate-dictated rollout
- Fake AI licenses without learning
From AI readiness reviews I sat in. Mirrors the same people gap pattern: tools first, humans last.
Share of interviewees who named each AI adaptation impact. Often the same people already reporting agile strain.
Workload stacking
AI adoption added on top of the day job and the agile programme. Same headcount, more output expected, no time removed.
Skill anxiety
Copilot licenses rolled out without training. People told to upskill on their own time while measured on usage quotas.
Trust erosion
Fear that honest questions about AI readiness will be read as resistance or replaceability. Silence in the room while the board deck shows progress.
Top performers, parallel strain
High-output teams asked to lead AI pilots while still carrying transformation debt from the agile rollout.
How we ran the survey. Anonymous interviews across telecom, financial services, retail, and technology. Percentages are share of respondents who named each impact, not a census of all IT workers. Budget figures come from programme and AI readiness reviews I attended. Agile and AI are mapped as two faces of the same transformation tabernacle. Negative impact lands on humans on average, not on industry scorecards. This is our field read, not peer-reviewed industry data. Cite: agilefragile.com/leaders.html#human-impact.
What our survey reveals
Anonymous voices from our field survey across sectors. These are not complaints about agile or AI. They are descriptions of what happened when the social system broke and the dashboard stayed green.
"I was burnt out before the transformation. After it, I had sprints on top of the work I already could not finish."Engineer, financial services
"We had a two-hour agile training. Then we were measured on velocity every week. Nobody asked if we understood why."Product team member, retail
"I raised concerns in the retro. Two months later I was managed out. The retro is not for learning here. It is for tracking."Developer, telecom
"Leadership sees green dashboards. We see people crying in the bathroom. Both are true in the same building."People lead, technology
"They rolled out Copilot to everyone on a Monday. By Friday we had usage targets. Nobody asked if we had time to learn it on top of the agile programme."Engineer, technology
"The AI steering deck shows 90% adoption. Half my team is afraid to say the tools are making more review work, not less."Tech lead, financial services
Leaders and scholars who put humans first
Human-centric agile is not a brand position. It is what the founders, practitioners, and researchers said when they still had the room's attention.
Individuals and interactions over processes and tools.Beck, Cockburn, Fowler, Jeffries, Sutherland, and twelve others
Psychological safety is a belief that one will not be punished or humiliated for speaking up with ideas, questions, concerns, or mistakes.The Fearless Organization, 2018
People's characteristics are a first-order success driver, not a second-order one. In fact, I have reversed the order, and now consider process factors to be second-order issues.Characterizing People as Non-Linear Components, 1999
If you want to succeed in doing software development, you need to find good people that work together at a human level, so they can collaborate effectively. The choice of what tools they use or what process they should follow is second order.The State of Agile Software, 2018
The hard part of software design is navigating the human relationships, full of conflict and conflicting incentives.Software Design is Human Relationships, 2023
The only thing of real importance that leaders do is to create and manage culture.Organizational Culture and Leadership, 2010
What the model offers leaders
The model offers a human-first path through the transformation tabernacle: name what is breaking people, remove what harms them, and scale only when the human gate clears. The ethic and adoption steps are on the model page.
Who is behind this
Anjish Bhondwe
Agile Fragile is the public home of the Human Agile model. I am proposing it based on years of enterprise agile scaling work across federated organisations, CoEs, and multi-country programmes.
For full background, credentials, and client work, visit the personal site. There is no team page here because there is no team yet.
A straight conversation
No pitch deck on the first call. I listen to what is happening to your people. Share your Human Agile Index score if you have one. I will tell you honestly if a conversation makes sense.
Start a conversation