Jonny Longden

Digital Experience Director | Helping you put experimentation at the heart of your business | Top Experimentation Influencer

My career has been driven by a singular obsession: fundamentally changing the way the business world makes growth and innovation decisions. For nearly two decades, I've forged a unique path using data and experimentation to ignite digital innovation and elevate customer experience.

While the term 'experimentation' has gained traction, I've observed a profound disconnect in its application. Too often, it's relegated to superficial A/B testing or seen as a tactical tweak, failing to deliver its true potential as a strategic driver of growth. This narrow focus leads to siloed learning, unquantifiable investment, and continued reliance on opinion and guesswork for critical business decisions.

This is why I champion Experimentation-Led Growth (ELG). At its core, ELG is about moving beyond mere 'optimisation' to build robust, evidence-driven operating systems for sustained business impact. It's about empowering organisations to de-risk ambitious innovation, accelerate learning velocity, and generate compounding commercial outcomes. My work focuses on integrating Growth Experimentation Systems & Engineering – connecting the strategic vision, the operational infrastructure, and the customer insights to ensure every investment drives tangible, measurable progress.

My unique blend of agency, client-side, and management consultancy experience at senior levels has equipped me with a holistic view of the digital landscape. From leading product analytics for Visa Checkout and building high-impact product discovery teams at Sky, to owning the P&L for front-end Digital Product at Boohoo Group, as well as scaling a digital experience consultancy from scratch, I've consistently applied an experimentation mindset to unlock significant commercial value.

I partner with visionary leaders – VPs of Growth, Product Directors, Digital Directors, and C-suite executives – who are determined to move past the limitations of traditional approaches. If you're looking to transform how your teams learn, ship, and grow; if you're seeking to embed genuine evidence-based decision-making across your organisation; or if you want to unlock step-change velocity and impact from your digital investments, let's connect. I'm keen to share insights and explore how a Growth Experimentation approach can redefine your trajectory.

Contact me at jonny@speero.com or +44 777 909 6222 to find out how we can help you grow.

Director, Digital Experience Services

90% of all changes made to websites are based on opinion and are a waste of time and money because…

It is impossible to know 1000s of customers
Your opinion is deeply biased
You can’t rationalise the irrational

Regardless of the size of your business, the answer is to put data and experimentation at the heart of your development operations. 

We help you do that. 

We take the guesswork out of decision making and provide consultancy and training, systems and frameworks, and experimentation agency services to help you reach your full growth potential.

Email me to learn how to put experimentation at the heart of your business: jonny@journeyfurther.com
Subscribe to my newsletter: jonnylongden.com/newsletter

A few months ago I wrote about evals - the AI equivalent of A/B testing - and why traditional experimenters need to understand them.

Well, the worlds just got closer.

Anthropic just released Claude Skills 2.0, and buried in the update is something that you need to take notice of: structured evals and A/B testing for AI skills.

What is it?

A "skill" is a reusable set of instructions that teaches Claude how to do a specific task - write a brief, triage research findings, summarise test results, generate hypotheses from analytics data. Think of it as a playbook you'd give a new analyst, except the analyst is an LLM.

The problem until now? No way to reliably test whether the skill actually works. Sound familiar? It should. It's the same problem we've spent years solving in experimentation programmes: how do you know a change is actually better, and not just different?

Skills 2.0 introduces:

→ Evals: structured test cases that benchmark a skill's output against defined success criteria
→ A/B comparison: run the same task with the skill on vs. off and measure the difference
→ Iterative optimisation: refine the skill based on results, re-test, repeat

This is literally the experimentation loop. Hypothesis → test → measure → learn → iterate.

Here's where it gets interesting for traditional CRO people specifically.

As the business world adopts Claude Code and similar tools to manage orchestration layers within both internal applications and external customer products, skills will become the playbooks for how things get done.

Your customer support team has a skill that triages tickets and drafts responses. Is it actually resolving issues faster, or just responding faster? Your product team has a skill that personalises onboarding flows. Is it increasing activation, or just generating plausible-sounding copy? Your marketing team has a skill that writes landing page variants. Are they converting better, or just different?

These are experimentation questions. And they need experimentation people to answer them.

The difference is that instead of testing what a user sees on a webpage, you're testing the quality of the instructions that power the system behind it. The skill IS the treatment. The eval IS the experiment.

This is where the experimentation discipline expands. Not replacing A/B testing on websites — but extending the same rigour to the AI layer that increasingly sits underneath everything.

If your organisation is building with AI, someone needs to be testing whether it actually works. That someone should be you.

My original article: https://lnkd.in/dT7c36Sx

#experimentation #AI #CRO #evals #LLMs

Chief Growth Officer @ Speero | Growth Experimentation Systems & Engineering | Product & Digital Innovation Leader

If you made 7-figures and fly around in private jets, why do you need to sell that playbook to people on LinkedIn? 🤔

Hello #Oslo. Just getting set up here for tonight's 𝗘𝘅𝗽𝗲𝗿𝗶𝗺𝗲𝗻𝘁 𝗟𝗶𝘃𝗲 𝗺𝗲𝗲𝘁𝘂𝗽.

If you are working in digital product, growth, or experimentation and fancy some genuinely good conversations without the usual agency sales pitches, we kick off in just a few hours.

We have a great crowd of local leaders confirmed, but there is always room to squeeze in a couple more if you want to make a last-minute decision to join us. We will be digging into the messy reality of scaling experimentation, the psychology of user friction, and optimising for long-term loyalty—the stuff that actually drives business outcomes, not just superficial metrics.

The details:
• 𝗪𝗵𝗲𝗿𝗲: Mesh Nationaltheatret (Tordenskiolds gate 2, 0160 Oslo)
• 𝗪𝗵𝗲𝗻: Today. Doors open at 17:00, talks start around 18:00
• 𝗪𝗵𝗮𝘁: Good food, drinks (on us), and evidence-based discussion.
  
Grab a free ticket via the link below (or just turn up and come say hello). See you later.

https://lnkd.in/dK-xWD5e

#experimentation #cro #productmanagement #growth #digitalexperience #experimentationledgrowth #elg #growthexperimentation

#OSLO, you did not disappoint. Last night’s 𝗘𝘅𝗽𝗲𝗿𝗶𝗺𝗲𝗻𝘁 𝗹𝗶𝘃𝗲 𝗺𝗲𝗲𝘁𝘂𝗽 was genuinely brilliant.

It is incredibly refreshing to get a room full of product and growth leaders together and not just talk about tactical A/B testing or surface-level metrics. Instead, we got into the actual, complex reality of human decision-making, customer behaviour, and building a culture of evidence.

A massive thank you to our speakers for bringing real substance to the table:

• Monica Solberg brought the reality check on what theory and glossy paper won't tell you about the messy reality of scaling experimentation.
• Håvard Maråk challenged the room to look beyond immediate conversions and optimise for actual, long-term loyalty.
• Christina S. gave a masterclass on the psychology of friction and how tiny moments make or break digital products.
  
Huge thanks to our co-hosts Jonas Moe and Joachim Furuseth for helping pull this together, and to our partners at AB Tasty for making it happen. Big thanks to Mesh Community for the amazing venue. 

To everyone from the local community who came out, shared a drink, and engaged in some proper critical thinking—thank you. It is exactly this kind of systems-level thinking that the industry needs to move forward.

Let’s keep the conversation going.

#experimentation #cro #productmanagement #growth #digitalexperience #experimentationledgrowth #elg #growthexperimentation

Building a sustainable experimentation program in a complex organization is not about running the most tests; it is about building a process that can withstand reality. In a B2C environment, the difference between a minor tweak and a strategic growth lever often comes down to the maturity of your experimentation culture.

This is why I am pleased to have Bjarn Brunenberg joining us at CIRCUS London this May.

Bjarn is the Experimentation Lead at TomTom and a recognized influencer in the A/B testing space. With deep experience in scaling growth for B2C teams—including his time at TomTom and Mollie—he specializes in turning raw data-driven insights into high-impact strategic growth initiatives. He is a two-time award winner known for his ability to organize and scale experimentation processes from the ground up.

At CIRCUS, Bjarn will be delivering a lightning talk—a high-energy, concise presentation designed to get right to the heart of a timely topic. These sessions are focused on sharing key messages and inspiring new conversations without the filler of a traditional conference format. He will also be presenting a session in our AI poster area, offering a more hands-on look at building automated growth workflows.

If you are a leader looking to move beyond siloed testing and toward a unified experimentation framework that drives real ROI, you need to see Bjarn in action.

Secure your seat at the table: https://lnkd.in/eXiYwFZu

See the full lineup for London (May 11 and 12): https://speero.com/circus

#Experimentation #GrowthMarketing #B2C #CIRCUSLondon #ABTesting #DataDriven #ProductGrowth #TomTom

Here's why every business should care about their product model.

The true product operating model has been demonstrated time and again to be, in most businesses, a more effective and efficient method for innovation and growth than the feature factory model.

One of the biggest, yet unseen, casualties of the feature factory is your data. Data is the lifeblood of your business and is essential for driving validated learning and growth.

You might not realise it, but your engineers and product teams are the ultimate custodians of your data. They are the ones who code the ability to capture it. Even if you have a fancy Tag Management System, it is often only as good as its implementation by your developers.

The significant problem with the feature factory is that nobody cares about the data. If product and engineering teams are just trying to deliver features on time, they rarely care whether a feature can be measured or optimised, or whether it provides any opportunity for learning and growth. They just want to ship it and move on.

While feature factories neglect data, true product teams are the custodians of growth through product experience, and to succeed, they absolutely need good data.

#experimentation #product  #productmanagement #growth #digitalexperience #growthexperimentation #featurefactories

Most growth teams are struggling with a "Frankenstein Stack" - a cobbled-together mess of isolated analytics, disconnected testing engines, and siloed data platforms that simply don’t talk to each other.

We recently collaborated on a research project with Amplitude and Eric Metelka to understand why organizations like FanDuel, Zwift, and Huckleberry Labs are moving toward a unified approach.

The findings confirmed what I see constantly in my consulting work: the biggest bottleneck to growth isn't a lack of ideas - it's the friction of disconnected systems. When your analytics and your testing engine don't share the same DNA, you aren't running an experimentation program; you're running a data-reconciliation program.

The technical "Unlocks" we found had massive commercial implications:

• 𝗩𝗲𝗹𝗼𝗰𝗶𝘁𝘆 𝗨𝗻𝗹𝗼𝗰𝗸: FanDuel more than doubled their experiment velocity in year one. By removing the friction of context-switching between tools, they shifted from simple "did we break it?" validation to genuine exploration and optimization .
  
• 𝗧𝗿𝘂𝘀𝘁 𝗨𝗻𝗹𝗼𝗰𝗸: Zwift has moved to testing in production. They trust their system enough to run live employee cohorts to catch bugs before a full rollout . Trust in data is the only way to kill the "culture of hesitation".
  
• 𝗜𝗱𝗲𝗻𝘁𝗶𝘁𝘆 𝗨𝗻𝗹𝗼𝗰𝗸: Huckleberry Labs solved the "identity crisis". In a world of shared accounts and multiple devices, they moved from device-level guesses to user-level resolution without any extra engineering lift .
  
If your experimentation program is stagnant, stop blaming your ideation. Stop blaming "opinion-driven" stakeholders. Look at your foundation .

Check out our new whitepaper, "𝗧𝗵𝗲 𝗙𝗿𝗮𝗻𝗸𝗲𝗻𝘀𝘁𝗲𝗶𝗻 𝗦𝘁𝗮𝗰𝗸 𝗣𝗿𝗼𝗯𝗹𝗲𝗺," attached. 

For the full strategic breakdown of why organizations are making this switch, read the companion article here: https://lnkd.in/dqut_CqF

𝗪𝗲𝗯𝗶𝗻𝗮𝗿 𝗔𝗻𝗻𝗼𝘂𝗻𝗰𝗲𝗺𝗲𝗻𝘁:

I’ll be sitting down with Eric Metelka from Amplitude on March 19th to discuss how to turn your stack from a graveyard of disconnected data into an engine for continuous innovation.

𝗥𝗲𝗴𝗶𝘀𝘁𝗲𝗿 𝗵𝗲𝗿𝗲: https://lnkd.in/dB6TrBv3

Stop optimizing the tool. Start optimizing the system.

#experimentation #cro #productmanagement #growth #digitalexperience #experimentationledgrowth #elg #growthexperimentation

Brand partnership • Chief Growth Officer @ Speero | Growth Experimentation Systems & Engineering | Product & Digital Innovation Leader

The Frankenstein Stack Problem

Experiment Live Meetups is coming to #Oslo #Norway!! 𝗧𝗛𝗜𝗦 𝗪𝗘𝗘𝗞 on 25th February and is ALMOST SOLD OUT. Register to secure your FREE place!

This is a FREE informal evening networking meetup for folks in digital growth, product, experimentation, data and UX, to be held in a TBC central Oslo venue. 

Experiment live meetups are a way to bring people together to meet, talk and learn from each other. There is NO SALES and NO PITCHING. The event is mostly chatting, eating and drinking (all provided), but we will also have a few talks from:

Monica Solberg, Head of Experimentation, Vend | “What theory and glossy paper doesn't tell you about scaling experimentation”

Christina S., Global Product Manager, Opera | "The Psychology of Friction: Why Tiny Moments Make or Break Digital Products"

Håvard Maråk, Manager of Dynamic Sales & Experimentation, Schibsted | "Optimising for Loyalty"

We're co-hosted by Jonas Moe of TV 2 and Joachim Furuseth of Coop Norge, and we have confirmed registrations from your peers at Norwegian, Hurtigruten, reMarkable, Tietoevry, Helly Hansen, Diakonhjemmet, Too Good To Go, Morrow Bank, FINN.no – mulighetenes marked, Apotek 1, Maverix AS,  Bislab,  Boxn, SuperOffice, EGGS, Part of Sopra Steria and more and counting. 

The event will be held at Mesh Community - Tordenskiolds gate 2, 0160 Oslo - https://lnkd.in/edaXyfNV

The event will be mostly informal chatting with above short talks;  light food and drinks will be provided. 

Thanks to our amazing sponsors AB Tasty; Kate Y. Feng, John Hughes 

REGISTER HERE: https://lnkd.in/dK-xWD5e

#experimentation #cro #productmanagement #growth #digitalexperience #experimentationledgrowth #elg #growthexperimentation

Imagine if, after JFK said this, NASA had got together and decided that the answer was simply to create a 'culture of experimentation' and to make sure that everyone had the tools they needed to run experiments.

There would never have been a person on the moon.

The modern business focus on experimentation is an obsession with tools, techniques and capabilities, often without any substance.

The moon landings succeeded because of the centralised, coordinated management of meaningful experiments and learning towards this goal and vision.

This is very different from just letting people run experiments.

Growth experimentation is, first and foremost, the critical thinking and management that tries to achieve strategy, vision and innovation. The technical aspects of running tests are merely one of the ways that this is achieved.

#experimentation #cro #productmanagement #growth #digitalexperience #experimentationledgrowth #elg #growthexperimentation

Host @ nohacks.show podcast | Chief Growth Officer @ FOSH

I help brands see what's actually broken on their site - and what to fix first.

Most sites are built for sighted humans only. But your site has three audiences: humans, assistive technologies, and AI systems. They all read your site through code. If it's broken for one, it's usually broken for all three. That's costing you traffic, conversions, and visibility.

I run technical audits covering performance, accessibility, and AI readiness. I also host a podcast about optimizing the web for AI agents. If your site is slow, inaccessible, or invisible to the next generation of search - let's talk.

---

Fifteen years in web development, web performance, technical SEO and CRO have shown me which levers move the metrics that matter. I work side by side with design, development and marketing teams to ship improvements instead of reports.

I’ve helped teams at Instacart, Klarna and Search Engine Journal launch products, increase conversions and speed up their websites. I'm also a keynote speaker, CXL-certified CRO (conversion rate optimisation) specialist, WordPress Core Contributor and WordCamp Porto 2024 and Future of eXperimentation 2025 organiser.

Slobodan Manić

First of all, Friday shoutout to Descript and their Studio Sound effect for saving the most important recording in history of No Hacks Podcast. Next door neighbour doing construction work at 19h is one of those uniquely Southern European things, but somehow my audio is usable. Happy customer here.

Second, no matter what kind of a website optimizer you are - SEO, CRO, or anything else or in-between, the future of your work is upstream. Adapt.

Episode with Rand Fishkin dropping next Wednesday. 

Subscribe to the podcast at nohackspod.com/subscribe

Web optimisation for the AI era | Keynote speaker | Host of the No Hacks podcast (200+ eps)

I got to interview Rand Fishkin for No Hacks and part of my brain is still processing the fact :)

Rand has been one of the sharpest voices in marketing for over two decades. And he recently published research that should make anyone selling (or buying) "AI visibility rankings" very uncomfortable. You can find the link in the comments.

You'd need to ask ChatGPT 1,500 times to get the same brand list in the same order twice. Less than 1 in 100 chance of an identical list. Less than 1 in 1,000 for identical ordering.

That's a lottery, not a ranking system.

We covered a lot:

AI tracking tools are selling phantom metrics. There's $100M+ being spent on tools measuring something that has no stable reality. Rand's advice: demand transparent methodology or walk away.

Your homepage is not your homepage anymore. It's what Google says about you. What ChatGPT says about you. What Reddit says about you. First impressions now happen on platforms you don't control.

The best brands of the last 30 years weren't built on digital marketing. Classic brand building never stopped working.

Google is still 210x bigger than ChatGPT. And ChatGPT usage actually declined last quarter. The narrative doesn't match the data.

Rand basically told me the optimization everyone's selling doesn't work. But he's not saying give up. He's saying the lever is different than what people think. It's not your website. It's what everyone else says about you.

This one's special. Get the episode link in comments.

P.S. If you're new here, I run a podcast about optimizing the web for LLMs and AI agents. Subscribe at nohackspod.com/subscribe

Hey LinkedIn, how about instead of focusing on killing everyone's reach if they don't want to pay for "thought leader ads" and completing the final stage of platform enshittification, maybe you could do something like this, at least for comments?

You know, just give users information about how each comment was created, at least under their posts? Or maybe something as radical as full transparency where every comment has this meta information for anyone to see? Number of keystrokes vs number of characters, keystroke log...

If you want the users to spend as much time on the platform as possible, interact with each other and comment, maybe focus on quality, rather than copy/pasted AI-generated nonsense?

With machines now consuming the internet as much, if not more than humans do, it seems very important to make sure the content they are consuming is created by humans, doesn't it? 

Too radical? Should I boost this post so people I care about can see it?

No podcast episode this Wednesday (planned week off), but there's something really cool hitting the feed next week. 

Here's a short clip from next No Hacks episode with Rand Fishkin, coming out on March 4th. It's crazy that someone I've been learning from and had massive respect for, for literally 20 years is now on the podcast, but here we are :)

Link to SparkToro study we discussed: https://lnkd.in/e_c78TaS

taplio

The Battle

Jonny Longden

Slobodan Manić

Followers

Posts

Engagements

Likes

Performance ScoreJonny Longden's LinkedIn Performance

Performance ScoreSlobodan Manić's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Jonny Longden

Slobodan Manić

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Jonny Longden's LinkedIn Performance

Performance ScoreSlobodan Manić's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreJonny Longden's LinkedIn Performance