The power trio for a faster webshop: Hypernode, RUMvision and Tideways

Black Friday falls on 27 November this year, and for most webshops the last releases before the code freeze are happening right now. A new plugin, a theme tweak, a Shopware update, a tracking tag the marketing team needs for the campaign. Each one is a small bet. Usually it pays off. Sometimes a release makes your product pages a few hundred milliseconds slower on mobile, or doubles the cost of a checkout query, and nobody notices until conversion dips or Google Search Console flags it weeks later.

Those milliseconds are not cosmetic. In a study Deloitte carried out for Google, covering 37 European and American brands and more than 30 million sessions, a 0.1 second improvement in mobile site speed lifted retail conversions by 8.4%, according to the Milliseconds Make Millions case study on web.dev. The research dates from 2019, but the direction has only become sharper as more buying happens on a phone.

At Memo we stopped treating speed as something you check once a quarter. We are a partner of all three parties, Hypernode, RUMvision and Tideways, and we connected their systems so that every release and every update we ship is measured on both page speed and technical behaviour. In this post you will read what each of the three does, why one tool alone always leaves a blind spot, and how the combination works around a release.

Three tools, three views of the same webshop

Hypernode is the foundation. It is a hosting platform built specifically for e-commerce, and Memo is a Hypernode partner. Caching, PHP workers, staging and deployment are tuned to how a webshop behaves rather than left as generic server settings. Tideways is built into the platform. Switching it on takes one setting or a single toggle in the Hypernode control panel.

RUMvision shows what your visitors actually experience. It is a Dutch real user monitoring (RUM) tool that measures Core Web Vitals in the browsers of your real visitors: how quickly the main content appears (LCP), how fast the page responds to a tap (INP) and how much it jumps around while loading (CLS). Unlike Search Console, which works with a 28-day window, RUMvision shows the data in real time, broken down by device, page type and third-party script, and stores it in the EU. Memo is a RUMvision partner, so we set it up, read the data and translate it into concrete fixes for your shop.

Tideways shows what happens inside your application. It is an application performance monitoring and profiling tool for PHP. Shopware itself added a profiler integration with Tideways to its core, as described in Shopware's architecture decision on profiler integrations. Tideways traces show which transaction, database query, template or plugin is eating the time, and for Shopware 6 it also measures how often pages are served from the page cache. As a Tideways partner, Memo configures it per environment and uses it daily in our own development work.

Why one tool is never the whole story

Picture a Tuesday morning. RUMvision shows that LCP on product pages got worse overnight. Is that a frontend problem, such as an oversized image or a new script, or is the server slower to respond? RUMvision can tell you that time to first byte went up. Tideways can tell you which transaction got heavier and which query or plugin caused it. Hypernode gives you the levers to act on it: capacity, cache configuration, PHP version.

It works the other way round as well. Tideways shows that the backend is as fast as ever, yet RUMvision reports worse INP on mobile. Then the cause is in the browser: a heavy third-party script, or a tag someone added via Google Tag Manager.

With only one of these tools, teams end up guessing, or worse, arguing about whose problem it is. With all three, the frontend and backend view sit on the same timeline, and the discussion moves from opinion to evidence.

How Memo connected the three around every release

Being a partner of all three is what lets us go a step further than installing three separate tools. The connection that makes this a trio rather than three separate subscriptions is the release itself. Every time we deploy to a Hypernode environment, that moment is marked in both Tideways and RUMvision.

In Tideways, a release event triggers a before and after comparison of performance, failure rate and errors. By default it compares the 90 minutes after the release with the period before it, and it lists the most improved and most degraded transactions. In RUMvision, the same moment appears as an annotation on the real user data, so you see Core Web Vitals per page type before and after.

The result is one timeline with two perspectives. Within roughly an hour of a release we know whether it was neutral, an improvement or a regression, both for the visitor and for the server. That applies to our own code, but also to Shopware updates, plugin updates and configuration changes. Tideways even supports markers for non-technical events, such as the start of an email campaign, so a traffic spike does not get mistaken for a bad deploy.

What that changes in practice:

• Regressions surface in hours, not weeks. You fix them before they cost a month of conversion and ranking.
  
  
• Rollback decisions are based on data. If a release makes checkout slower, the comparison shows it, and the choice to roll back is not a gut feeling.
  
  
• Improvements become provable. When we optimise something, you see the effect on real visitors, not just a better lab score.
  
  
• Performance becomes part of the definition of done. A release is only finished when it has not made the shop slower.

This is the same measure-first approach behind our work for Kempers.de, where Memo and Hypernode made the shop three times faster in load time without a full migration. That project is described in the Memo x Hypernode case study.

What this means for your webshop in Q4

The weeks before Black Friday are exactly when this setup earns its keep. Measure your baseline now, while traffic is normal. Measure every release leading up to the code freeze, so that the version you go into the peak with is one you know is fast. During the peak itself, alerts in both tools warn you when performance drops, and afterwards you have real data on what the peak did to your shop, which is the best input for next year's preparation.

If you are planning larger changes to your platform next year, this kind of measurement also gives you an honest starting point. It tells you whether your performance problems call for a different platform or simply for better Shopware development and tuning on the one you have.

Frequently asked questions

Learn more about release monitoring for your webshop

A faster webshop is not a one-off project. It is the sum of hundreds of small releases that each make things slightly better or slightly worse. With Hypernode, RUMvision and Tideways connected, you see which of the two every single release does.

Curious what your last release did to your page speed, or want this setup in place before 27 November? Book an introduction call or get in touch and we will look at your shop's performance together.