Goals, Objectives, Strategies, Tactics: A Planning Framework That Works Because It Stays Small

GOST Planning Framework — Carousel.pdf

Download PDF • 1.49MB

By Paul Shotton, Advocacy Strategy

Something has shifted in how senior leaders ask public affairs teams to account for themselves. The question used to be “what are you working on?” Now it is closer to “what are you trying to change, and how will you know whether you are winning?” That is a harder question, and most teams cannot answer it in one breath. They can list activities — meetings taken, papers filed, consultations answered. What they struggle to state, cleanly, are the goals those activities serve, the objectives beneath them, and the indicators that would show progress.

That gap is the most common reason public affairs functions find it hard to demonstrate value to a board, to members, or to funders. The good news is that the discipline which closes it is neither new nor complicated. If anything, its whole value lies in staying small.

Why a framework, and why this one

Peter Drucker named the underlying problem in 1954. He called it the activity trap: the tendency of managers to become so absorbed in activity that they lose sight of its purpose. Public affairs is unusually prone to it, because the work is relational and continuous, and because being busy feels like being effective.

The framework most useful for escaping the trap is GOST — Goals, Objectives, Strategies, Tactics — popularised by Rich Horwath in his 2014 book Elevate. Horwath’s point was simple: managers were using these four words interchangeably, and as a result were planning at the wrong level. GOST forces each word to mean one thing. It is a synthesis of older management thinking rather than a new theory — Richard Rumelt’s work on what actually makes a strategy strategic sits underneath it — but its practical contribution is the insistence that the four layers are distinct and ordered.

You will sometimes see a fifth letter added. I have used variants with an extra layer in client work, but there is no basis for it in the original framework or the wider management literature, so I leave it out here. Four layers are enough.

The point is simplicity

Before defining the layers, it is worth being clear about what the framework is for. GOST is not a device for adding structure. It is a device for reducing it. A good cascade has a small number of goals, one objective per priority file, strategies drawn from a fixed and limited set, and a deliberately thin layer of indicators. Every one of those is a constraint, and the constraints are the point. A framework that lets you write twelve objectives and invent a new strategy for every issue has not organised your thinking; it has simply given your sprawl somewhere to live.

The four layers

A goal is a long-term, directional outcome in a defined policy domain — the way your organisation’s mission applies to that domain. Goals are qualitative and multi-year, and they are not directly measurable. Their job is to set direction, not to be tracked quarterly. Keep them policy-facing, and keep them few. The temptation to add an internal goal, such as “build the team’s capability”, is best resisted; capability matters enormously, but it is the scaffolding beneath the framework, not a goal within it.

An objective is where direction becomes commitment. It is a specific, time-bound outcome whose achievement advances the goal. The most useful habit here is to write an objective in two horizons at once: a standing outcome that holds across the life of the file, and a dated near-term focus that says what you are actually pursuing this year. One sentence carries both. You do not need a separate short-term objective and long-term objective; you need one outcome with a focus line you refresh annually. This is also where the familiar disciplines belong — SMART, from Doran’s 1981 checklist, for testing whether a single objective is well formed, and OKRs, popularised by John Doerr, for teams that prefer to pair an ambitious objective with a few measurable results. Both are tools inside the objective layer, not alternatives to the framework, and both descend from Drucker’s management by objectives.

A strategy is the chosen route to the objective — the guiding policy, in Rumelt’s terms, that selects some levers and excludes others. This is the layer where proliferation does the most damage, and where the simplicity discipline matters most. The practical fix is to stop inventing strategies file by file and instead draw them from a small, fixed set of standing pillars: direct institutional engagement, member or national-association mobilisation, coalition and value-chain building, technical evidence, and thought leadership. Most files use two or three. The menu stays stable; only the combination changes. That single decision is what keeps a public affairs plan from dissolving into a list of unrelated initiatives.

A tactic is the concrete deliverable that implements a strategy: an owner, an action, a deadline, and a target. The test of a good tactic is whether it names who will do what, by when, and aimed at which forum or stakeholder.

The order is not decorative. Each layer is the test of the one below it. A tactic is only a good tactic if it serves a chosen strategy; a strategy is only sound if it advances a defined objective; an objective only earns its place if it operationalises a goal.

A worked example

Take a manufacturer of finished goods that depend on a material covered by the EU’s Carbon Border Adjustment Mechanism — steel, say. The example is deliberately generic; it is the shape that matters.

The goal is directional: secure a carbon-border regime that protects European competitiveness without penalising downstream manufacturers that rely on covered inputs. The objective makes that concrete across two horizons. The standing outcome is to ensure CBAM’s scope does not place a disproportionate cost on finished products through their covered inputs, and to prepare for any future extension to finished imports. The near-term focus for the year is to shape the current downstream-extension proposal so the covered input is treated proportionately, and to clarify the route for the later imported-product question.

From the fixed pillar set, three strategies carry most of the weight: technical evidence, direct institutional engagement, and coalition-building. The tactics follow directly — a technical note on the input’s role in the product, circulated to the relevant Commission units by a set date; a defined number of targeted meetings with the rapporteur and the Council working party; and a coalition submission agreed with affected partners and priority member states. Nothing here is invented. Each tactic inherits its meaning from the strategy above it, and each strategy from the objective.

Making it measurable — without measuring everything

The question that usually derails this is measurement. Leadership wants numbers, so teams push hard numbers to the top and end up counting tactics while calling them strategy. The fix is to assign indicators by layer, and to accept that the character of the indicator should change as you move up.

GOST layer	What you measure	Indicator type
Goal	Long-term direction	Qualitative, narrative
Objective	Policy outcome + near-term focus	One light outcome indicator
Strategy	Whether the route is influencing the file	Outcome / influence indicator
Tactic	Whether the deliverable was produced	Process KPI

David Parmenter’s distinction between result indicators, which tell you whether you have succeeded, and performance indicators, which tell you whether you are likely to, maps onto this neatly, as does Kaplan and Norton’s older split between lagging and leading measures. The higher the layer, the more the indicator must capture meaning rather than motion — and the more qualitative it becomes. That inverts the usual instinct to put the hardest num

A closing thought

None of this requires new technology, but it does make new technology more useful. A team with a clean GOST cascade can ask an AI tool to draft candidate objectives from a priority list, propose indicators at the right level, or flag where the language has drifted back into activity. A team without one gets fluent output built on the same confusion it started with. The framework is the precondition; the tool is only the amplifier.

So before reaching for the next platform, it is worth asking the simpler question: can you state the goal, the objective, the strategy, and the tactic for your most important file — and say which indicator belongs at each level? If that fits on a single page, you are already in better shape than most.