It’s been a while since I sent a newsletter, things have been quite chaotic on my end. Normally a piece takes me 2-3 weeks to research & write, and to be honest, I did not have that type of time in the last month. Between parenting, work, life and events, I’ve spent most of my free time hiding in a corner, surfing Reddit and watching trashy reality tv & Swedish/ Danish Border Control TV series.
But, I’m slowly getting back in the game, so let’s get to it.
What a German conference reminded me about b2b sales business models
I was at d3con in Hamburg this week. The conference is about 98% German, which, fair enough, it is Germany, but it meant I spent half of the day nodding along like a dumbass, realizing my level 5 German Duolingo skills meant fuck all. There were only 4-5 sessions in English, including my own.
I won’t talk much about my session in this edition, but, I will say that even if I had the graveyard speaking spot, thanks to the team for having me. I had a good time connecting to the audience and thanks to the people that showed up to hear something different, something radical and not complacent. Also appreciated the folks that reached out to me at the end and shared their love for my talk, wish I'd connected with you all on LinkedIn but I got hit by the Oscar's song 🎵 😂
But hey, here is a pic!
Now, funny enough, I found out halfway through the event that translation headphones were sitting at reception the whole time. Classic.
Once I started understanding what was going on, I realized I’d stumbled into a consultant’s paradise. I’ve never seen a higher % of brand-side speakers on stage. There were no hot takes (BTW, why does every time someone says they have a hot take, in fact, it’s quite a lukewarm one?…), no paradigm shifts lol, and no one announcing that everything we knew was wrong.
Instead, it was some sort of honesty from these folks that are running marketing inside large, complex, traditional businesses, that you don’t often hear at events where vendors and agencies set the agenda. They were just describing their businesses in the most matter-of-facty way.
There were 2 panels that really stuck with me. And man, I’m not a fan of panels in general. The first one was about programmatic advertising, with a focus on CTV, here the panelists were folks from Beiersdorf, Cosnova, Spotify, Netflix & Google, and the other one was about campaign optimisation and attribution, with folks from Deutsche Bahn and Unilever.
Now, the reason they stuck with me wasn’t really the topics discussed, although the panelists were genuinely great and a lot of good stuff was covered, but because they reminded me once again how much business models matter.
I especially liked Tanja Leutschacher’s (Cosnova) take about how important it is for them to be where the audience is. Cosnova has 2 brands, Essence and Catrice and their marketing strategy is social first, because their audience is social-first. What is the point in their case investing in CTV if their audience lives and breathes TikTok or Instagram and learn about their products from GRWM’s, influencer content? And keep in mind, Cosnova has a very complex cross continents retail machine and distribution models.
Another brand (Beiersdorf) mentioned that for Nivea alone, a huge chunk of their core customer base in Germany is still watching linear television, and that their main buyer is someone, 40+, who takes their news about their latest products from traditional TV. They aren’t going anywhere, and their strategy reflects that reality rather than the one being dictated by the trade press.
Unilever Foods pointed out that when you’re selling products that cost 2 euros in a grocery aisle, “attribution” isn’t a tech problem to be solved with a better dashboard. It’s just the physics of the business.
Neither of these things is surprising, for sure, but, the more I think of it, the more I realise the industry treats them like edge cases.
We’ve built a narrative based on digitally native brands. Companies that sell direct, own their data, and have feedback loops that close in days. Then, we hand that playbook to giant FMCG brands and act shocked when it doesn’t fit.
For example, a brand like Nivea wasn’t built on a retargeting ad but over decades of mothers passing it down to daughters. When the sale finally happens, it’s in a supermarket in Warsaw or a pharmacy in Hamburg, separated from the advertising by layers of retail margins and shelf placement negotiations. That relationship exists between the customer and the product, not a digital touchpoint. The feedback loop is long, indirect, and complicated in ways no “modern” framework is going to untangle.
Let’s not forget, this is also a geography problem. The industry consensus, that is headlining conferences and writing “the playbooks” operates in a very specific US bubble. They talk about “global strategy” while describing a tiny slice of venture backed commerce. But if the dominant industry narrative doesn’t even map onto Germany, the largest economy in Europe, then it begs the question of how actually useful is it in other geographies?
Also, don’t mistake the lack of “disruptor” jargon in my take for a lack of modernity. This is not what I am saying at all. These businesses are innovating at a level of complexity that makes most tech first brands look like sitting ducks.
The reason they don’t “move fast and break things” is because they operate in the physical world where actions have actual consequences. If a VC backed DTC brand messes up a landing page, they fix it in ten minutes and send an apologetic email. If a global FMCG giant messes up a product formulation or a supply chain rollout, they are looking at a nine figures disaster across 60 different legal jurisdictions.
What we might perceive as “slowness” is actually a highly evolved form of risk management. Nobody is ignoring digital, but they play their cards right. They go where their buyers are, and provide their buyers with what they need. Simple as that. Also, these companies manage global trade spend, and logistics networks that span continents. That is a level of technical and operational sophistication that a Shopify based skincare brand doesn’t even have a department for… nor experience, nor awareness off….
Rethinking about this sent me in a deep rabbit hole. It’s easy, in agency and consulting work, to get frustrated when a client isn’t moving at the pace you expected. But I think that frustration is usually a signal that we didn’t do enough homework upfront. If we spent more time at the beginning actually understanding how the business makes money, the distribution, the retail relationships, the customer, the real constraints, we’d pitch more realistic solutions and set better expectations.
Maybe it’s us showing up sometimes with the wrong context and then wonder why nothing sticks.
Personally, in the past few years I’ve been way more disciplined about this, especially in pre-sales, and it genuinely changes the quality of the conversation. This event was a good reminder of that.
Every time I deal with an inbound lead, an RFP, or a sales opportunity, I genuinely try to spend as much time as I can in discovery. And I don’t mean asking 2-3 questions and then present a credentials deck. In fact, I personally hate showing up at a first call with a potential new client partner with a deck. If anything, who cares what we do and who we worked with previously, if we don’t understand the problem at hand?
I mean actually trying to understand how the business makes money, how decisions get made, who has budget, who has opinions, and what the real constraints are.
Is there enough time to fully understand a business in a one or two week pre-sales process? No. But there are ways of approaching that window to set honest expectations, and more importantly, establish business model understanding as something that continues after the contract is signed!!
I can already hear some of you pushing back, lol. Since most of you all reading my newsletter work in digital analytics, your reaction might be: “Juliana, I’m doing some GA4+BigQuery or setting up server-side tracking, I don’t need to understand a company’s retail distribution strategy to do that.”
WELL, I actually think you do. Let me give you some scenarios…
Let’s say a retailer brings in an analytics team to build out their ecommerce attribution model. Months of work, measurement frameworks, KPIs definitions, choosing the right models, the whole shebang. However, nobody thought to mention that 70% of their actual sales happen in physical stores. This makes your work technically great, but commercially useless.
A subscription business wants help optimising their acquisition funnel. Conversion rate, CAC, channel performance, you build all of it properly. However, the real problem was churn, which nobody brought up because internally it was considered a product problem, not a marketing one. Two different teams, two different conversations, and the analytics work landed in the wrong one. Silos ftw.
Or, you invest heavily in tracking and audience building, run remarketing campaigns, and wonder why performance feels off. Turns out you’ve been targeting existing customers the whole time, because nobody in the discovery process ever got into how the business actually defines a new customer versus an existing one. The data was technically flowing…but flowing around the wrong definition of the problem.
This is why I say, the business model is THE strategy, everything else comes after, and the more time we spend away from the truth, the further we are from how the money is made, the less likely our work will matter… or produce meaningful results.
PS: This reminds me of a slide from one of my talks from 2023…
Other stuff I’m up to
I've partnered with Stape to create educational content about the commercial POV of server-side tracking. If you missed the announcement, this is the short version: there’s a lot of great technical content about server-side tracking in this industry, and almost nothing that explains what it actually means for marketing performance in plain language. That’s what we’re going to build together.
The first piece drops in April, and the third example above is exactly where we’re starting. It’s a problem I’ve been hearing consistently for the past year, and it has more layers than people realise. Most assume it’s a tracking issue. It’s a bit more complicated than that. It starts much earlier, with nobody in the business ever agreeing on what a new customer actually means. Everything downstream of that missing conversation is built on shaky ground, and we’re going to unpack the whole thing properly, and how can Stape help once the definition part is out of the way.
Next week I’m at Measurecamp Helsinki, where together with Matt Gershoff, we will combine 2 sessions in 1 (so approx 40’) to talk about Main Effects and Interactions. If you remember, I previously wrote about Creative Intelligence, and have already spoken at 5+ conferences across Europe about this topic.
Matt has a fabulous talk he did at Superweek on the mechanics of how I propose solving for measuring creative asset performance, and I quote Matt: I will come at it from thinking about an effective complexity lens. How maybe we should think a bit about expected regularities versus noise and how for many of the simple problems in A/B Testing land we can recast many of our analysis into partial F-tests - which can be loosely thought of as a complexity ratio that captures and tests for marginal returns to an increase in the effective complexity in our solutions. In fact even a simple A/B test can be see in this light.
If you are there too, make sure to join our session. If you aren’t but you’d love to watch it, I’m actually planning to stream it live on my Substack, so if you are a subscriber you will get an email when we go live from Helsinki :)
The latest episode of Standard Deviation Podcast, Season 5 is live and me and Simo had the pleasure to welcome Johan van de Werken as a guest. Johan walks us through the origin and vision of GA4 Data Form, an open-source project that transforms messy raw GA4 BigQuery exports into clean, analysis-ready tables with minimal SQL knowledge required. He talks about how it all evolved from a community contribution into a bootstrapped EU-based business, what the free vs. premium tiers offer, and where the roadmap is heading (Google Ads, Search Console, and beyond) » Make sure you check it out here.«
One last thing, I know a lot of my readers work in universities and academic institutions, and I genuinely appreciate the backlinks and referrals to my content.
Teaching is something I care about a lot. I’ve taught a course at IHM Business School, and next week I have the pleasure of giving a guest lecture in my brilliant friend, Lotta Holm, class at the Metropolia University of Applied Sciences, in Helsinki, which I’m SO LOOKING FORWARD TO.
If you’re building a curriculum around AI, machine learning, or commercial and business preparedness for AI, and you’re looking for someone to come in and make it a bit more real-world, I’d love to hear from you.
As always, thank you for subscribing to this newsletter. While I do not know everyone in my emailing list, I will take a moment to be vulnerable and say how privileged I feel that more than 1300 of you think that what I have to say matters, or find what I have to say useful.
I tend sometimes to have quite shitty self talk and feel less than, but every time I sit down and write, I get great comfort thinking there will be some of you out there that really appreciate what I am trying to do with this newsletter.
