Usually because it explains your process instead of the visitor's experience. A good How It Works section answers what the reader has to do and what they get back. Most versions describe internal workflow steps that matter to the company and mean nothing to the person reading.
This section shows up on almost every service and software homepage, and it is one of the most reliably wasted blocks on the page. It takes real estate in a prime position and often says less than the headline above it.
When we get it right, it does a specific job. It removes the fear of the unknown that stops people from enquiring, and it does that in about fifteen seconds of attention.
It exists to reduce perceived risk. Someone interested in your offer is quietly asking what happens after they click, how much effort this will take, and how long before they see something. The section answers those three questions so the reader can commit without guessing.
That framing rules a lot of content out. Your internal quality checks, your discovery methodology, and your team's review process are not what the reader is anxious about. They are anxious about whether this will be complicated and whether they will look foolish for starting.
The best test we know is to read each step and ask who does the work in that sentence. If every step has your company as the subject, you have written an about section wearing a How It Works costume.
Three, usually. Four if the process genuinely has a distinct fourth stage. Beyond four, the section stops reducing anxiety and starts creating it, because a long list of steps reads as a long commitment. Fewer steps signal a simpler experience even when the underlying work is identical.
This is a presentation decision, not a lie about your process. A build that involves twenty tasks can honestly be described as scoping, building, and launching. You are choosing the altitude at which to describe the work, and the reader wants the high altitude version.
Length aversion is well documented in adjacent research. Baymard Institute found that 17% of shoppers who abandon a checkout cite a process that is too long or too complicated, and its work notes that the average site presents 23.48 form elements when something closer to 12 to 14 is workable. People react to apparent effort, not actual effort.
Below the hero and the primary value proposition, but above pricing and testimonials. By the time someone reaches it, they should already know what you do. The section answers the second question, not the first, and moving it higher tends to confuse rather than clarify.
The exception is a genuinely unfamiliar offer. If your product category is new enough that people cannot picture the experience, the mechanism becomes part of the value proposition and belongs much higher. That is a rare case and most companies assume it applies to them when it does not.
Placement interacts with how people scan. Nielsen Norman Group's eyetracking research on the F-shaped pattern, first published in 2006 and revisited in 2017, found that first lines of text receive more gazes than the lines below them, and that the first few words on the left of each line get more fixations than the words after them. Whatever sits at the start of each step carries most of the weight.
Screenshots when the product is visual, icons when it is not, and illustrations only when they show something. A generic icon of a rocket next to the word Launch adds decoration and zero information. If the image could sit next to any of the three steps without seeming wrong, it is not working.
Screenshots are underused because they date faster and require maintenance. That cost is usually worth paying. A real interface answers questions a stylized illustration cannot, and it signals that the product exists, which matters more for newer companies than most founders expect.
Whatever you choose, keep the visual treatment identical across the steps. Mixed metaphors, where step one gets a photo and step two gets an icon, break the sense that these are parts of one sequence. Our guide to visual hierarchy in web design covers why that consistency does so much work for comprehension.
A short label and one sentence. Two at the absolute most. People do not read this section, they scan it, so the label carries the meaning and the sentence adds the reassurance. Anything longer will be skipped entirely and you lose the label too.
The research here is blunt. Nielsen Norman Group analyzed 45,237 page views and concluded that on an average web page, users have time to read at most 28% of the words during an average visit, with 20% being more likely. Jakob Nielsen published that finding in 2008 and nothing since has made people read more.
So write for the 20%. The label should make sense alone, because for most readers it will be alone. If your step only works when someone reads both the label and the sentence beneath it, rewrite the label until it stands by itself.
Add a video, do not replace the section with one. A short walkthrough serves the minority who want depth, while the three steps serve everyone else. Replacing text with video hides your explanation behind a play button that most visitors will never press.
Video also costs you in ways that are easy to miss. An embedded player from YouTube pulls in third party scripts that slow the page, and an autoplaying background video competes with the text it is supposed to support. If you embed, load it on click rather than on page load.
Where video genuinely wins is complex software. If the value only becomes obvious in motion, a fifteen second silent loop of the actual interface will outperform any amount of copy. Lottie and Rive are both worth considering for that, since they produce far lighter files than video for interface style animation.
Use a verb the reader performs, then the outcome they get. Book a call, get a plan, approve and launch. Avoid abstract nouns like Discovery, Alignment, and Onboarding, which sound professional and communicate almost nothing about what will happen.
Abstract labels are the single most common failure we see. Discovery is a word that means something specific inside an agency and nothing at all to a founder comparing three suppliers. The reader has to translate it, and most will not bother.
Timing helps more than adjectives. Adding a concrete duration to a step turns a vague promise into something a person can plan around, which is why we tell clients to state real timelines rather than hide them. Our guide to microcopy in web design covers how these small pieces of text carry disproportionate weight.
Describing internal process, using abstract labels, adding too many steps, and decorating rather than illustrating. A fifth is placing it correctly but writing it last, so it repeats the headline instead of advancing the argument the page is making.
Repetition is the sneaky one. Because this section usually gets written after the hero, writers often restate the value proposition in three parts rather than explaining the mechanism. The page then says the same thing twice and the reader learns nothing new from scrolling.
The last mistake is forgetting the end. A How It Works section should finish with what the reader ends up holding, not with your final internal task. Ending on Launch tells them the project stops. Ending on a live site that loads in under a second tells them what they get.
Watch scroll depth and what people do after they reach it. If visitors consistently scroll past the section and keep going to pricing, it is doing its job. If they stall there or bounce, the section is raising questions instead of settling them.
Heatmap and session tools like Hotjar or Microsoft Clarity make this visible quickly, and scroll depth in Google Analytics gets you most of the way for free. Look for the drop off point rather than the average, because averages hide the exact place people give up.
The cheaper test is to show the section to someone unfamiliar with your business for ten seconds, then ask them what they would have to do to get started. If they cannot answer, no analytics tool is going to tell you anything more useful than that.
Getting this block right is a small change with a real payoff, and it is one of the levers behind the 3.5x average conversion lift we see across our projects. If you want a second opinion on whether yours is explaining or just decorating, take a look at our thinking on homepage design first, then reach out through phoenix.studio and let's talk it through.
Tell us where you want to go. We'll tell you how we'd get you there.