An operator asked us a fair question after we walked through his own homepage. It has a video near the top, a strip of five-star reviews, a row of association badges, and a quote form at the bottom that works. Why doesn’t any of it add up to anything?
The honest answer starts somewhere else. In a recent session we went through a set of real, live moving-company websites, the kind of businesses that clearly invest in their sites: current photography, working forms, real testimonials, nothing broken and nothing amateur. That is the part worth sitting with. None of these were neglected pages built by someone who stopped caring. They were well-built, well-regarded operators, and nearly every one made the identical mistake once its page was read start to finish instead of scanned in pieces. Good material, presented with no order connecting it, adds up to nothing in particular. The mistake was not confined to one company. It repeated across a well-resourced part of the industry.
The highlight-reel homepage
Most operator homepages get built one good element at a time. Someone commissions a video. Someone else pulls the strongest five-star reviews and pins them near the top. A designer adds the association logos the company earned over twenty years. A developer builds a form that submits correctly and routes to the right inbox. Every one of those decisions is defensible on its own. A different person usually makes each one, at a different time, with no brief connecting it to the others.
What lands on the page is a highlight reel: the best clips, stacked in whatever order they happened to get finished. It asks the visitor to do nothing specific next, and a highlight reel is not a story. It is evidence with no argument attached, and a visitor who has never worked with the company has no way to supply that argument themselves. They are left to infer what the company does, who it is for, and why they should care, from a stack of good but disconnected proof.
Compare that to the pages these operators are actually competing against for the first click: a national brand’s landing page, a marketplace’s listing, an aggregator’s funnel. Those pages are not necessarily better designed, and often carry fewer badges and thinner testimonials. But they are usually built by one team, working from one brief, in an order chosen on purpose. Order does work there that no individual element can do by itself.
What a story requires
Ask most operators why their homepage is built the way it is. The answer is some version of: we put our best stuff up top. That instinct is not wrong so much as aimed at the wrong question. The question was never what to put on the page, but what order convinces a stranger of anything at all. “Our best stuff first” answers neither.
Donald Miller’s StoryBrand framework describes the seven-part shape underneath almost every story that persuades anyone of anything: a character, the customer, not the company, has a problem, meets a guide who hands them a plan, calls them to action, and shows them what they stand to gain or lose. Each beat depends on the one before it. Put the plan before the problem, or the success before the plan, and the sequence stops making sense, no matter how strong any individual beat is on its own.
Apply that shape to a moving company’s homepage and the reordering is not subtle. The character is the person moving, worried about their possessions, their timeline, and getting lied to by a lowball quote. The problem is not “I need a mover.” It is closer to “I don’t know who to trust with everything I own.” The guide is the company, and a guide earns that role through empathy and competence, not by going first on the page. The plan is the three or four steps between requesting a quote and moving day. The call to action is the form. Success is the move going the way it was promised. Failure, named honestly, is the thing every visitor is afraid of: a truck that shows up late, a price that changes, a piece of furniture that doesn’t survive the trip.
A page built in that order does something a highlight reel cannot. It tells the visitor, in sequence, that the company understands their specific fear, has a specific plan for it, and is asking for one specific action. A page with the identical five elements in the wrong order never gets to make that case: video first because it finished first, badges next because someone was proud of them, form last because it has to go somewhere. Nothing on it is wrong. Nothing on it is in the right place either.
Why a video, some reviews, and a form still add up to nothing
A second reason a highlight reel fails has nothing to do with narrative theory and everything to do with how people actually look at a web page. Nielsen Norman Group’s original eyetracking research found that visitors scan pages in a rough F shape: a horizontal pass near the top, a shorter horizontal pass further down, then a vertical scan along the left edge, picking up whatever bolded words or headings sit on that path. A reaffirmation of the same research eleven years later found the pattern still holds, with one honest caveat: a page with strong visual hierarchy can pull a visitor into other scanning patterns instead. Either way, almost nobody reads a homepage top to bottom the way its builders assembled it.
That changes what “our best stuff up top” actually does. A visitor’s F-shaped scan hits the video, catches a fragment of a review, notices a badge shape, and lands on the form. None of those four fragments tells the visitor anything about what happens next or why it matters. The video answers a question nobody in scan mode asked. The badges answer a question a competitor might ask, not a stranger deciding whether to trust a company with everything they own. What a scanning visitor needs at each stop is a fragment that advances an argument, not a fragment that showcases an asset. A page with real sequence gives them that at every stop. A highlight reel does not, which is the mechanical reason the video, the reviews, and the form can each be good and still add up to nothing together.
A related, more technical version of this same problem lives underneath the page, in the machinery rather than the story: your website isn’t failing, it was never infrastructure.
The mystery box, correctly understood
A tempting shortcut is worth naming here, before it can take hold. A homepage built in sequence sounds close to a homepage built around suspense: withhold the good stuff, tease it, make the visitor scroll to get it. That is not what the filmmaker most associated with mystery in modern storytelling actually argues, and getting this wrong produces the manipulative homepage nobody should want to build.
J.J. Abrams’ 2007 talk on what he calls the mystery box is not a formula for withholding and revealing. He opens by showing an actual sealed box he bought decades earlier from a magic shop, a “$50 of magic for $15” novelty package. The point of the talk is that he has never opened it and never will. In his own words, the box “represents infinite possibility. It represents hope. It represents potential,” and mystery itself, not the payoff behind it, is “the catalyst for imagination.” His own box stays sealed forever. That is the opposite of a setup waiting for its reveal.
Applied to a homepage, the distinction matters. A page built to manipulate withholds information the visitor needs, the price range, what happens after the form is submitted, purely to force another scroll. That is not mystery so much as friction wearing a marketing label, and it usually loses the visitor at the point of withholding. What Abrams is describing is different: a coherent world where curiosity is generative because a real, complete story underlies it, even when not every detail sits on the first screen. The visitor is not manipulated into wondering what comes next. They are given enough of a coherent story, character, problem, guide, that wondering what happens after they submit the form becomes a reasonable, motivated question, not a trick.
That is the difference between narrative order and playing games with visitors. One earns curiosity by being coherent; the other manufactures it by staying incomplete on purpose. A stack of unconnected elements does neither. That is the actual failure mode this piece is about. It is not coherent enough to earn curiosity, and it is not withholding anything deliberately. It just does not add up.
The thread worth leaving open
If mystery-as-manipulation is the wrong lesson to take from Abrams, a real, narrower psychological case still supports leaving exactly one thread open on a homepage rather than resolving everything the moment it appears. It comes from an older, more specific piece of research, its caveat stated plainly rather than buried.
In 1927, psychologist Bluma Zeigarnik ran an experiment in which participants worked through a series of small tasks and were interrupted partway through roughly half of them. Her original paper found that participants recalled the interrupted tasks dramatically better than the ones they had completed without interruption, usually summarized as roughly a 90% memory advantage for the incomplete over the complete. That figure comes from one study run nearly a century ago, and it has not held up cleanly since. A 2025 meta-analysis pooling 59 studies found almost no memory advantage for unfinished tasks specifically. It did find a separate, more reliable effect: people tend to want to go back and finish something they started, at a rate well above chance. The narrow memory claim is weaker evidence than its reputation suggests. The pull to resume something left incomplete is the part that has actually held up.
That distinction, not the 90% figure, is the useful one for a homepage. A page that resolves everything on first contact gives the visitor no reason to keep going, because nothing is left unresolved to pull them forward. The video already explains everything. The reviews already summarize the whole experience. The form asks for a full history before the visitor has committed to anything. A page that opens one real thread instead works differently: it states the problem clearly, names the guide, and holds the plan and the fuller proof for the next section. That gives a visitor an actual reason to keep scrolling. Not suspense for its own sake. An unfinished thought they were already having about their own move, met partway.
What Amazon and Airbnb build before they build anything
Two companies outside the moving industry treat narrative sequence as infrastructure, not decoration, and neither one is guessing.
Amazon’s own practice is to write the press release and the FAQ for a product before writing a line of code, a discipline described by CTO Werner Vogels in 2006 and unchanged in substance since. In his own words: “we start by writing the documents we’ll need at launch, the press release and the faq, and then work towards documents that are closer to the implementation.” The press release has to state, in plain language, what the thing does and why anyone would want it, in story order, before the company is allowed to build it. If the story does not work on paper, the product does not get built. Sequence comes first, by design.
Airbnb ran a version of the same discipline in the other direction, on a product that already existed. In 2011, a cofounder read a biography of Walt Disney and noticed that Disney’s animators built full storyboards before animating a single scene of Snow White. Airbnb hired a former Pixar animator and built its own storyboard: fifteen frames tracing a guest’s actual journey from first hearing about Airbnb through leaving feedback after a stay, and fifteen more for the host’s side. Laying the whole journey out in sequence exposed a real gap: the company had built almost everything around the online booking moment and almost nothing around what happened once a guest actually arrived somewhere unfamiliar. That gap led directly to a real product: neighborhood guides that helped a guest feel oriented once they landed. The storyboard did not decorate a homepage. It found what was missing from one.
Neither company treated narrative order as decoration layered onto a finished plan. Both treated it as the plan itself. Most operator homepages get their sequence added last, if at all, after the video has been shot, the badges designed, and someone remembers a form still needs to exist.
Neither Amazon nor Airbnb treats that discipline as a one-time project, and most operator sites do the opposite the moment they go live: your website was finished the day it launched.
A homepage in the right order
Put the case together and a rewritten homepage looks different in kind. The same five elements from the top of this piece, video, reviews, badges, form, run through the sequence a real story requires instead of the order they happened to get finished.
Open with the character and the problem, stated in one sentence a visitor can read in the first F-shaped scan: who this is for, and what they are actually afraid of, not what the company does. Follow immediately with the guide’s claim to competence: one specific, credible reason to believe this company understands that exact fear. This is where a testimonial belongs, as evidence for that claim, not as a badge of general approval. Then the plan: three or four steps, visible before the visitor is asked for anything, so the form that follows is not a leap into the unknown but the next expected step. If it earns a place at all, the video belongs here, showing the plan in action rather than sitting up top as generic brand content. The badges that survive the cut belong last and small: reassurance for a visitor who has already decided, not the opening argument.
That is not a radical rewrite. It uses every element the original page had. It just puts a character before a problem, a problem before a guide, a guide before a plan, and a plan before a form, instead of stacking finished assets in the order they happened to arrive.
It is also not a moving-industry-specific idea. Apple’s page for its current flagship phone runs the identical shape end to end: a hook, several sections of escalating, specific proof, materials, camera, performance, battery, each claim sharper than the last, a section that states outright whether an upgrade is worth it, and a closing set of plainly answered objections. Nowhere on that page does a badge, an award logo, or an unearned superlative appear before the specific claim that justifies it. A company selling more phones than almost anyone on earth still puts its proof in an order that builds a case, not a stack that displays assets. An operator with forty years of real reputation behind them has more raw material for that same case than most companies ever will. Almost none of them are arranging it that way.
What a visitor leaves holding
None of this requires new material. Every operator whose homepage reads like a highlight reel has a video, real reviews, earned badges, and a working form. The problem was never the assets. A stack of good evidence with no story connecting it asks nothing of the visitor and leads nowhere in particular. A visitor given nothing to follow leaves with no clear idea of what the company does, who it is for, or why they should come back.
That is the real cost, and it is easy to miss because nothing about it shows up as an error. The page loads. The video plays. The form submits when someone fills it out. A homepage that fails this way looks, from the inside, like one that is working, right up until someone asks why the enquiries never quite match the traffic.
If a visitor left after this page’s first ten seconds, what single idea would they be left holding? For most of the operator sites reviewed for this piece, the honest answer was: none in particular, just an impression of competence with nothing specific attached. That is worth sitting with: the fix is not a bigger video or a better badge. It is an order.
Back to where this started: a video, some reviews, a form, and why none of it adds up to anything. It adds up to nothing because nothing on the page asks anything of the visitor in sequence. A character, a problem, a guide, a plan, a call to action, in that order, turns the identical elements into a case instead of a highlight reel. The assets were never the problem. The order was.
