What is strategy, really? The ultimate guide for marketers and non-marketers alike

Strategy is the general's craft, not the soldier's

The word comes from the Greek stratēgós, a general, not the soldier who swings the sword fastest. A general does not win by fighting everywhere, but by choosing the battlefield where they have the upper hand and pouring in more force there than anywhere else. And that is exactly what most companies have lost: strategy is not a list of everything you want to get done, it is the decision of what you give up to win, when budget, time and talent are always in short supply.

First the uncomfortable part: what strategy is NOT

Nine out of ten documents proudly titled "strategy" are no such thing. They are wishes in a nice font. Before we get to what strategy is, let us clear up what it is routinely confused with:

• A goal is not a strategy. "We want to double revenue" is an ambition. Strategy answers what will make that happen when it did not happen for your competitors.
• Vision and mission are not strategy. "Change the world for the better" is a poster in the canteen. Lovely, but it tells nobody what to do on Monday morning.
• A plan is not a strategy. A to-do list with deadlines is work organisation. It can be flawless and still point the wrong way.
• A list of tactics is not a strategy. "We will run PPC, TikTok, a newsletter and a podcast" is a shopping list of channels, not a reason any of them should work.
• Doing more and better is not a strategy. Operational improvement (faster, cheaper, higher quality) is necessary but copyable. Michael Porter puts it bluntly: operational effectiveness is not strategy.

In Good Strategy Bad Strategy, Richard Rumelt named four unmistakable signs of bad strategy: fluff, a mush of buzzwords that hides the fact that nothing is being said; failure to face the real problem; mistaking goals for strategy; and bad strategic objectives that miss the core. When a document has no single hard decision and no clear what we will not do, it is not a strategy. It is a noticeboard.

Strategy versus goal versus plan versus tactic

This is the most common confusion in the world, and it can be settled with a single table. Four terms, four different questions. The Czech marketing strategist Michal Krutis puts it elegantly: strategy answers what and why, tactics answer how. Add the goal (where) and the plan (when and how much) and you have the whole chain:

Roger Martin, one of the most cited strategists alive, adds a provocation: strategy is not planning. A plan is comfortable because it is under our control (we will do this and that). Strategy is uncomfortable because it is a bet on customer and competitor behaviour we do not control. That is why firms love to soothe themselves with thick plans instead of one bold decision. A plan gives you a feeling of certainty. Strategy gives you an advantage. They are not synonyms.

The anatomy of good strategy: Rumelt's kernel

The most usable definition of what a strategy must contain comes from Richard Rumelt. He calls it the kernel, and it has three parts that must sit in order:

• 1. Diagnosis. Naming the true nature of the problem. Not the symptom ("revenue is falling") but the cause ("our offer is indistinguishable from three cheaper rivals, so it comes down to price alone"). A good diagnosis reduces tangled reality to the one or two things that actually matter.
• 2. Guiding policy. The overall approach to dealing with the diagnosis. Not a goal, not a specific action, but a direction that rules out a lot of other directions. For example: "we stop competing on price and become the fastest, most convenient option for the impatient."
• 3. Coherent action. A set of steps that reinforce each other and all pull the same way. The key word is coherent. When every department runs "its own strategy", the result is a tug of war, not a strategy.

The beauty of the kernel is that it is merciless. If you cannot write the diagnosis, you do not have a strategy, you have a feeling. If your guiding policy implies nothing you say no to, it is not a policy, it is a slogan.

Where to play and how to win: Playing to Win

The second pillar of modern strategic thinking is the Playing to Win cascade by Roger Martin and A. G. Lafley (the former head of Procter & Gamble). For them strategy is an integrated set of choices that answers five interlocking questions. They are not independent boxes, every answer has to fit all the others:

• What is our winning aspiration? What winning means for us, not merely taking part.
• Where will we play? Which segments, categories, channels, geographies. And just as importantly: where we will not play.
• How will we win there? Through what advantage we prevail on precisely this field. Price? Experience? Speed? Brand?
• What capabilities do we need? What we must be above-average at for the win to be sustainable.
• What management systems support it? Which processes, metrics and structure must exist for it all to work day to day.

The most important pair is where to play and how to win. Most firms skip "where to play" (they want to be everywhere and for everyone) and jump straight to arguing about tactics. But when you play everywhere, you win nowhere. Choosing the field where you have a real chance is the single most strategic decision you will make.

Strategy means being different: Porter

Michael Porter, the father of competitive strategy, added an idea that sounds simple yet the vast majority of firms ignore it: the essence of strategy is choosing what not to do. Competitive advantage does not come from doing the same things as everyone else, only slightly better. It comes from doing things differently and being willing to sacrifice something else for it. Porter calls that trade-offs. Whoever refuses to sacrifice anything has no strategy, only an ambition to be best at everything, which is impossible.

Porter's tools worth knowing:

• Five forces. An industry's attractiveness is set not only by direct rivals but by supplier power, buyer power, the threat of new entrants and the threat of substitutes. Before you enter a market, run all five.
• Generic strategies. You win either on lowest cost, or on differentiation, or on narrow focus on a segment. Getting stuck "in the middle" (neither cheapest nor best) is the most common road to mediocrity.
• Activity system. Real advantage is not one thing but the way dozens of your activities fit together, so that copying you is not worth it for a rival because they would have to copy everything at once.

The types of strategy and how they nest

"Strategy" is not one thing. It is a hierarchy in which each lower level should flow from the one above it. When the levels come apart (marketing does something the business did not ask for, and creative does something marketing did not ask for), you get exactly the "every dog barks a different tune" that the Czech strategist Pavel Cahlik describes. The British strategist Julian Cole, who has taught over five thousand planners, built an entire school on this hierarchy. Here it is at a glance:

The golden rule: a lower level must never contradict a higher one. Brilliant creative that does not serve the communications strategy is money thrown away. Smart communications that contradict the brand destroy the brand. When someone sells you a "marketing strategy" and actually means a list of social posts, they have mistaken the ground floor for the whole building.

Where strategy is born: tension and insight

Frameworks tell you what structure a strategy should have. They do not tell you where to find the spark that makes a strategy clever rather than merely correctly filled in. This is where the contemporary strategists who pushed the craft forward come in. The Australian strategist Mark Pollard (author of Strategy Is Your Words and founder of the Sweathead community) champions two ideas worth tattooing:

• Strategy is your words. Until you can say it clearly and concisely, you are not thinking clearly. Fog in the language is fog in the thought. A sharp strategy is a sharp sentence.
• Strategy is born from tension. The best strategies sit on a contradiction: between what people want and what they get; between how the world pretends to work and how it actually does. Without tension there is nothing to hang an idea on.

Rob Estreitinho, a strategist at London's VCCP, extends this in his newsletter Salmon Theory: most good advice for writers is really advice for strategists in disguise. Clear writing forces clear thinking, and an unclear strategy is almost always just an unclearly written one. If you cannot sum your strategy up clearly in a few lines, you do not have a writing problem, you have a thinking problem.

Add the notion of insight, a non-trivial truth about the customer that competitors overlook. An insight is not a statistic ("80% of people use a phone"). It is an understanding of why people behave as they do, so precise that when you voice it the customer nods, "exactly, how did you know?". A good strategy almost always rests on one such insight and one tension. The rest is honestly working them through.

Hypotheses and research: where you get the evidence

The diagnosis we talked about does not fall from the sky. For it to be more than a clever-sounding opinion, it needs evidence. And because nobody knows the future, a good strategy is really a hypothesis: a considered bet on what will work, framed so it can be tested. Roger Martin has a useful turn of phrase for it, instead of "is this true?" ask "what would have to be true for this choice to work?". You then test those conditions cheaply before betting the whole budget on them. Eric Ries calls this the leap-of-faith assumptions in The Lean Startup, the riskiest beliefs that must be validated first, and Rita McGrath builds her discovery-driven approach on the same logic.

The types of research most often confused:

• Qualitative vs. quantitative. Qualitative research (in-depth interviews, focus groups, ethnographic observation) answers "why" and reveals motivations. Quantitative (surveys, panels, analytics) answers "how many" and measures how widespread something is. A handful of interviews is not proof of scale, and a big survey will not tell you why.
• Primary vs. secondary. Primary data you collect yourself for your specific question. Secondary (desk research) is reading data and studies that already exist. Always start with secondary, it is fast and cheap, and you often find the answer already exists.
• Exploratory, descriptive, causal. Exploratory research finds out which questions to even ask. Descriptive describes the state of things (who, what, how many). Causal, typically an experiment or A/B test, looks for cause and effect, meaning what actually happens when you change something.
• Stated vs. actual behaviour. What people say in a survey is not the same as what they do. That is why Ehrenberg-Bass and behavioural economics trust observed behaviour over attitude polls.

The doyen of applied research, Alan Hedges, summed it up back in the 1970s: more important than the data itself is understanding the broad, living context in which people decide. Research is not a polish for the deck. It is a tool for testing your hypothesis and not being confidently wrong.

Brand strategy in depth

The brand is the most patient and most valuable asset a firm owns, and also the one most often fobbed off with a logo. Brand strategy answers the question of what a person should recall and feel when they run into you. The Czech strategist Tomas Hrivnak likes to break it into three questions (drawn from semiotics): what the brand is, what it means and what it evokes. And the key variable is consistency: incoherent signals build a weak, contradictory web of associations in the head, whereas repeating the same thing builds instant recognition.

The tools and concepts we actually use when building a brand:

• Positioning (STP). Market segmentation, choice of the target segment and a clear place in the customer's head. Positioning is a sentence the customer should say about you, not one you say about yourself.
• Archetypes. The system of twelve archetypes (Mark and Pearson, The Hero and the Outlaw) helps give a brand a consistent personality and a story framework the customer can navigate.
• Brand pyramid. A model from the Millward Brown (now Kantar) researchers: the relationship with a brand is built in layers, from presence through relevance and performance up to bonding. A diagnostic tool for finding which layer a brand is "stuck" at.
• Distinctive assets and availability. The Ehrenberg-Bass Institute (Byron Sharp, Jenni Romaniuk) proved with data that brands grow mainly through penetration, mental availability (I recall you in the buying situation) and physical availability (I can buy you easily), not narrow loyalty. That means building recognisable distinctive assets: colour, logo, face, sound, word.
• Long and short (60/40). Les Binet and Peter Field, on IPA data, showed the most effective split is roughly 60% of budget into long-term brand building and 40% into short-term activation. Do performance only and you harvest demand nobody sowed.

The strategist's frameworks and tools (and when to use which)

A framework is not a strategy, it is a form that helps you think. The danger is filling it in and believing the work is done. Here are the most useful ones by phase, so you know when to reach for each:

How you know the strategy is working: measurement

Strategy without measurement is a religion. But measuring the wrong things is worse than measuring nothing, because it confidently drives you into a wall. Three principles that will save you a lot of money:

• Distinguish leading and lagging metrics. Revenue is a lagging metric: it shows the result once nothing can be done. Leading metrics (brand recall, pipeline quality, activation) give you a signal in time.
• Have one North Star. One main metric that best captures the value delivered, with a handful of supporting ones around it. Twenty equally important metrics means no priority.
• Beware attribution hubris and vanity metrics. Not everything can be cleanly attributed to a channel, and likes do not pay salaries. Les Binet warns that overvaluing easily measured short-term performance systematically undermines long-term brand growth. Measure what pays the bills, not what reports nicely.

How to write a strategy: the one-page strategy

When a strategy will not fit on one page, it usually means it is not finished yet. Here is the skeleton we use. It is not exhaustive, it is a forcing function for decisions:

• Diagnosis: one or two sentences on the real nature of the problem, not the symptoms.
• Guiding decision: one sentence on where we play and how we win (and where our advantage comes from).
• What we say no to: three things we deliberately sacrifice. Without this line it is not a strategy.
• Three moves: three coherent actions that reinforce each other and all serve the guiding decision.
• How we measure: one North Star and two or three leading metrics.

Mark Ritson has a harsh rule for it: diagnosis always precedes objectives. Whoever jumps straight to goals and tactics without an honest diagnosis is building a house from the roof down. And one more thing the Czech strategists Krutis and Cahlik both stress: a strategy written at a desk and handed over "finished" ends up in a drawer. The best strategy is built with the team that will then own it, because they understand why each thing is done.

The most common mistakes that kill strategies

• A goal passed off as a strategy. "We want to grow 30%" is not a plan for growing 30%.
• A strategy for everyone. When you do not want to offend anyone, you become invisible and interchangeable. A brand for everyone is a brand for no one.
• A carbon copy of the competition. Doing the same thing slightly better is Porter's operational effectiveness, not an advantage. Rivals catch up.
• Too many priorities. Ten priorities is zero priorities. Strategy is violent triage.
• Mixing types of spend. Tomas Hrivnak points out that fixed, performance and brand spend serve different purposes, and mixing them means chaos. Each needs its own logic and its own metrics.
• Changing course every month. A strategy that changes the moment it stops being fun is a mood. Give it time to work.

Who to read: today's best

If you are serious about the craft, this is your reading list. The staples: Richard Rumelt (Good Strategy Bad Strategy), Roger Martin and A. G. Lafley (Playing to Win), Michael Porter (competitive strategy). Brand data: Byron Sharp and Jenni Romaniuk at the Ehrenberg-Bass Institute, Les Binet and Peter Field under the IPA. Sharp pens and practice: Mark Ritson, Mark Pollard (Sweathead), Julian Cole (Strategy Finishing School), Rob Estreitinho (Salmon Theory), Rory Sutherland, Dave Trott, Adam Morgan (challenger brands), Martin Weigel, Tom Roach, JP Castlin. Creativity and attention: Orlando Wood and the System1 institute. For research and industry sources follow WARC, The B2B Institute and the Effie and Cannes Lions awards. And to not stay only in English: Czech strategic thinking is well served by Tomas Hrivnak, Michal Krutis and Pavel Cahlik, each from their own angle.

Special mention goes to a legend of Czech advertising, Josef Havelka, founder of Leo Burnett's Czech branch and author of columns in the magazines Strategie and Marketing & Media from 1996 to 2020 (later collected in a book). Havelka reminds us of two things strategists like to forget. First, that advertising is largely showbusiness, and without courage you get only a harmless average that nobody notices. Second, the so-called paradox of branding: to grab attention a brand must bring something unexpected, yet it also has to resonate with what the person already holds in their head. A good strategy does not hide that tension, it resolves it.

And let us not forget the classics who invented the craft. David Ogilvy (Ogilvy on Advertising) and Bill Bernbach (DDB's creative revolution) showed that great advertising rests on a big idea and respect for the customer. Claude Hopkins (Scientific Advertising) and Rosser Reeves (the USP) laid the foundations of measurable advertising and a clear offer. Al Ries and Jack Trout (Positioning) named how you win a place in the customer's head, and Philip Kotler organised the whole field into a textbook. For research and data beyond Ehrenberg-Bass, follow Kantar, Nielsen, Gartner, Forrester, McKinsey and the academic Marketing Science Institute.

Frequently asked questions

Sources: Richard Rumelt, Good Strategy Bad Strategy · Roger Martin & A. G. Lafley, Playing to Win · Michael Porter, Competitive Strategy and What Is Strategy? (HBR) · Byron Sharp & Jenni Romaniuk, Ehrenberg-Bass Institute, How Brands Grow · Les Binet & Peter Field, IPA, The Long and the Short of It · Mark Pollard, Strategy Is Your Words (Sweathead) · Julian Cole, Strategy Finishing School · Avinash Kaushik (See-Think-Do-Care) · Alexander Osterwalder (Business Model Canvas) · Mark & Pearson, The Hero and the Outlaw · Roger Martin (what would have to be true) · Eric Ries, The Lean Startup · Rita McGrath (discovery-driven planning) · Alan Hedges, Testing to Destruction · WARC, System1, The B2B Institute. Czech perspectives: Tomas Hrivnak, Michal Krutis, Pavel Cahlik.

This is what we do: marketing strategy and go-to-market strategy. It leads into media strategy. And for what an idea is (and where strategy ends and creative begins), see this article.

Got a goal but missing the one hard decision on how to reach it? Let us take 30 minutes. We will tell you straight whether you have a strategy or a nicely written wish list.