A Strategy Is Not a Plan

Read any localisation strategy published in the last decade and you will find the same commitments. Shift decision making closer to the people affected. Increase direct funding. Build local leadership, and where the strategy is government facing, build national ownership too. The language has converged across the humanitarian and development sector as a whole. What has not converged, nearly as consistently, is what happens after the strategy is signed off - and that gap tends to show up in the same three places. One at the moment a strategy is written, one at the moment the strategy is meant to be delivered, and one before either of those (at the moment someone decides a strategy is even needed).

Its important being precise about terms before going further. Localisation is often read as shorthand for local NGOs and civil society alone. Localisation never meant only that. The Grand Bargain's own definition covers "national and local responders comprising governments, communities... and local civil society," meaning national authorities sit inside the term, not beside it, and the definition also covers the work of international organisations whose job is strengthening those authorities' own systems. This piece uses localisation as that full umbrella, and national ownership as the specific term for the government layer within it, because all three failures below show up identically whether the strategy in question belongs to a Government ministry or a local NGO.

The first failure: mistaking the plan for the strategy

Part of why so many strategies stall starts even before the launch, at the moment the strategy is actually written. The word "strategy" is being asked to do two different jobs at once. Henry Mintzberg stated the distinction plainly: strategic planning is not strategic thinking, one is analysis and the other is synthesis. Strategy is the judgement call - deciding what to prioritise, what to stop doing, and what success actually requires. The plan is what follows that judgement - the timeline, the budget lines, the results framework, the document that gets printed and launched.

The plan and the strategy are two different skills, and treating them as a single skill explains more of what goes wrong than any single resourcing or capacity problem does. A plan can be well written, thoroughly consulted, and entirely professional, and still not represent a real strategic choice, because the harder judgement a plan is supposed to encode (what the organisation should actually prioritise, and what graduation or national ownership would genuinely require) was never made, or was made once and simply carried forward unexamined. When that happens, what gets renewed every few years is not the strategy. The plan is what gets renewed.

Tajikistan's mine action programme is a clear, documented example. Its national authority developed a first strategy covering 2006 to 2010, then a second covering 2010 to 2015, with support from international partners over that period. Both strategies were genuinely consultative processes, and by most measures the programme performed reasonably well against them. But a 2012 evaluation of the programme found that after these two full strategy cycles, the country still needed to begin planning an actual exit strategy. In plain terms, after roughly a decade of strategic planning, the question of when and how the programme would eventually stand on its own (aka graduate) had not yet been answered as a real decision. The question had simply been left for the next planning cycle to pick up. This finding says nothing about the competence of anyone who worked on those strategies. The finding shows what happens when a plan gets renewed on schedule while the harder question underneath the plan goes unexamined, cycle after cycle.

The second failure: mistaking the launch for the delivery

Even where the underlying strategic judgement is sound, a plan still has to survive contact with an actual day to day operations, and this is the point at which most localisation strategies quietly stop. A strategy gets a launch event, some initial momentum, and then nothing that tracks whether any of the commitments are actually being delivered a year later.

This half of the problem has already been solved elsewhere, which makes it more frustrating that the sector keeps failing to learn from that solution.

When the UK government hit the same wall on public service reform in the early 2000s (sound strategy but weak delivery) the government built a small team known as the Prime Minister's Delivery Unit, with one job: track a short list of top priorities against hard targets and trajectories, and clear whatever stood between a decision and the decision's delivery on the ground. The approach, now dubbed deliverology, has since been adapted by governments from Canada to Ghana to Colombia, and by philanthropies and multilateral organisations well beyond governments.

What makes a delivery unit work has less to do with where the unit sits on an organisational chart and more to do with a few disciplines. A narrow, named mandate rather than an attempt to fix everything at once. Real backing from leadership, alongside working relationships with the teams actually implementing. And an honest sense of the unit's own limits: a delivery unit clears bottlenecks and tracks progress, but a delivery unit cannot rewrite an organisation's procurement rules or rebuild an organisation's systems on its own. The good delivery units also plan their own closure from day one, with a clear route for handing their tools and habits to the permanent structure rather than becoming a permanent fixture in their own right.

Localisation plans, government and NGO alike, typically get a launch event and nothing resembling this kind of dedicated delivery function. What most localisation plans actually need is a small, accountable team whose only job is turning commitments into tracked, resourced, monitored action, for as long as it takes to turn theory into practice.

The third failure: mistaking the timing for a formality

The first two failures assume the strategy process itself was worth starting. That assumption is not always true, and this third pattern sits earlier than either of the other two, at the point someone decides a new strategy is needed at all. Strategies for specific technical fields, mine action, WASH, health, education, are usually commissioned in isolation from each other and from the country's own national development plans. Each strategy driven by its own donor and its own consultation timeline, with little structural connection to what the ministry next door is planning.

Ethiopia's experience with nutrition strategy shows what that isolation produces concretely. Nutrition is, by its nature, a problem that spans health, agriculture, and social protection at once, yet the ministries responsible for each of those areas have largely worked in silos, at times giving conflicting guidance to the same households. The community health workers expected to deliver the strategy on the ground carry a workload the strategy itself was never built around, and the data that should be feeding back into course corrections stays disconnected from the programmes the data should be shaping. The country ends up with well written national strategies and, at the community level, a delivery reality that looks nothing like those strategies.

This is a timing problem as much as a coordination problem. Strategy processes tend to get commissioned on the funder's or organisation's own clock, a grant cycle, a leadership change, rather than because anyone has checked whether the country or technical field in question is actually ready to connect a new strategy meaningfully to everything else already in motion in that country or field.

What actually fixes this

None of these three failures needs a new theory. Each failure has a known, workable fix. We have simply not been in the habit of applying all three fixes together.

For the first failure, the fix is separating the two conversations properly: doing the strategic judgement work explicitly, as its own distinct exercise, before the planning document gets written, and being willing to say that a graduation date or an ownership transition has not actually been decided, rather than letting a plan imply that the decision has been made.

For the second failure, the fix is the delivery unit model described above, adapted to scale: a small team, a short list of tracked priorities, and a defined lifespan rather than an indefinite one.

For the third failure, the fix is a readiness assessment done before a strategy process is commissioned, not after. A readiness assessment is a well established practice in change management more broadly, used to ask plainly whether an organisation or a country is actually positioned to take on a given change right now. The same practice translates directly to localisation: is this the moment, does this technical field connect meaningfully to what else is happening nationally, and is delaying a strategy process by six months a better outcome than launching that process on schedule and unconnected.

Held together, not bought separately

These three fixes rarely get applied as a set. An organisation might commission a genuine readiness assessment and still hand the resulting strategy to a team with no mandate to track its delivery. Another might build excellent delivery machinery around a plan that was never actually a strategic choice to begin with, simply the same document renewed on schedule. The fixes work as a package because each one closes a gap the other two leave open, and a strategy only survives its own launch when judgement, delivery, and timing are treated as one connected piece of work rather than three separate purchases.

That is the standard we have built ImpactRoot to work towards: strategic judgement that is honest about what is actually being decided, a delivery function that outlasts the launch, and a willingness to tell a client the moment is not right before recommending a process at all. A strategy that is renewed rather than rethought will always look, on paper, exactly like the one that came before it. The only real test is whether it survives contact with the years that follow its launch.

Sources: Emerald Publishing, study on localisation of logistics preparedness in international humanitarian supply chains (2024); Start Network / Humanitarian Advisory Group, "The Missing Link in Localisation" (2024); INTRAC, "How to do Strategic Planning: A Guide for Small and Diaspora NGOs"; evaluation of UNDP support to the Tajikistan Mine Action Programme, Journal of Mine Action / CISR; Delivery Associates, "What Makes a Delivery Unit Successful? What It Leaves Behind" (2025); PMC/NCBI, "Heavy on Plans, Light on Delivery: The Structural Failures of Ethiopia's Nutrition Policies" (2025); Henry Mintzberg, "The Fall and Rise of Strategic Planning," Harvard Business Review (1994); Grand Bargain, definition of national and local responders; change management literature on organisational readiness assessments.