pixelgravitas
Free Assessment

Home/Resources/Ecommerce website structure

Guide · With a 10-point store check

Ecommerce website structure that actually sells, with real numbers from three stores.

Short answer: an online store that sells is boring on purpose. The homepage is a door, not a catalogue. The menu has five to seven items, search is on every page, the price sits next to the product photo, the cart shows the full total, and checkout works as a guest on one page. Most stores that get visitors but no sales have broken one of those rules.

Written by the engineer who builds the stores Real before-and-after numbers, not estimates
Checkout82% → 64%Abandonment after switching to guest checkout.
Homepage71% → 48%Bounce rate after cutting 12 sections to 3.
Prices shown2×Product-page conversions the month prices came back.
Mobile target< 3 sLoad time on a mid-range Android phone.

The question behind this guide

Why does my online store get
visitors but no sales?

Usually the layout, not the products. In the stores we build and fix, the same four problems come up again and again, and each one sends a ready-to-buy visitor somewhere else.

The homepage is a catalogueEvery collection, offer, video and feed on one page. It looks rich in a mockup, loads slowly on a phone, and nobody scrolls far enough to reach a product. See the boutique case.
The price is hidden or hard to find"Call for price", "login to see price" or a price below the fold. Shoppers use price to decide whether to keep looking. See the furniture case.
Checkout asks too muchA required account, extra fields, several steps. Each one is a reason to leave with a full cart. See the supplement case.
It's slow on a phoneSliders, autoplay video and heavy scripts. If the first product takes eight seconds to appear, most visitors never see it. See the speed target.

The standard, page by page

What pages does an
ecommerce website need?

A good ecommerce website structure is predictable: a customer should never have to think about where to go next. This is the structure we build every store on, and what each page has to do.

PageWhat it must doWhat to leave out
HomepageOne clear promise, one hero product or collection, bestsellers, a little trust proof, then the footer. It's a door.Every collection, three banners, an autoplay video, the Instagram feed and the blog.
NavigationFive to seven top items: Shop, the categories people search for, Offers, About, Contact.A dropdown inside a dropdown.
SearchVisible on every page: top right on desktop, in the sticky header on mobile. Handles typos and shows product photos in the results.Hiding search to keep the design minimal.
Category pageFilters for price, size, colour and availability (on the left on desktop, one tap to open on mobile). Sort by popularity, price and newest.Endless scrolling with no way back to the top.
Product pagePrice up front, clear photos, size or variant selector, delivery information, an Add to Cart button that stays visible, reviews below.Making people hunt for the price.
CartEdit quantity, remove items, see the total including delivery, reach checkout in one click.Delivery costs that appear for the first time at checkout.
CheckoutGuest checkout by default, one page if possible: name, contact, address, payment. Cards, Apple Pay and Google Pay, PayPal.A required account, extra questions, animated steps.
AccountOptional, offered after purchase: order tracking, reorder, saved addresses.An account as a gate before buying.

Check your own store · 5 minutes

Does your store pass
these 10 checks?

Open your store on your phone, as if you'd never seen it before, and tick each one that is true. Be honest: this is what a first-time visitor experiences.

Your score: tick each check your store passes.

Real case · A fashion boutique

Should my homepage show all
my products and offers?

No. The homepage is a door, not a catalogue: one promise, one hero, bestsellers, one offer. A fashion boutique we rebuilt asked for the opposite, and the numbers show what it cost.

What she asked forTwelve homepage sections, an autoplay video, a lookbook slider and three different offer banners.
What we warnedMobile load time would go over six seconds, and people would leave before they saw a single product.
What happenedMobile load hit 8.2 seconds and homepage bounce reached 71%. Sales fell about 30% in the first month compared with the old, simpler site. Most support messages said the site was slow, or asked where the dress from Instagram was.
The fixAfter six weeks we cut it back to one hero, bestsellers and one offer. Bounce fell to 48% and sales recovered within two weeks. The detour cost about three weeks of rework plus the lost month.

Real case · A supplement brand

Should customers have to
create an account to buy?

No. Guest checkout should be the default, with an account offered after the purchase. Forced accounts are one of the most common reasons shoppers abandon checkout, and every extra field adds another.

What he asked forForced account creation, extra checkout fields (date of birth, company name, "how did you hear about us") and a three-step checkout with a progress bar that looked premium.
What we warnedEvery extra field costs sales, and a required account is one of the biggest reasons people give up at checkout.
What happenedCart abandonment hit 82% in the first ten days, and more than 40 support tickets arrived in a week, most saying some version of "I just wanted to buy one bottle, why do I need an account?"
The fixAfter 11 days: guest checkout on one page, asking only for name, contact, address and payment. Abandonment fell to 64% within a week. The detour cost him his launch-offer window.

Is a one-page checkout better than a multi-step one? For most small stores, yes. Ecommerce checkout best practices come down to one rule: ask only for what you need to deliver the order and get paid. A redesigned checkout that adds steps, animations or custom fields to look premium usually does the opposite of what it was meant to.

Real case · A furniture store

Should I hide prices and use
"request a quote" instead?

Only for genuinely custom work. Shoppers use price to filter; when they can't see it, many assume it's too expensive and leave. "Login to see price" and "call for price" suit a luxury showroom, not a normal online store.

What he asked forNo prices anywhere, only an "Enquire now" button, because he felt it made the brand look high-end.
What we warnedPeople use price to decide whether to keep looking. If they can't see it, they assume it's expensive and leave.
What happenedEnquiries rose a little, but sales fell and product-page bounce reached 78%. People enquired, received a quote two days later, and went quiet.
The fixAfter a month we put prices back and kept "Enquire" only for custom sizes. Product-page conversions doubled the following month, because the people who enquired already knew the price range.

More questions store owners ask

Speed, popups and looking
like a big brand.

How fast should an online store load?Under three seconds on a mid-range Android phone, tested on mobile data. That's the phone and connection many of your customers actually use, not your office computer.
How many popups are too many?One. An entry popup, a chat popup after ten seconds, an exit popup and a spin-the-wheel together make the site feel like it's begging, and they cover the Add to Cart button on phones.
Should my store look like Apple or Nike?No. Big brands can afford an unusual flow because people already trust them and will fight through it. A small store has a few seconds to earn that trust.
Is a minimal, artistic design a problem?When it costs usability, yes: tiny grey text on white, image-only buttons with no labels, no search because it spoils the look. Style is fine; making the buyer guess is not.

Where these requests come from

Why store owners ask for
layouts that cost them sales.

The ecommerce design mistakes on this page are rarely a developer's idea. They come from one of four places, and none of them involves a customer.

A big brand does itCopying the look of Nike or Apple feels like copying their trust. It doesn't come with the look.
A designer's mockupA concept on Dribbble or Behance that looked stunning and was never tested with a real shopper.
A competitor did itWithout knowing whether it worked for them, or whether they've quietly changed it back.
Personal taste"I like it dark, moody and mysterious." Taste belongs in the brand, not in the buying path.

Where we hold the line

What can I customise, and
what should stay standard?

The buying path is not a design choice. It's where the sales happen. Everything around it is where a brand's taste genuinely helps.

We never compromise onThe store owner decides
Guest checkoutColours
The price visible on the product pageFonts, as long as they stay readable
Add to Cart always visible on mobilePhotography style
Working searchTone of the copy, playful or serious
Under 3 seconds on a mid-range Android phoneThe order of storytelling sections below the fold

We stopped arguing about taste. The sentence that actually changes minds is this one: if a first-time visitor needs instructions to use your site, they'll just buy from Amazon instead.

It moves the conversation away from my taste versus yours, and towards the person who matters: a tired shopper on their phone at 10pm with five other tabs open.

Before you hire anyone

Six questions to ask whoever
builds your online store.

Whether you hire us or someone else, these questions show quickly whether a developer will build for your buyers or just for the mockup.

AskA good answer sounds like
Will customers be able to check out as a guest?"Yes, by default. Accounts are offered after the purchase."
How fast will it load on a phone, and how will you test that?A number, a device and a tool, not "it will be fast".
What will you push back on if I ask for it?Specific examples. A developer who agrees to everything is building your mockup, not your sales.
How will shoppers find a product in two clicks?A clear plan for the menu, search and category filters.
What happens after launch?Who updates WooCommerce and plugins, tests them first, and fixes checkout if it breaks.
Can I see a store you built, on my phone?A live store you can try yourself, from homepage to checkout.

Who wrote this

Written by the engineer
who builds the stores.

Naveen Kumar, founder and lead engineer of PixelGravitas

Naveen Kumar

Founder & lead engineer · Working 9am–6pm US Eastern with clients across the US, UK, Europe and Australia

I've built and rebuilt many WooCommerce stores, and the cases on this page are real ones, anonymised. The structure above is what we build every store on, through our website redesign services, and what we protect afterwards through WooCommerce maintenance. More about how I work →

Questions, answered

Ecommerce website structure:
common questions.

Why does my online store get visitors but no sales?

Most often because of the layout, not the products. The usual causes are a homepage so crowded that visitors never reach a product, prices that are hidden or hard to find, a checkout that asks for an account or too many fields, and pages that load slowly on phones. Check those four before spending more on traffic.

What pages does an ecommerce website need?

A homepage, category pages with filters, product pages, a cart and a checkout, plus search on every page and an optional customer account for order tracking. Add About, Contact, shipping and returns pages for trust. Anything beyond that should earn its place by helping someone find or buy a product.

Should customers have to create an account to buy?

No. Make guest checkout the default and offer an account after the purchase, for order tracking and faster reorders. When one store we built forced account creation, checkout abandonment reached 82% in the first ten days; switching to guest checkout brought it down to 64% within a week.

Should I hide prices and use a request-a-quote button instead?

Only for genuinely custom products. Shoppers use price to decide whether to keep looking, and when they can't see it many assume it's too expensive and leave. A furniture store we worked with doubled product-page conversions the month after putting prices back and keeping quotes only for custom sizes.

Is a one-page checkout better than a multi-step checkout?

For most small stores, yes. Keep checkout to one page if you can and ask only for what you need to deliver and get paid: name, email or phone, address and payment. Every extra field, step or required account adds a reason to leave before paying.

How many items should an online store menu have?

Five to seven top-level items is enough for most stores: Shop, the categories people actually search for, Offers, About and Contact. Avoid menus nested inside menus. If a visitor needs more than one level to find a category, the categories need rethinking, not the menu.

How fast should an online store load?

Aim for under three seconds on a mid-range Android phone over mobile data, not on your office computer. A boutique homepage we rebuilt went from 8.2 seconds on mobile, with a 71% bounce rate, to a fast, simple page with a 48% bounce rate once the extra sections came out.

How many popups are too many on an online store?

One. A single, well-timed offer or newsletter popup is fine. An entry popup, a chat popup after ten seconds, an exit popup and a spin-the-wheel together make the store feel like it is begging, and they get in the way of the Add to Cart button on phones.

Should my online store look like Apple or Nike?

No. Big brands can afford an unusual layout because people already trust them and will work through it. A smaller store has to earn that trust in a few seconds, so it gets more from a familiar layout that a first-time visitor can use without thinking.

Which parts of my store's design can I customise freely?

Colours, fonts as long as they stay readable, photography style, the tone of your copy, how playful or serious the brand feels, and the order of storytelling sections below the fold. Keep the buying path standard: visible price, guest checkout, a visible Add to Cart button and working search.

Start here

Want your store checked against this list?

Send the URL. We'll go through it as a first-time shopper on a phone and tell you, in writing, which checks it fails and what to fix first.

  • 1We test the live store on mobile: homepage, search, a product, the cart and checkout.
  • 2You get written findings against the 10 checks, most costly first.
  • 3You decide what happens next. No follow-up sequence if you don't reply.

Get a free website assessment

Three fields on purpose. We reply within one business day, and we don't add you to a mailing list.

Get a free website assessment