One idea, proved
in every channel.

The brief is the problem and the concept is the answer. A brief states the business objective, the audience, the proposition, the reasons to believe, the constraints, and what success looks like, and it is written before creative work begins, usually by a strategist or the client. The concept is what the creative team brings back, and it should be judged against the brief rather than on its own merits, which is precisely why the two should be visible at the same time. The common failure is a concept that has quietly redefined the brief, answering a more interesting problem than the one that was set. That is sometimes the right outcome and it needs to be named out loud rather than discovered three months later, and it is much easier to name when both documents are on the same canvas.

Three is the convention and two genuinely distinct routes are usually better than three where one is padding. The reason three persists is that one route feels like a take it or leave it, and two can feel like a binary the client did not agree to. But the real requirement is that every route presented is one you would be happy to make, because a route included to make another look good will occasionally be chosen and you will have to build it. Presenting them side by side rather than sequentially matters more than the number: a client comparing route three against their memory of route one is not really comparing, which is how the last route shown tends to win. Laid out on one canvas, all three are in view at once and the decision is about the work.

Four tests, in roughly this order. Does it translate, meaning does it work in at least three channels while staying the same idea. Is the insight true, meaning would the audience recognise it as something they know about themselves rather than something a brand decided about them. Is it specific to this brand, and the sharpest version of this test is whether a competitor could run the same campaign with their logo on it, because if they could you have made a category ad. And can you picture the finished thing, which is where most concepts fail in the room, not because the thinking is weak but because nobody showed enough craft reference for it to feel real. A board that carries all four is a board that survives the meeting.

Art direction is the practice of deciding how something looks and making every visual element serve one idea. It covers the choices that add up to a coherent feel: photography style, colour, typography, casting, location, wardrobe, composition, and the treatment applied across every asset. The distinction people usually want is between art direction and design. Design executes; art direction decides what is being executed and holds the line across everything, so a campaign shot by three photographers in two countries still looks like one thing. It is also different from a creative concept, which is the idea, where art direction is how that idea is made visible. In practice an art director works from reference rather than description, which is why the material is a board of images rather than a written brief: nobody has ever successfully specified a tone in adjectives.

Creative strategy is the reasoning that connects a business problem to the creative work meant to solve it. It sits between the brief and the concept, and it answers a specific set of questions: who exactly are we talking to, what do they currently believe, what do we need them to believe instead, what is the one thing we can say that would shift that, and why should they believe us. The output is not an idea, it is the proposition an idea has to express, which is why strategists and creatives are usually different people doing sequential jobs. The most common failure is a strategy stated so broadly that any idea satisfies it, which means it is doing no filtering. A useful test: if the proposition would work equally well for your competitor, it is a category observation rather than a strategy.

Not during early access. Storyflow is paid-only right now, with plans from $7.99 a month billed annually, and a Free plan launches before the end of 2026. Anyone a paid member invites to a board joins free today, so a client or a collaborator can view and comment without paying. If you need something free this minute, Figma's free tier handles concept boards well and Milanote's free tier is genuinely pleasant for collecting reference. What you give up in both is the AI that produces territories from a brief and reads the whole board as context, which is the part that helps most when you are concepting alone.

It can produce territories worth reacting against and it should not produce the idea you present. What it is genuinely good at is range and speed: ask for six territories from a brief and you get six, quickly, and the fact that four are obvious is useful information about what the obvious answers in this category are, which is exactly what you need to avoid. It is also good at stress-testing, because asking how a territory would express in a channel you have not tried will expose a thin idea fast. What it cannot do is have the insight, because insights come from noticing something true about people, and a model trained on what has already been said is structurally biased towards the middle. Storyflow's AI proposes and arranges; the idea stays yours, which is also the only way the work ends up sounding like you.

Per territory: the insight in one line, the idea in one line, three or four channel expressions with real headlines rather than placeholder text, and enough visual reference that the tone is felt rather than described. Above all three territories, the brief itself, so the conversation stays anchored. What to leave off is the process, meaning the routes you abandoned and the reasoning behind the shortlist, which is interesting to you and noise to a client. Keep that material on the same canvas but in a separate region you can scroll to if asked. The advantage of a canvas here is that you do not have to choose in advance between a tight presentation and having the depth available, which is the choice a linear deck forces on you.

Move between the two rather than proceeding from one to the other, because the loop is the method. Write the insight as plainly as you can, then try to make one execution from it, and pay attention to what happens. If the execution will not come, the insight is often not actually true, or it is true but not uncomfortable enough to produce tension, and going back to sharpen it is faster than pushing harder on the execution. If a strong execution arrives, work backwards and ask what insight it is really expressing, because it is frequently not the one you started with and the real one is better. On a canvas both directions are the same gesture: put the insight and the executions next to each other and draw the connection, and the loop stays visible instead of being lost between drafts.

Yes, and it is a better presentation than a deck for this particular job. Walk the canvas, showing the brief first, then the three territories side by side so the client can compare rather than remember, and zoom into whichever route they lean towards to show the depth underneath. Send a view-only link afterwards and they can revisit the whole board in a browser with no account, which is what usually happens anyway when they need to show a colleague who was not in the meeting. Comments attach to the territory they concern, so the feedback arrives with its referent intact instead of as an email listing six points about four routes.

The board becomes the production plan rather than being archived. On the same canvas, the approved territory carries into the things that have to be made: shot lists, the shooting schedule, the channel plan, the asset list per format, and the timing. The reason to keep them together is that the questions that come up in production are almost always questions about intent, meaning whether a particular execution is still expressing the concept, and answering those from a board where the insight and the original expressions are one scroll away is far quicker than reconstructing the argument from an approved PDF. The routes you did not take also stay on the board, which settles the version of that conversation that reliably resurfaces in week six.