Constraint
No analytics survived the rebuild, so no number is claimed
Client
W Communications, a strategic communications and speaking practice
Scope
8 pages across 4 audiences, brand system
Stack
WordPress
A television producer needs a communications expert on air in 90 minutes. They have one name, one browser tab and no patience. They want the credential, the topics, a photograph they are cleared to use, and a way to reach a person who answers. The site greets them with a paragraph about the practice. The credential is further down somewhere. The topics are inside a document. The photograph is 400 pixels wide. They do not email for a media kit, because they cannot wait 90 minutes for one. They book the second name on the list, whose page they could read in 40 seconds.
One page cannot serve four audiences. Four pages can, if each one knows who just arrived.
What was actually happening?
Journalists, clients, event bookers and readers all landed on the same page and were handed the same paragraph. The proof was there, below the fold. Nothing was structured so that a producer, or a machine answering a question about the practice, could pull a fact out and use it without first asking a person.
1. What should a consultancy homepage do when four audiences arrive at once?
Before: one homepage that opens with a general paragraph about the practice and names no audience.
After: a homepage that names each audience and sends it to the page built for it.
UX and UI analysis, before against after
Before: what the design cost the visitor
A general paragraph, nobody named. Four distinct audiences reached one homepage that spoke to none of them. A reader not named in the first screen cannot tell whether the page is for them, which the headline research lists as the first reason people leave. Source: Nielsen Norman Group, 2015.
Prose where a scanner wanted structure. Opening with a paragraph asks the visitor to read, and the research consistently finds most people scan instead. Source: Nielsen Norman Group, 1997.
Scent that ended at the door. A producer, a client, an event organiser and a reader each arrived for a different reason, and the page gave none of them a path. Source: Xerox PARC, Pirolli and Card, Psychological Review, 1999.
After: what changed and why it works
Each audience named and sent on its way. The homepage says who it serves and routes each group to its own page. Naming the reader is the strongest scent a page can give. Source: Nielsen Norman Group, 2003.
Four short routes, not one long paragraph. Dividing the first screen into audience blocks lets a scanner find theirs in seconds. Source: Nielsen Norman Group, 2016.
One decision at the top. The visitor picks which audience they are, and everything after that is relevant to them: the choice reduction the research links to action. Source: Columbia University and Stanford University, Iyengar and Lepper, Journal of Personality and Social Psychology, 2000.
2. Where does the credential belong on a personal practice site?
Before: a biography page with the credential sitting below several paragraphs of narrative.
After: a biography page that opens with the credential and tells the story beneath it.
UX and UI analysis, before against after
Before: what the design cost the visitor
A buried credential. The former Starbucks senior vice president credential sat under several paragraphs of story. Three quarters of viewing time falls within the first two screens, so the strongest proof on the page sat outside most readers' attention. Source: Nielsen Norman Group, 2018.
Story before evidence. Research on About pages finds visitors arrive with a specific question, usually who is this and why trust them, and want it answered quickly. Source: Nielsen Norman Group, 2008.
Credibility delayed is credibility lost. Demonstrated expertise is one of the four trust factors, and it only counts if it is seen. Source: Nielsen Norman Group, 2016.
After: what changed and why it works
Credential first. The page opens with the fact that establishes authority and tells the story under it, matching how people actually use an About page. Source: Nielsen Norman Group, 2008.
Proof inside the attention zone. Placing the credential in the first screen puts it where more than half of viewing time is spent. Source: Nielsen Norman Group, 2018.
A story worth reading. Once the credential has earned attention, the narrative beneath it reads as context rather than as an obstacle. Source: Nielsen Norman Group, 2016.
3. How should expertise be structured so it can be quoted?
Before: expertise described in prose, with the topics implied rather than listed.
After: an expertise page with one topic per section, each quotable on its own.
UX and UI analysis, before against after
Before: what the design cost the visitor
Topics implied, never named. Expertise written as flowing prose left the actual subjects unstated, so a reader, a producer or a search engine looking for a specific topic could not find it. Source: Nielsen Norman Group, 2017.
Nothing to lift. Answer engines and journalists quote discrete passages, and prose with no headings offers no passage that stands alone. Source: Google Search Central, 2024.
Nothing to stop on. With no headings, the F-shaped scan passed over the page and found no anchor. Source: Nielsen Norman Group, 2017.
After: what changed and why it works
One topic, one section. Each area of expertise has its own heading and paragraph: chunking applied to a subject list, which is how a scanner finds the one topic they came for. Source: Nielsen Norman Group, 2016.
Each section stands alone. A heading plus a self-contained paragraph is exactly the unit a search engine, an answer engine or a producer can quote. Source: Google Search Central, 2024.
Headings that promise something. Every topic heading tells the reader what they get by continuing, which is the test the headline research sets. Source: Nielsen Norman Group, 2015.
4. What does a client page have to show a buyer?
Before: a list of client names with no context and no description of the engagements.
After: a client page that says what the work was beside each name.
UX and UI analysis, before against after
Before: what the design cost the visitor
Names without context. A list of clients with no description of the work leaves the buyer guessing whether it was a keynote, a campaign or a crisis. Evidence of expertise only builds trust when the reader can see what it was applied to. Source: Nielsen Norman Group, 2016.
Logos as ornament. The eye treats a logo wall as decoration, and eyetracking shows decorative imagery is skipped. Source: Nielsen Norman Group, 2010.
The buyer's question left open. A buyer asks what you did for them; a page that only says who was served does not answer. Source: Stanford Persuasive Technology Lab, Fogg et al., 2002.
After: what changed and why it works
The work stated beside each name. Every client entry names the engagement, so the name becomes evidence instead of ornament. Source: Nielsen Norman Group, 2016.
Informative, not decorative. With a line of context next to it, each logo turns into information the eye pauses on, which is the distinction the photo research draws between studied and ignored imagery. Source: Nielsen Norman Group, 2010.
Claims a buyer can check. Naming the work lets the buyer verify it, and verifiable specifics are the form of proof the Stanford guidelines rank highest. Source: Stanford Persuasive Technology Lab, Fogg et al., 2002.
5. What does a producer need from a media page?
Before: a media page that required an email before anything usable was available.
After: a media page a producer can use without sending an email first.
UX and UI analysis, before against after
Before: what the design cost the visitor
An email before anything usable. A producer booking a guest on deadline needs a bio, a headshot and a topic list immediately. Gating them behind an email puts a form before its purpose, which is exactly what the form research shows people abandon. Source: Nielsen Norman Group, 2018.
The most hurried visitor was the most blocked. The person under the most time pressure had to wait for a reply, the foraging cost the research says sends people to the next source. Source: Xerox PARC, Pirolli and Card, Psychological Review, 1999.
No status, no timeline. Nothing told the producer whether or when anything would arrive: the visibility failure the heuristics list first. Source: Nielsen Norman Group, 2024.
After: what changed and why it works
Usable without an email. Bio, photographs and topics sit on the page and download directly, so a producer can use them in the minutes they have. Source: Nielsen Norman Group, 2018.
Laid out in scan order. The assets a producer needs are chunked and labelled, so they can be found and taken without reading. Source: Nielsen Norman Group, 2016.
Contact kept, as a second step. The email route remains for anything the page does not cover, placed after the value rather than in front of it. Source: Nielsen Norman Group, 2024.
6. What should a speaking page state before the enquiry?
Before: a speaking page with no formats, topics or audience.
After: a speaking page that lists formats and topics, then offers one enquiry action.
UX and UI analysis, before against after
Before: what the design cost the visitor
No formats, topics or audience. An organiser deciding whether to book a speaker needs to know which talks exist, how long they run and who they suit. Without that, the enquiry form asked for interest the page had not earned. Source: Nielsen Norman Group, 2018.
No way to scan for fit. With nothing listed, a scanning reader could not match the speaker to their event. Source: Nielsen Norman Group, 1997.
Trust asked for without evidence. A speaking page that shows no past formats and topics offers no demonstrated expertise, which the credibility research treats as essential. Source: Nielsen Norman Group, 2016.
After: what changed and why it works
Formats and topics listed. The page shows what is on offer, so an organiser can judge the fit before contacting anyone: disclosure before the ask. Source: Nielsen Norman Group, 2016.
One enquiry action. After the facts, the page asks once. A single action at the end of a clear page is the pattern the choice research supports. Source: Columbia University and Stanford University, Iyengar and Lepper, Journal of Personality and Social Psychology, 2000.
Built for comparison. Formats laid out as separate blocks let the organiser weigh options the way the comparison research says people prefer, one difference at a time. Source: Nielsen Norman Group, 2017.
7. How should published work be presented?
Before: books shown as cover images with no description and no route to purchase.
After: a books page with a description and a purchase route beside each title.
UX and UI analysis, before against after
Before: what the design cost the visitor
Covers with no description. A cover image on its own tells a reader nothing about the book, and the photo research finds an image without accompanying information is treated as decoration and passed over. Source: Nielsen Norman Group, 2010.
No way to buy. The page showed the books but offered no route to purchase, a dead end exactly where the visitor's intent was clearest. Source: Xerox PARC, Pirolli and Card, Psychological Review, 1999.
Nothing an engine could cite. Without a title, a description and a link, the books did not exist as answerable facts on the web. Source: Google Search Central, 2024.
After: what changed and why it works
A description beside every title. Each book now carries a paragraph on what it is and who it is for, which turns a cover into information. Source: Nielsen Norman Group, 2010.
A purchase route next to each book. The action sits beside the item, so a reader who decides can act without hunting. Source: Nielsen Norman Group, 2013.
Entries an engine can cite. Title, description and link give search and answer engines a complete fact to return when someone asks what this author has published. Source: Google Search Central, 2024.
8. How many ways to make contact is the right number?
Before: a contact page with several competing routes and no hint of which is fastest.
After: a contact page with one primary route and the reply time stated.
UX and UI analysis, before against after
Before: what the design cost the visitor
Several routes of equal weight. Email, phone, a form and social links presented equally make the visitor choose a channel before making contact, and more equal options lowered the rate of action in the choice research. Source: Columbia University and Stanford University, Iyengar and Lepper, Journal of Personality and Social Psychology, 2000.
No hint of which was fastest. A producer on deadline needs to know which route gets a reply first; the page did not say, so the fastest visitor had to guess. Source: Nielsen Norman Group, 2013.
No reply expectation. With no stated response time, the visitor had no idea what would happen after sending. Source: Nielsen Norman Group, 2024.
After: what changed and why it works
One primary route. The page names the channel to use and keeps the others secondary. A single recommended path removes the choice the research shows people stall on. Source: Columbia University and Stanford University, Iyengar and Lepper, Journal of Personality and Social Psychology, 2000.
Reply time stated. The visitor is told when to expect a response, which closes the status loop the heuristics require. Source: Nielsen Norman Group, 2024.
Plain words, one screen. The whole page fits where attention is concentrated and says what it needs to in plain language. Source: Nielsen Norman Group, 2017.
Before against after, with the measured column
What | Before | After | Measured |
|---|---|---|---|
Audience routing | 1 page for 4 audiences | 8 pages, 1 audience each | Not measured |
Credential | Below the narrative | Leads the biography page | Not measured |
Media access | Email required first | Usable on the page | Not measured |
Citability | Prose, no structure | Structured headings and metadata | Not measured |
What was built?
A brand system covering the mark, type, colour and layout rules, applied across all eight pages.
Eight pages, each answering one question for one audience: media, clients, speaking and published work.
A media page a producer can use without emailing anyone first.
Structured headings and metadata, so an answer engine can cite the practice with no person in the loop.
The result
What changed, and what is not claimed?
Eight pages shipped, each carrying one audience and one question, on a brand system that covers mark, type, colour and layout. The media page works without an email, and every page is structured so a fact can be lifted from it and cited accurately.
The honest limitation
No performance data exists for this rebuild; no analytics survived it. Everything on this page is a claim about architecture and should be read as one.
Other Projects
Want results like these for your brand?
Book a free call and we will map out what is possible for your project.
Book a call
Demand Designer®
Conversion-focused web design, ad creative and video editing for EOS®-run businesses in the United States.
© Demand Designer 2026. All rights reserved.
EOS® is a registered trademark of EOS Worldwide, LLC. Demand Designer is not affiliated with or endorsed by EOS Worldwide.