10 Nov Ideation: Goal Statements, How Might We, Rapid Sketching, and Competitive Audits
Part 8 of Start the UX Design Process — a series on the UX principles I use in client work.
There is a moment in every website project when somebody falls in love with the first idea. It usually happens fast — somebody sketches a homepage, everybody nods, and from then on every meeting is about defending that sketch instead of improving it. The project did not choose the best idea. It chose the first idea that felt good enough to stop thinking.
Ideation exists to prevent that. It is the deliberate practice of generating lots of ideas before judging any of them. Quantity first, judgment later. That sounds wasteful until you realize how cheap ideas are compared to launched websites. A sketch costs minutes. A launched page built on the wrong idea costs weeks and a second redesign.
One more rule before the methods: ideation is not guessing. It is grounded in everything from the earlier posts — the empathy maps, personas, stories, journey maps, and the problem statement from post seven. Ideas that ignore that research are just the designer’s own assumptions in sketch form, and assumptions are what the whole process exists to replace. You may need more than one ideation session, and you will not use most of what you generate. Both are normal. The research narrows the field; the ideation fills it.
Write the goal before you sketch
Between the problem statement and the sketching sits one more short sentence, and it is worth writing before you generate anything: a goal statement. Where the problem statement names the problem, the goal statement names the ideal solution in one or two sentences — the product, the specific action it lets people take, who it affects, how it helps them, and how you will know it worked.
For a festival site, it might read: “Our festival site will let day-of attendees find today’s schedule in one tap, which will help visitors arriving by phone get to the right stage without asking on social media. We will measure it by whether those ‘where is this happening?’ messages drop.” Notice what that sentence does: it pulls the who from the persona, the what from the user story, the why from the journey map, and it adds a finish line. Sketching against a goal statement is a different exercise from sketching in the open air — every idea can be checked against it, and the next post’s choosing gets much easier because the success test is already written.
Turn the problem into a question: How Might We
A problem statement can feel heavy to sketch against. A useful trick is to turn it into a question that starts “How might we…?” — a question that treats the problem as an opportunity instead of a verdict. “Day-of attendees cannot find today’s schedule on their phones” becomes “How might we help a day-of attendee find today’s schedule without opening a menu?” That one question can produce a dozen sketches. A different question — “How might we make today’s schedule work on a weak phone signal?” — produces a dozen more, different ones.
Good questions are broad enough for many answers but narrow enough to stay focused. “How might we make the website better?” is too broad to sketch against; it could mean anything. “How might we help a first-time guest picture Sunday morning before they visit?” is narrow enough to guide a session and open enough for many answers. If a question produces nothing useful, rewrite it — questions are drafts, not commitments.
Reframing helps when you are stuck. Flip a positive in the problem into a solution: make finding the schedule feel like a game for the family in the car. Solve the opposite: how might we make a schedule impossible to find, and what does that reveal about the real blockers? Change the process entirely: what if there were no menu at all? Remove the painful step: what if a visitor never had to register to see the schedule? Compare it to another experience: how might we make finding a stage feel like finding a gate at an airport? Or bring in a resource the problem never mentioned: what could a text message do that a webpage cannot? Each reframe is a new question, and each new question is a new batch of ideas.
The rule that makes ideation work
One rule, and it is harder than it sounds: no judging during the generating phase. No “that won’t work,” no “we tried that,” no “the client won’t like it.” Judgment has a job — it comes later, in the next post — but if it shows up early, it kills the weird ideas, and the weird ideas are often where the good ones hide.
I have seen this play out on real projects. The first idea for a festival schedule page is always the obvious one: a list, in order, with times. Fine. The third or fourth idea is where it gets interesting: a “happening now” view for day-of visitors, a simple filter by stage, a schedule that loads as plain text first so it works on a weak signal. Those ideas only show up because the first two were allowed to be ordinary. If you marry the first idea, you never meet the fourth.
The same is true for a church site’s guest path, or a nonprofit’s donate flow. The first donate idea is “make the button bigger.” The fourth idea is “tell the donor what their gift does before asking for payment, then make the form shorter.” The first idea decorates the problem. The fourth idea solves it — but only if you gave yourself permission to generate four.
Crazy Eights: eight ideas in eight minutes
My favorite method for forcing quantity is called Crazy Eights, and it is exactly what it sounds like. Take a sheet of paper, fold it into eight boxes, set a timer for eight minutes, and sketch one idea per box — one minute each. No polishing, no perfect drawings, just enough of a sketch that somebody else could understand the idea. A box with a rectangle and an arrow and three words counts. A box left blank because you were perfecting box one does not.
The time pressure is the point. One minute per idea leaves no room for self-censorship. You cannot talk yourself out of an idea in sixty seconds, so the ideas come out raw — including the ones you would have dismissed as silly in a meeting. Some of them are silly. Some of them, on second look, are the seed of the solution.
Crazy Eights works alone, and it works better with a small group: the client, a staff member, anybody who knows the visitors. Each person fills a sheet, then everybody talks through their eight. The conversation that follows is completely different from a meeting that started with one homepage mockup, because now there are twenty or thirty ideas on the table instead of one idea to defend.
A few practical notes from using it on real projects. Sketch the page, not the art — boxes for content, lines for text, arrows for flow. Label each idea in a few words so you can remember it tomorrow. Warm up first with a silly minute of drawing — it loosens the hand and the judgment at the same time. Work on one problem statement or one How Might We question at a time, not three at once. Stick figures are fine; clarity beats artistry. And when the eight minutes are up, have each person share their top two or three sketches, or let the group vote on favorites, then develop the winners in more detail. That sharing step is where a solo exercise becomes a team decision.
And do it against the problem statement from the last post, not in the open air: “eight ways a day-of attendee could find today’s schedule on a phone” produces useful ideas, while “eight homepage ideas” produces decoration.
Other fast methods that do the same job
Crazy Eights is not the only way to force quantity. A timed list works too: ten ways to answer the visitor’s top question, written in ten minutes, no editing allowed. So does “worst idea” — deliberately generating the worst possible solutions first, which loosens everybody up and often reveals a good idea hiding inside a bad one turned inside out. So does sketching the same page three ways for three different personas. The method matters less than the discipline: generate first, judge later, and generate more than feels necessary.
Look sideways first: a competitive audit
Before you sketch, spend a little time looking at what already exists. A competitive audit is a structured look at other sites in the same space — direct competitors, who serve the same audience with something similar, and indirect competitors, who serve the same audience with something different, or a similar audience with something different. For a festival site, direct competitors are other festival and event sites; indirect ones might be ticketing or community-calendar sites. Both teach you something: direct competitors show you the standard visitors expect, indirect ones show you ideas your own field has not tried yet.
Set the audit’s goal first — for a redesign, usually “compare how visitors experience each site” — then list the competitors (ask the client who they worry about, then add what your own research finds) and compare the same aspects across all of them: first impressions on phone and desktop, how easy the navigation and user flow are, whether the visual design stays consistent, whether the content is clear and findable, and who the site seems built for. Rate each aspect on one consistent scale, in your own words, from “needs work” to “outstanding,” so the comparison is honest instead of impressionistic. Then write it up as a short report: what each competitor does well, where they all fall short — that gap is your opening — and what it means for your design. Present the highlights, not the whole spreadsheet, back your conclusions with specific examples from the sites you reviewed, and watch your own biases: an audit that only confirms what you already believed is not an audit. Done this way, an audit does not tell you what to copy. It tells you what visitors already expect, what nobody is doing well, and where your site can be genuinely different — which is exactly the information ideation needs.
What ideation is not
Ideation is not a brainstorming meeting where the loudest person wins. If one voice dominates, you do not have ideation — you have a presentation. Silent sketching first, discussion second, is the fix: everybody generates alone, then everybody shares. The quiet person’s fourth sketch deserves the same hearing as the confident person’s first.
Ideation is also not a promise to build everything. Most ideas will die, and that is success, not waste. An idea that dies on paper cost you a minute. An idea that dies after launch cost you a redesign. The goal of ideation is not a pile of features. It is a short list of strong candidates, earned by surviving a fair fight — and choosing among them is the next post’s job.
So before your next redesign locks in a layout, try eight minutes and eight boxes against your one-sentence problem. You might keep the first idea in the end. But you will keep it because it won, not because it was first.
Want a website process that explores options before locking one in? I build and redesign websites for small businesses, non-profits, and churches across Eastern NC. Get a free consultation or see recent work.
Previous in this series: Define the Problem · Next: Choosing What to Build First: Judging Ideas by Impact and Feasibility
This article is my own take, in my own words, on UX principles I use in client work. It is not affiliated with or endorsed by Google or Coursera.