NESDIS — the National Environmental Satellite, Data, and Information Service — is the 700-person, $3-billion federal agency inside NOAA responsible for operating the nation's weather satellites. Its own strategic plan committed it to a new paradigm: incorporating innovative approaches, pursuing emerging partnerships, and adapting to a rapidly changing technology landscape. By September 2021, with a record NOAA budget proposed and the IPCC issuing alarm-bell climate reports, the window for that transformation had never been larger, and the pressure to act had never been higher.
The problem was not ambition, and it was not analysis. The direction had been set, written down, and repeated. What the agency did not have was a mechanism for converting a stated direction into changed behaviour across 700 people who answered to different offices, different funding lines, and different disciplinary norms.
That strategic plan had been written in 2016. Five years later, meaningful progress had been difficult to sustain. NESDIS remained an organization of scientists and engineers working within their own disciplines and domains, with limited structural incentive to weigh the downstream effect of their work on colleagues or end users. Priorities accumulated faster than they were retired, and decisions that could have been resolved close to the work tended to travel upward instead.
None of that is unusual in a large technical agency, and none of it yields to instruction. The deepest barrier was not resistance but distance: the people being asked to change had never been asked to step back from their day-to-day work long enough to see the system they were part of. As one senior NESDIS leader put it:
Long-term adaptive capacity was not one obstacle but six interlocking ones, each owned by a different part of the enterprise. None could be resolved by the strategy office, or by any single office, alone. These six dimensions became the six topics the room was built around.
| Dimension | Why it could not be solved alone |
|---|---|
| Value proposition | The agency measured its value by the products and services it produced rather than the problems it solved, and had no common operating picture of the full mission end to end and no unified view of its users. Assembling one required the line offices and partners who held the other fragments of the picture. |
| Prioritization | A culture in which failure was not an option meant that everything was a priority and nothing could be stopped, and no shared criteria existed for saying no. Deciding what to stop reaches into budget lines, portfolio management and other line offices' expectations, so no single office could declare it. |
| Innovation | There was no structure to surface, evaluate or fund new ideas, and no shared definition of what innovation meant at NESDIS. Changing that required leadership, budget holders and staff to accept a different risk posture together; none of them could set it alone. |
| Whole system view | Strategic planning was framed inside organizational structures, budget lines were not shared, and offices had limited visibility of one another's work. A view of the whole system cannot be declared by one part of the system. |
| Partnerships | Partners, users and customers were conflated, partnering was slow and heavily formal, and no inventory existed of the agreements already in place. Making partnering faster required other agencies, cooperative institutes and commercial actors to accept new terms. |
| Workforce transformation | Knowledge sat with an ageing workforce with no mechanism for transfer, advancement paths for scientists uninterested in management were unclear, and much of the recruitment timeline sat outside the agency's control entirely. The levers are held by HR authorities beyond NESDIS. |
MMG convened 24 participants over three days via Zoom and MURAL on September 27, 28 and 30, 2021 — programme managers, scientists, budget and strategy staff, communications and data leadership from across NESDIS, together with colleagues from other NOAA line offices and from NASA. The selection was the design choice that mattered most: this was deliberately not the senior leadership group that had been trying to drive the change.
| Stakeholder | Role in the problem | Why their absence would have stalled the solution |
|---|---|---|
| Stars and change agents below the senior tier | Live the dysfunction daily and carry the most accurate diagnosis of why change had not stuck. | Four rounds of direction from above had already demonstrated the limits of top-down instruction. Recommendations built by the people who would have to change the behaviour carry a mandate that recommendations handed to them do not. |
| NESDIS programme and office directors | Own the operational products, ground services, data centres and system architecture that any change has to move through. | A transformation plan written around the programme offices rather than with them would have been technically naive and would have stalled at the first implementation step. |
| Budget, portfolio and strategic management staff | Control funding lines, the annual operating plan, and the five-year product portfolio process. | Recommendations that cannot enter the budget and portfolio cycle expire as good intentions. Every recommendation about stopping work depended on these people to make it real. |
| Other NOAA line offices | Are the internal customers whose needs define what NESDIS data is actually worth. | NESDIS could not define its own value proposition in isolation from the offices that consume its output. A value statement written without them would have described what the agency produces rather than what it is worth. |
| NASA | Holds the longest-standing and most consequential external partnership in the enterprise. | Half the partnership discussion concerned speed, duplication and shared visioning with NASA specifically. Reforming how the agency partners, with its largest partner absent, would have been theoretical. |
| The workshop sponsor, in a single Critic seat | Holds the authority to act on whatever the room produced. | A sponsor who is absent cannot own the output; a sponsor who participates normally dominates it. The Critic role gave full visibility into how the recommendations were built without allowing seniority to determine what they said. |
NESDIS had not been passive. By the time MMG was engaged, the agency had produced four separate sets of governance guidelines since 2018, brought in outside consulting support, and issued new direction from the top. The diagnosis in that work was not wrong. The pattern was that decisions were made at the top, communicated downward, and then quietly failed to change anything.
That happened for a structural reason rather than a lack of effort or goodwill. Analysis produces a correct answer; it does not produce an owner. Recommendations that arrive as external instruction ask people to change behaviour they were never consulted about, in a system whose constraints the recommendation may not have accounted for. In a 700-person technical agency, the rational response to such an instruction is to absorb it and continue — not out of obstruction, but because nobody in the room where it was written had a stake in making it work.
What was missing was not another diagnosis. It was a mechanism for the people who would have to live with the answer to build it themselves, under a structure that forced each proposal to be tested by colleagues who would bear its cost.
NESDIS had a mandate to transform, a strategic plan five years old, and a record budget environment — but priorities that only accumulated, a structure that resolved little close to the work, and a track record of agreeing on direction and then not moving. The three days convened around a single, deliberately time-boxed question:
The words that did the work were stop and 6 months. Asking what an organization should start invites an unbounded wish-list that costs nobody anything. Asking what it must stop, within six months, forces the room onto prioritization, budget, authority and risk — the things a strategy document cannot resolve on its own.
Six topics were the instrument — Value Proposition, Prioritization, Innovation, Whole System View, Partnerships, and Workforce Transformation — designed during discovery so the challenge would be opened along every seam at once: what the agency is uniquely for, what it should stop, how it generates and protects new ideas, how it sees itself as one system, who it works with, and who it needs to become. Six teams worked those seams in parallel rather than in sequence, which is what made thirteen working hours across three days sufficient.
Each team is split by role. Members own the topic and are accountable for producing three recommendations that answer the Question through the lens of that topic. Critics sit in the same session with a standing mandate to attack the work and are invited to do so at fixed points. Challenge is therefore structural rather than a matter of who feels safe speaking up — and because every participant sat on more than one team, the reasoning behind each recommendation travelled across topics instead of staying with the group that produced it. The sponsor held a Critic seat and no other.
Each topic ran the same three-stage discipline. In Analyze, teams surfaced the real constraints. In Diverge, they generated options widely before committing to any. In Converge, each team built the three recommendations it would own, and the full room allocated ten votes each across the resulting 18 recommendations to prioritize them. What follows is how that played out, topic by topic. The mechanics were identical; the content differed.
Analyze. This team's job was to articulate what NESDIS uniquely offers and to whom. The binding constraint was that the agency had no single answer. Value was being counted in products and services rather than in problems solved; there was no common operating picture of the full mission end to end; and there was no unified view of users, to the point that the team could not say with confidence who qualified as a NESDIS customer. Underneath sat a harder question the agency had avoided: what could it stop providing and still be the nation's trusted environmental data source?
Diverge. Options ranged from defining the high-value customers and building a process with them that could be refined over time, to making the obligations between NESDIS and the other line offices explicitly two-way, to developing a common language for value and an economic benefit analysis with NOAA's chief economist, to placing user representatives alongside users, to opening forums with external customers organized around large thematic problems rather than around products.
Converge. The team committed to three recommendations: leverage the agile value stream concept to prioritize and deliver value, defining at least two priority value streams with key partners; develop value statements across three dimensions — the issue view, the line office view, and the full NESDIS capability view; and use those value streams to refocus the User Engagement Council and other engagement groups toward value. The value stream recommendation ranked second of all 18 in the room's vote.
Analyze. This team worked on how the agency chooses, and the constraint it found was cultural before it was procedural. In an organization where failure was not an option, everything became a priority and nothing could be retired. There were no shared criteria for saying no and no transparency about who decided. The default was to prioritize whatever had been done before, and so much time went to commitments already made that little was left for planning or innovation. The team also named a habit that quietly sustained the problem: the agency rewarded heroics rather than consistent process.
Diverge. Options included prioritizing across line offices so NESDIS took on work only where no one else could; rallying around a NOAA-wide problem and working backwards to the capabilities required; making prioritization continuous with defined stop points rather than annual; delegating continuity decisions for lower-priority products to division-chief level; setting explicit risk tolerance at working and leadership levels; building a decision tree that made the risk at each level visible; and supporting staff emotionally through the loss of work they had invested in, so that stopping something was survivable.
Converge. The team committed to: clarify and articulate NESDIS priorities, emphasizing unique capabilities and disproving the assumption that the agency must do everything; create a clear mechanism, directive and enforcement to stop non-priority work, including follow-up to confirm that identified projects actually stopped; and refine the five-year product portfolio plan to enable real and adaptive prioritization, adding an impact focus to what had been a product-focused plan. The first ranked third of 18 and the second ranked fourth — the highest-placed pair from any single topic.
Analyze. This team examined why good ideas did not travel. The agency had no structure to showcase, evaluate or fund innovation, and no shared definition of what innovation even meant — whether it required a home run or whether many small improvements counted. Risk and innovation were treated as inseparable, which made every new idea a risk decision. Staff reported having no obvious place or support to bring ideas forward. And a quiet assumption ran through the organization that anything not built in-house was not good enough.
Diverge. Options included giving all staff dedicated time, budget and a safe space to develop ideas; writing innovation activity into annual operating plans and performance plans; a “screw up chip” that let someone take a swing without putting their performance rating at risk; hackathons and crowdsourced competitions run with partners; rotations into other offices and outside companies; prize competitions; judging scientists on service rather than only on publication; and reimagining NESDIS as a badgeless organization in which federal staff, contractors and cooperative institute researchers were developed alike.
Converge. The team committed to: make innovation a NESDIS core value, signalled by leadership as a transformational change and backed with defined budget, staff and time rather than encouragement; demonstrate that culture through diverse pilot project teams working real problems, deliberately staffed with people who do not normally work together; and create an innovation lab and “barrier buster” — a cross-functional team that takes ideas from any level, funds some, tracks progress, and clears the obstacles that stall them.
Analyze. This team worked on why a 700-person agency did not behave as one system. Strategic planning was framed inside organizational structures rather than across them, budget lines were not shared, and offices had limited understanding of what other offices did — which made it hard to see the value in anyone else's work. The team named the underlying pattern precisely: the organization operated like a university, on an investigator model where you protect your own projects and screen out everything else. Product orientation reinforced it, because a product belongs to somebody and a problem does not.
Diverge. Options included organizing around thematic problems rather than products; framing work around problem life-cycles, since related problems were being treated as independent; focusing on application systems instead of product systems to attack data interoperability at its root; standing up a designated team to identify and remove cross-office barriers; leveraging existing change champions and cross-office working groups rather than building new structures; and partnering beyond NOAA entirely — academia, private sector, NGOs, other governments — to reach a genuinely whole-earth perspective.
Converge. The team committed to: create a whole system view demonstration project, picking a thematic area such as algae or fire, assembling the necessary datasets in the cloud, and championing it through a cross-line-office integrated product team; help NESDIS staff embrace a whole system view through generalist hiring, rotational assignments, change champions and peer recognition; and align product portfolio management with the whole system view, moving from mission-unique to source-agnostic products with explicit plans to sunset some. Aligning portfolio management ranked seventh of 18 and the demonstration project sixth.
Analyze. This team found that the agency had not defined its own terms. Partners, users and customers were used interchangeably, which made it impossible to say what a partnership was for; the team's working definition — a partner brings something back, it is a two-way relationship — had not been shared. Creating a partnership was slow enough that NESDIS struggled to match the speed of the agencies it wanted to work with. There was no inventory of the agreements already in place, no clarity on who held authority to establish new ones, and no answer at all to the question of how to exit one.
Diverge. Options included matching resources honestly to partner commitments; defining what a better partnership looks like and piloting it on one or two smaller-scope relationships; stopping work a partner was already doing; bringing in social scientists to understand downstream societal impact rather than only economic value; leveraging the agency's regional presence; investigating whether an interagency agreement could be negotiated in two to three months instead of the prevailing timeline; and establishing tiers of formal and informal agreement so that a short-term collaboration did not require a permanent instrument.
Converge. The team committed to: assess partnership capabilities against two key NESDIS priorities using a gap analysis, explicitly targeting work the agency could stop doing and delegate; maximize the value of partnerships by engaging a broader set of partners across the whole value chain, including the last mile to the end user; and make partnering faster and easier through an accessible inventory of existing partnerships, a measurement matrix for effectiveness, and tiered agreements that speed up the start of work.
Analyze. This team was given an explicitly enabling brief: its job was to support the recommendations of the other five. The constraints it found were concrete. Knowledge sat with a workforce approaching retirement, with no mechanism to hire ahead and transfer it. Contractors were leaving for better conditions elsewhere, which the team read as the early warning rather than the problem. Scientists uninterested in management had no clear way to advance. And much of the recruitment timeline sat outside the group's control entirely, which meant the answer had to lie in what the agency could change: how it develops, empowers and retains the people it already has.
Diverge. Options included structured mentoring and a cadre of internal coaches; leadership at every level rather than a leadership track tied to pay; a rotational assignment programme with defined positions at multiple levels; pulse surveys to measure engagement continuously rather than episodically; a unified inclusion strategy backed by training and follow-up coaching rather than piecemeal programmes; and a mechanism to staff a retiring person's position before they leave.
Converge. The team committed to: develop leaders at all levels and hold them accountable, redefining a leadership position as an obligation to develop and empower others and embedding that expectation in performance plans; maximize workforce flexibility to increase retention, recruitment and inclusion; and attract new talent pools from underrepresented groups through an accelerated recruitment strategy and recruitment ambassadors across the agency. Workforce flexibility was the single highest-voted recommendation of all 18 — the room's clearest signal, from a group that had been convened to talk about strategy, that the constraint they felt most acutely was people.
Across the six topics the room produced one coherent strategy rather than six wish-lists. The 18 recommendations clustered into three themes and, when the room allocated its votes, the priority order was unambiguous: prioritize activities based on desired impact and unique value proposition (94 votes); adopt a culture of innovation, flexibility, cognitive diversity and accountability (81); and maximize impact with a flexible, adaptable workforce and partner network (65). Read together, the three form a single sentence — this mix of people needs to work together in this way and focus on this work.
Leadership left with 18 prioritized recommendations and three things that years of strategy documents and outside analysis had not produced:
In the post-workshop survey, 82 per cent of respondents agreed the format helped the group make significant progress in a short amount of time, 88 per cent endorsed the recommendations of the teams they sat on, 88 per cent wanted NESDIS to move forward with them, and every single respondent said they would take part in another workshop run this way.
What participants pointed to was rarely the content of any single recommendation. It was the composition of the room and the discipline imposed on it — the experience of working a problem alongside people they had no routine reason to meet.
The effect on how NESDIS was regarded elsewhere in NOAA registered immediately, and for an agency whose transformation depended on partners choosing to work with it differently, that shift was not cosmetic.
Because the people expected to change their behaviour had no ownership of the decisions. Analysis diagnoses correctly and recommends sensibly, but the recommendations land as external instruction — and in a 700-person federal agency, instructions written without the people living the constraints get absorbed rather than acted on. Not out of obstruction: the recommendation may simply not account for what the work actually requires, and nobody who wrote it is present to find out. The approach here was the inverse. Convene the internal change agents, let them build the strategy, and give the output organizational weight by connecting it directly to the leadership meeting where decisions get made.
Because the senior leaders had already demonstrated the limits of top-down direction. Several rounds of strategy documents and outside support had produced nothing durable. The people with the most accurate account of why change kept stalling were the ones living it — and they had never been asked to build the plan. Convening them, rather than the leaders who would ultimately approve the output, was itself a structural choice about what kind of mandate the recommendations would carry into the room where they were decided.
The sponsor was not excluded, but sat in the room for all three days in the designated Critic role, with a standing mandate to challenge the work and full visibility into how every recommendation was built — but without a Member's authorship. That distinction matters in both directions. It prevented seniority from quietly determining what the room concluded, and it meant the sponsor had watched the reasoning develop rather than receiving a document at the end. Recommendations are easy to dismiss when you first encounter them finished.
By designing the output to connect to decision-making structures that already exist. The engagement was structured so the 18 recommendations fed directly into the agency's Fall Strategy Meeting — a five-day gathering of 50 senior leaders — rather than producing a standalone report that would compete for attention with everything else on a leadership agenda. A workshop cannot manufacture authority. It can position its output where authority is already being exercised.
The typical model positions the consultant as the diagnostic authority and the client as the recipient of recommendations. That works for technical problems with knowable answers. It does not work when the real barrier is that the people who need to change have no stake in the decisions being handed down. The question worth asking before any engagement is not who will do the analysis. It is who will build the strategy, and what it will cost them if it fails.
If your organization has a strategy it believes in and cannot convert into changed behaviour — where the direction is set, the analysis exists, and the movement does not follow — that is the conversation worth having. Get in touch.