A Dutch building-materials merchant with 36,000+ SKUs, two branches and its own delivery fleet moved its Magento 2 storefront to the full Hyvä Commerce suite. Revenue up 21.3% on fewer sessions, checkout entries up 41.7%, and mobile, where the customers actually are, the biggest winner of all.
Website: sleiderink.nl
Country: The Netherlands. Dutch-language webshop, nationwide delivery, both branches in Overijssel
Industry: Building materials: timber, sheet materials, insulation, facade cladding, doors, paving, masonry, roofing
Business model: B2B and B2C, omnichannel. Webshop plus two branches (Denekamp head office, Oldenzaal service branch), counter sales, click-and-collect and own delivery
Catalogue: 36,000+ SKUs
Platform: Magento 2 Open Source 2.4.9 (PHP 8.4)
Hyvä products: Hyvä Commerce: Hyvä Theme, Hyvä UI, Hyvä Checkout and Hyvä CMS
A trade that moved online early and then had to keep up
Frank Sleiderink founded an independent building-materials wholesaler in Denekamp in 1986. A second branch followed in Oldenzaal in 2000, online selling in 2012 and a dedicated webshop in 2014 well ahead of most of the sector. In 2026 the company marked its fortieth year, still family-owned and tied to no buying group or manufacturer.
The proposition is a genuine one-stop shop: everything for a project from foundation to roof, from one supplier, with an in-house sawing facility for cut-to-size timber and sheet material and advisors who guide customers from plan to delivery. Professional buyers join Sleiderink Premium for personal pricing, a dedicated contact and project discounts. Customers rate the company 9.2 out of 10 on Kiyoh.
The customer base splits in two. Contractors, carpenters and installers reorder the same things on account at negotiated project prices; self-builders compare and ask before they buy. Both are strongly mobile: a large share of orders is placed on site, on a phone, on a patchy connection, often with gloves on.
The ERP was fine. The frontend was not
Everything commercially interesting about Sleiderink lives in its ERP. Customer-specific pricing, project pricing, per-branch stock, orders and quotations all run through Kerridge K8. The problem sat in the layer on top.
Performance. The old Magento 2 frontend carried the weight of a traditional theme. Field data before the migration put mobile LCP at 2,117 ms and mobile TTFB at 1,450 ms at p75, on exactly the surface where most orders are placed.
Legacy code. Years of accumulated overrides made every frontend change slower and more expensive than it should have been.
Step-by-step design. The site had grown feature by feature and needed a refresh, without disorienting professionals who order weekly and know exactly where everything is.
Checkout. In the words of the lead backend developer, the part of the shop nobody wanted to touch.
Content. Landing pages and campaign content needed a developer. Any change that needs a developer is a change that waits.
So the brief wrote itself: replace the frontend with Hyvä Commerce, leave the commercial engine alone.
How Sleiderink found Stimmt
There was no RFP and no referral. Sleiderink kept hearing good things in the market about the work Stimmt was doing, and reached out on that basis. What decided it was the opposite of a pitch.
Instead of a deck, the first sessions were spent going through the database and the configuration of the existing shop together with the Sleiderink team, an honest assessment of what was there, what was solid and what was not. That set the tone for the daily, direct collaboration between the two teams that the project still runs on.
Why a suite, not a shopping list
The second decision was what to buy. Sleiderink took Hyvä Commerce: the whole suite, under one licence, rather than assembling a frontend from separate parts.
For a merchant of Sleiderink's shape, buying the suite rather than the pieces changed four things:
One stack, end to end. Theme, checkout and CMS are built the same way, on Alpine.js and Tailwind, from the same component library. There are no seams between the storefront and the checkout to design around, and no third component to make it compatible with the other two.
One upgrade path. New features arrive in the same subscription instead of as separate purchases, each with their own version and their own compatibility question.
The marketing team gets the keys. Hyvä CMS moves content and landing pages out of the development queue. Existing Page Builder content stays intact and can be migrated page by page, which mattered for a site with a large knowledge base.
Predictable cost of ownership. One licence, one renewal, one support channel: against a Magento extension ecosystem where the same functionality would arrive as several subscriptions.
Magento 2 Open Source stayed. Magento remains the right fit for a catalogue this size and this complex, for customer-specific and project pricing, and for the depth of the Kerridge K8 integration, none of which a lighter SaaS platform handles well. Migrating the frontend rather than replatforming cut the risk of the project roughly in half.
One honest caveat, and Stimmt raised it themselves: Sleiderink took every part of the suite at once: Theme, Checkout, CMS and Commerce, alongside a complete redesign and a migration off a custom page builder.
All of it had to be tested in detail, and the content migration into Hyvä CMS took longer than expected. It landed well, but a phased route would have shipped sooner. More on that below.
What Stimmt built
A full design refresh and a technical migration to Hyvä Commerce, shipped together and deliberately keeping the recognisable black-and-yellow Sleiderink branding so returning customers were not disoriented.
A new three-column product page layout, which came out of UX testing as the best fit and has performed well since.
Hyvä Checkout for a shorter, faster and more stable purchase flow.
Content and landing pages rebuilt on Hyvä CMS, so marketing can publish independently.
A Premium customer portal for professional buyers: personal pricing, an approval flow, and reusable task lists (order templates) for recurring purchases.
Custom Stimmt modules for quotation ordering, cut-to-size unit quantities and product deposits.
The integration surface stayed rich: Kerridge K8 (orders and quotations out, product master data in, per-branch configuration, with outage monitoring), Mollie payments, DPD Connect plus custom delivery-promise, preferred-delivery-day, branch pick-up and own-fleet logic, Postcode.nl address validation, Snowdog megamenu, Channable product feeds, MailPlus, Kiyoh reviews, and Google Tag Manager via Yireo GTM2 with a custom Stimmt tracking layer.
A fast frontend needs a fast server
Stimmt and Hypernode have worked together for years, and the lines are short – questions during the migration came back answered quickly. Hypernode is one of Hyvä's strategic hosting partners and its platform is tuned for the way Hyvä works, so the brief went beyond hosting the theme: caching, PHP and OpenSearch were configured around it.
A faster frontend alone does not make a fast webshop. Hypernode's slow-module check reads the PHP slow log to show which extensions hold up page requests, so Stimmt fixed the modules that counted rather than guessing.
Testing ran on Hypernode Brancher, a temporary isolated copy of the live shop, where the theme, checkout and every integration were tested against real data – and load-tested on its own resources – without touching production.
Hyvä made the storefront faster, Hypernode made sure that speed holds under traffic, Stimmt cleared the remaining bottlenecks and delivered the build. One client, three parties, one result.
The results - performance converts to business outcomes
Faster, on the connections that matter
Sleiderink's customers order from building sites, so the gain that counts is the mobile one. Largest Contentful Paint at p75 fell 39.0% origin-wide, from 1,933 ms to 1,180 ms.
On mobile every loading metric moved together: LCP −39.5% (2,117 → 1,280 ms), First Contentful Paint −40.5% (1,850 → 1,100 ms) and Time to First Byte −40.7% (1,450 → 860 ms). Desktop improved in step: LCP −40.6%, FCP −44.6%, TTFB −36.8%.
The share of page loads meeting Google's "good" LCP threshold rose from 84.0% to 96.2%, so roughly one in eight visits that used to feel slow no longer does.
Before and after
Swipe left or right to navigate. Double-tap to zoom. Pinch to zoom. Swipe down to close. Use left and right arrow keys to navigate between images. Press plus to zoom in, minus to zoom out, zero to reset. Press Escape to close.
Source: sleiderink.nl, mobile field data, two days before go-live and six months after. Source: PageSpeed Insights, Chrome UX Report.
More revenue, without buying more traffic
Revenue grew 21.3% while measured sessions fell 3.2%. Sleiderink sold more by converting the demand it already had, not by spending more to attract it. Orders rose 8.3%, from 7,762 to 8,408: 646 additional orders and the conversion rate improved 11.9% (0.714% → 0.798%), with the purchaser rate up 14.1%.
Mobile carried the growth
Mobile conversion improved 21.3% (0.544% → 0.660%) against 8.9% on desktop, 2.4 times the gain, and mobile revenue rose 23.3%. Mobile's share of sessions grew from 53.3% to 56.5%. The surface that was slowest before the migration became the biggest commercial winner after it, which is the clearest line in the dataset between the technical work and the money.
Customers get to the order faster
Pages per session fell 21.9% (4.46 → 3.48) while engagement time per session stayed flat (98.7 s → 98.1 s). Professionals reordering the same materials reach them in fewer steps without disengaging. Sessions per user grew 4.1%, so they come back more often.
The checkout gain is the sharpest of all: abandonment between adding to cart and entering checkout fell from 70.7% to 65.2%, and to 63.9% across the most recent three months, while Hyvä Checkout lifted checkout entries 41.7%. Two in five more shoppers now reach the point of paying.
Bigger orders from the buyers who matter most
Average order value rose about 4.2% and items per order rose 25.3%, from 72.7 to 91.0. Orders for the Premium customers grew 36.1%. The programme built for professional buyers is doing what it was designed to do.
The regulars stayed, and spent more
Returning customers generated 21.8% more revenue at a 20.8% higher conversion rate, and accounted for 76.8% of total revenue. For a business built on repeat professional purchasing, a redesign that keeps the weekly buyers and grows their spend, is the result that mattered most.
Search: better rankings, and an honest wrinkle
Average position improved from 14.4 to 12.6 and click-through rose 6.1%. Product and category pages, the commercially relevant half of the site, gained 2.1% more organic clicks on 17.8% fewer impressions as their CTR rose 24% (1.37% → 1.70%), fewer people saw the listings, more of them clicked.
Total organic sessions were down 17.7% in the measured window, concentrated in informational knowledge-base content. That is consistent with the sector-wide shift of informational queries to AI answers, and a new AI-assistant referral channel appeared after launch (2,214 sessions). Commercially the picture held: organic conversion improved 13.1% and 49 new SEO landing pages delivered 5,226 additional organic sessions. 404 errors spiked briefly at launch and normalised within two months.
Marketing stopped waiting for developers
Since go-live the team has averaged roughly 71 merged frontend releases a month, and marketing publishes content and campaign pages in Hyvä CMS without a developer in the loop. Campaigns ship when the market moves, not when the sprint allows.
What the teams say
This result is 100% the result of the intensive and pleasant collaboration we have had for years. I have enormous respect for the people behind this result. Without their involvement and endless effort, we would not be where we are now with Sleiderink.nl
Stimmt's lead backend developer saw some risks during the start of this project. Backend developers usually are, he says, when someone promises that the frontend will resolve the issues at hand.
What changed his mind was what the migration forced the team to confront: overrides that only existed to patch Luma's quirks simply did not come along, and everything that survived the move earned its place. Checkout is his clearest example, the part of the shop nobody wanted to touch became normal, readable code again.
As lead backend developer I measure a stack by where my team's hours go. They used to go into front-end firefighting. Now they go into integrations, order flows and B2B logic, the things that actually move and help grow our clients' business.
On the developer side more broadly: working in Hyvä made the team's day-to-day simpler, leaning on Tailwind to build efficiently against new client requests, and restructuring the codebase during the migration made backend implementation of new features easier.
What they would do differently
Stimmt used what they call the holy grail of Hyvä: Theme, Checkout, CMS and Commerce, all in one go, while also restructuring all content and redesigning the entire website. Everything had to be tested in detail, and because Sleiderink was coming off a custom page builder, the migration into Hyvä CMS took real time. It went well in the end, but the site could have launched somewhat earlier with a different strategy.
Their advice, and Sleiderink's, points the same way: do the theme and checkout migration first, then bring in the rest of the suite step by step. If your existing content sits in regular WYSIWYG pages rather than a custom builder, that migration will be far simpler. And when you migrate, resist carrying things over: remove modules where needed and replace them with Hyvä-compatible ones. It is a good moment to clean up the codebase and drop maintenance load.
Six things Sleiderink would tell another merchant
Migrate the frontend, do not rebuild everything. The commercial logic sits in Kerridge K8: customer-specific pricing, project pricing, orders and it was running fine. Replacing the frontend with Hyvä while leaving that untouched cut the risk roughly in half and shortened the timeline. Be honest about which part of your stack is actually the problem.
Do not use a replatform as an excuse to reinvent your brand. Customers recognise the black and yellow, and professionals who order weekly do not want to relearn where things are. A faster site that feels unfamiliar can still cost you orders in the first weeks.
Test on the conditions your customers actually order in. A lot of orders are placed on a site, on a phone, on a weak connection, with gloves on. That is the benchmark that matters, not a desktop on office wifi and it is where the speed gain pays off most.
Measure before you migrate. Fix your baseline first: page speed, Core Web Vitals, conversion, average order value, organic traffic, over a clean period before go-live. It is the one thing you cannot recover afterwards.
Give the marketing team the keys. With Hyvä CMS, content and landing pages belong to marketing. Check during the project which parts your own team will be able to manage.
Pick a partner who knows your business, not just your platform. Pricing rules, cut-to-size items, delivery and pick-up logic. An agency that already understands that will save you months over one that has to learn it.
About Stimmt
Stimmt builds brand websites and ecommerce powered by AI, perfected by humans. The agency develops strategic custom websites, scalable webshops and applications that integrate with clients' existing systems, using AI as an accelerator for process, performance and conversion.
Stimmt has been doing this for more than 15 years, working mainly in Magento, Shopify, Craft CMS, Prepr CMS and WordPress. Hyvä clients include Sleiderink, Top1Toys, Schipper Techniek and PouchDirect.
