From SEK 1 million to SEK 10 million a month: what to do when growth stalls #106

A warm welcome to a new episode of Zero to 100. With me as usual is Fredrik Ankarsköld. This week we've had some questions come in. I've also had the honor of taking a look under the hood of a company that's grown incredibly quickly in a year: from zero to 100,000 kronor, then some stagnation, and then from 100,000 to a million a month. Young guys, but now they can't quite get it working.

Why isn't the methodology working going forward? A million a month is just over twelve million a year. That's the first reasonable plateau they've reached, but there are more plateaus to talk about. How do we move a company forward once it's leveled out? Sometimes the level is twelve million a year, sometimes twenty. It can vary, but there are often plateaus you reach. How do you work your way beyond them? That's the question.

This feels like a really fun topic. I can imagine quite a few people have got stuck in one of the phases we'll talk about. There are different phases in different kinds of company building. In this case, young guys have started a company with a good idea in a well-established market. They want to make it more automated, scalable and cost-effective for the end customer. They have a fantastic product and a good website, so there are underlying factors that make scaling possible. I'd say the reason you manage to sell to your first, say, 10,000 customers is simple: Facebook or Google wants the best for you. Many people think it's a black hole and they just want your money.

But no, they want the best for you. Trust me. They want you to make money and will do everything in their power to help you find your customers. If your first 10,000 kronor of spend gives nothing back, you're put off the platform. Few people invest hundreds of thousands of kronor in a platform that delivers no return.

So you often get fairly quick ROI on your ads at the start. Facebook looks for the easiest customers first. The platform reads your website, information and creatives. Hopefully you've done good preparation: a solid product page, knowledge of the competitors and what works in their ad libraries. Then you do something equivalent or better. On that assumption, you find your first customers relatively cost-effectively. Meta groups the audiences and knows who the customer for this product is. You end up in the same audience as your competitors and hopefully stand out a little on price, with a better product or some unique quality.

It could be a unique feature, faster or free shipping, or something else that gets you chosen. At the start it feels like everything is going your way. Then you hit the famous wall, the plateau: now I have these 1,000, 2,000 or 5,000 customers a month. What do I do now? More budget doesn't always solve the problem. CPMs shoot up and it doesn't feel like there's much more to adjust on the website. There's always more to do, but we're talking incremental improvements. You've already worked on the big things, like a clearer button, purchasing journey or checkout. What's next?

Most people think they should scale up advertising and make it more efficient. But simply pouring in money can be like spraying it straight into a fire. You might get more traffic, but it's no longer as relevant. Remember what I said: Facebook wants the best for you and wants to find the easiest customers first. When you move beyond that easy customer group, let's call them early adopters, you're more on the periphery. That's where the brand starts to matter. Parameters other than just having a slight edge become important.

And as we've discussed before, awareness of the problem matters. The first audience that buys may already know they have a problem and be looking for a solution when your ad appears.

Then comes the audience that knows it has a problem but doesn't know what the solution might be. That takes more. Initially you can move forward with greater creative variety in your ads. Perhaps you've found a style that works, repeated it and only made small departures. Start with broader variety: carousel ads, GIF ads, ads leading to listicles, lead ads and better email flows. Perhaps much better creatives. Maybe you're moving from UGC towards [unclear in the Swedish source: employer brand content] or doing other things.

All this builds the brand more broadly. Instead of doing one thing really well, you start broadening the work. Together, the efforts can educate the audience that knows the problem but not the solution. They encounter the solution in several ways, at several touchpoints. It's classic Marketing 101: the first thing you learn about marketing at university. You need different points of contact, because otherwise the end customer isn't that easy to reach.

So the company you're looking at now has reached a million a month, and advertising isn't the problem? Not just advertising. Creative breadth is one of my strongest recommendations. I'm going to give them a whole battery of recommendations, so they can work on several things in parallel. The question isn't just how to take it from one to two or three million, but how to take it from one to ten million, relatively quickly and at a fast pace.

What's in that battery that will take the company from one to two and on towards ten million? Lots of things, but creative breadth is absolutely fundamental. We need more variety and more ads.

So ads are still an important factor. But it's not about spending more money on the same ads in Meta. How do we change the structure and build processes that let us bring in more ads? Great creative breadth should appeal to different audiences at the stage they're at. We move a bit more upper funnel, if we're using the traditional funnel. I hate the funnel and usually turn it upside down, but let's forget that shit. The point is to reach a broader audience. There are different ways to do that. Another thing is variety in platforms.

You might reach new audiences on other platforms, or the same people with more touchpoints. Again, classic marketing: you want to reach and educate the end customer in different ways.

But isn't there a risk? A company usually has limited time and limited resources in these phases. Someone might think it's not working perfectly on Facebook and therefore want to invest in YouTube, TikTok and Pinterest too. Surely that can be the wrong approach?

Yes, and that's why it's good to have someone who's made the journey several times and can help prioritize. Prioritization is among the hardest things a founder faces. We usually create a backlog in Asana and ask: Are these the most critical things for us to work on every week? Then we have a weekly meeting to follow up, so everyone is constantly working towards the same goal rather than pulling in different directions.

It's easy to set up, but clarity and communication are really important. Often communication is what breaks down when something fails. You get frustrated and don't know what the others are doing. If communication is clear, expectations are reasonable and everyone understands what takes time, usually nobody needs to tear their hair out. Then everyone works on their own tasks, doesn't disturb each other, grafts and scales. You said the first recommendation is breadth. How should a company go about getting it? It can feel abstract.

What do you actually do? There are several things. One is to call existing customers and ask how they found the product, what they like about it and whom they'd recommend it to. I love picking up the phone and calling customers.

How do you do that without being a stalker? It's simple; they've bought from you. You can say: Hi, you've bought from us at Såpa. I see you've shopped twice. Then you're not a stalker; they like hearing from you. I have a few questions. Help me and I'll help you. You can get a discount code, join the testing panel for the next product or something else. Want to be part of the journey? Some people are really busy and ask you to call later; others are happy to join in. Many share their experiences of the brand and the website.

Suddenly you learn there's a problem with the bottle you didn't know about. Or that the customer didn't understand at all how to use the product after delivery, so it's still sitting on the dishwasher.

Then you can address problems you didn't know existed. In other cases you get new angles. There's a great deal to learn from existing customers.

What questions do you usually ask in an interview like that? Problems are part of it. You want to find pain points in the customer journey: Was there anything that made you not buy, or almost abandon the purchase? Which competitors did you compare us with before choosing our products at Såpa? If we didn't have X or Z, what other product would you have bought?

If you hadn't bought from us, what would you have bought instead? How did you hear about us? Was there an ad or something else that stood out? I have a battery of questions, but I want the conversation to flow naturally. You only have a few minutes and can't ask everyone everything. It isn't a scientific, data-driven survey where a thousand people get exactly the same questions.

It's not an interrogation; it's qualitative insights that can improve the marketing. Then there's qualitative research into how other brands in the same niche work. Is someone doing something really well? Can we use their strategy in our approach? How do we refine the processes? Many improvement suggestions will be about not hiring lots of people straight away, but working more with AI and automating processes that are currently manual. That might be filling in Excel to track the numbers every day, week and month. Right now there's no structure; you look in Shopify. There, data can lag or be wrong.

It can also involve data you don't get from Shopify, such as advertising costs. Is the problem that Shopify lags or that you don't have enough data from the different platforms?

It can be both. Perhaps not all the data comes in at the same time, you're missing data from different platforms, or your own KPIs for measuring their effectiveness. What kind of data are we looking at?

It might also be that three employees at the company are looking at three different types of data. Perhaps everyone looks at revenue, returning customers and conversion rate. I'm guessing, because that's on the dashboard when you log in. But then what? What products are we selling, when are we selling them, and which platforms do the sales come from?

Some things are in the platform, others aren't. How much we've spent on each product is a really interesting metric, I think. Also the return rate.

Yes, the return rate. If I'm going to optimize something, I want to know how much you spend on the same product each day, week and month, to see whether the product is profitable.

So how do we collect the data from the platforms where we advertise? If it's Meta and Google, I want the naming to be correct so I can track it. Usually you drive traffic to a product page. If you don't, I have another, underlying problem. A tip from the coach: Drive traffic to a product page and do it well. How much money have you spent driving traffic there? What does the product return in revenue and ROI? Fundamentally it's simple mathematics. All sales beyond the product you've driven traffic to, we call add-on. You have a few hero products that …

… you drive traffic to and hopefully generate sales on. Then you have add-on products on the website that hopefully mean you make money. Those are different categories we divide them into. How do you think about stores that have showstoppers? In a shop window there might be garments that are very colorful and eye-catching.

Not because they're the biggest sellers, but because they're meant to stand out as someone passes and draw the person into the shop. Once inside, they might still buy …

… a black bag or a black tank top. I know, and that's a typically old-school way to work in e-commerce. When you advertise on Google or Meta, you're making an ad for a product. If you drive traffic to it, people are unlikely to click on to the next product unless they first put the first one in their basket.

You primarily have to work on the product you're advertising. That [unclear in the Swedish source: peaking] or showstopper method can work in physical retail. There's a broader range there and you can get people to walk through a maze and pick things up along the way. That maze doesn't exist online in the same way. People go from an ad to a product page and back to the ad, if they found it on Meta. It's only an X away, as I usually say. One click and they're out of the store. They'll disappear if you don't get them to buy the product you're advertising.

What ROAS do you think they should have? If you sell makeup, perhaps you don't need as high a ROAS on the first purchase, because you want the customer to return and buy the same product again.

No, and you don't really need that with the cleaning products at Såpa either. But I still damn well fight for it to be profitable from the first purchase. If it isn't …

… competitors can also take the customers at the bottom of the funnel. People aren't as loyal to a brand as we might like to think. They might buy makeup once, twice, three or seventeen times and switch brands the eighteenth time, because someone else has better ads, prices, products or a better launch. People don't always fundamentally have the loyalty we think exists. We're not as loyal as the statistics can make it seem.

Being profitable on the very first purchase is my strongest advice. If you're not, you need a substantiated calculation showing you can scale profitably. We talked about collecting data from different platforms. Do you use any particular software for that?

Not really; I use Excel. I collect everything using Supermetrics. Admittedly that's software, a connector that pulls data from the platforms into Excel.

So it pulls from Meta, Snapchat and the others? Yes, but you can also do it manually. We automate with Supermetrics, but it's not essential. It's a relatively cheap platform where you pay per API call to retrieve data. But sitting looking at data also has risks. You can draw conclusions about things that aren't connected, such as more people drowning in summer because lots of people eat ice cream.

That's a story I've heard. Is it because of eating ice cream, or because people swim more in summer? It's easy to look at a data point and think it's the cause, even though it may not explain the result at all.

It's about correlation and causation. I understand what you mean. But what's important in this context is that we know roughly what we want to measure, measure what we want answers to, and don't measure too much. Here we're measuring how much money we spend on product X and how much of product X we sell. That's relevant data. It would be different if we looked at what we spend on product X and what we sell of product Y.

Do you separate returning customers and new customer acquisition? Always. That's really relevant, good point. You have to distinguish them. Otherwise you might just be buying your old customers over and over again. There's nothing wrong with that as long as you know that's what you're doing, because you'll spend money on returning customers too. But on average it's five times easier to dig where you stand.

Meaning: five times easier to reactivate an old customer than acquire a new one. That tends to make growing companies, like Såpa, more efficient over time. We see that in the P&L, provided everything else holds up and new customer acquisition doesn't shoot up. It can develop like a hockey stick: initially it's really cheap. Remember the beginning, when Facebook takes the easiest customers first. Then it gets more expensive, even more expensive and eventually really expensive.

We want to avoid new customer acquisition becoming really expensive. That's why we must constantly improve at acquiring new customers and measure it as a clear KPI. So how do you split the campaigns? You have the customer acquisition cost and Facebook likes optimizing for purchases. If we take Juniper, which has both soaps and bedding sets …

Duvet cover sets, yes. Then Facebook would actually want to optimize for the soaps. Is it even soap, or perfume? What do they have? Yes, hand soap.

So it doesn't skew that way, since the customer acquisition cost is lower for soap. You can instruct Facebook on what the platform should do. If you want as many purchases, meaning transactions, as possible, a large share of the budget will …

… go to the hand soap. If instead you select total order value, more will go to the bedding. Since they're two such separate products, I recommend moving them into different campaigns. One can target the fragrance line, the other the bedding. They can have different repurchase patterns. It's almost two different companies within the same company, with different products. At Såpa we have the deck soap in its own campaign. That's something you buy once a year, maybe twice if you have a boat or something similar. Detergent, on the other hand, you might buy every month or every other month, or at least …

… every third month. There's a completely different regularity there. It's important to split up the work so Facebook gets a chance to bid on both the fragrance products and the bedding in this example.

What had the company that got stuck at a million a month done wrong? There are lots of things, but it's a very young company. There's nothing wrong with that in itself. They've come to me with the thinking: We're looking for an investor who wants to help us take the next step. We need strategic advice. We can graft, as long as you help us review this.

There's a world of opportunity when someone has made the journey before. I don't need many minutes to see things that can have a big impact in that small company.

The alternative is for them to make the mistakes themselves, learn and let another year pass. We've seen companies stagnate, not just for a year but four years running. Eventually you might shut down because you're fed up. As an entrepreneur you want to maintain momentum as much as possible. When things are going well and you have a tailwind: keep pushing, so the tailwind continues. But what had they done wrong?

Above all, they've done roughly the same thing from October–November, when things started taking off, until today. Meanwhile advertising gets more expensive and the season changes. November–December is a hotter season than January–February. Now we're heading towards summer and that's different again.

Is it a year-round product or seasonal? It's a year-round product. But online shopping still has considerable seasonal variation. We know many people shop online during certain months.

Even a painting, TV, piece of furniture or other year-round product has peaks in certain months. You need a basic understanding of how traffic moves in your segment. They don't have that. I ask a simple question: What do search volumes look like in your segment? They don't know. Then we have to find out. If we say the product were camera equipment, what does the search volume for camera equipment look like month by month?

And compared with the previous year: Is the trend growing or shrinking? We need to know that too before looking at the reasons for stagnation. Do you go into Google Trends, or how do you look at the data?

Initially I look at Google Trends. Then there are good tools in Semrush and Ahrefs. Agencies often have access to tools others don't. You might not initially pay ten or twenty thousand kronor just to get certain data, but most agencies already have access to some of these tools.

There we can see search volumes and which websites are growing or declining in popularity. Has Google made an update, a core update, during the period? Could that explain a change, or has something else underlying happened?

What else had they done wrong? They've done a great deal right. But they've scaled from an apartment to a slightly larger apartment and then to a space of just over 300 square meters. My recommendation is that 3,000 square meters may be needed if we're to hit the budget for this year and, above all, next year. It's about perhaps not daring to invest in what you know will cost money.

Is it a retail space? There is a risk in tying yourself down, after all. These are production premises. Okay, so they have their own production, like Såpa, but a different type of product.

It's an entirely different category and industry, without the same repurchase patterns. The product doesn't solve a problem in the same way; it's more about lifestyle. They have an angle they've worked on a lot on Meta, because they see it working for others. Now it's about looking up and building the brand. Those who've bought may have bought the product, not the brand. I'd venture to say they haven't cared about the brand at all. You mentioned Juniper. In that case I think people care a lot about the brand. They've been good at PR and marketing across many channels, being visible on LinkedIn and telling Matilda's story. Many elements …

… are woven together and make me buy bedding from Juniper. That's different from buying, say, camera equipment from this website. They don't sell camera equipment, but I'm using that example to avoid revealing the company. Should they broaden the product range too, or is it the ads that need the most work? If they can't increase sales, why should they increase production?

They will succeed in increasing sales; I'm completely convinced of that. They've tested a little abroad and sell a lot in Sweden. I think they can dig where they stand and sell enormously more here. Partly through small improvements here and there, and partly through the big baseball swings: simplifying checkout, broadening and varying the creative work, and building a much stronger brand. That has an effect after three weeks and three months, and after three years it has an enormous effect.

It's the same as when we took over the work with Såpa. Instead of just trying to get a purchase here and now, we built the brand from the start. Today, three and a half years later, according to these [unclear in the Swedish source: Genpop] research companies, we have brand awareness among 18–28 percent of the total population.

Even if you've never received a single ad from Såpa, you might know the brand. How do you build that from the start? Look at engagement with the ads, build organically and in a community. Many things build a brand today. It's definitely not just about color, form and what the logo looks like, as many seem to think. A lot lies in communication: making it engaging, educational and inspiring. How have you done it? I'm curious.

We've brought the founders along and have a lot of instructive, educational and entertaining content in the ads. When someone signs up, they don't receive a single sales email. Instead, the emails are about education, inspiration, entertainment and a little behind the scenes, things not everyone emails about. Then you might want to open our emails more often than others in your inbox and become more drawn to the product. People follow people, not companies. That's why we want a front person people want to follow.

In every company I've been involved in, having a clear person fronting the company has been very successful. But the small company we're talking about doesn't have a front person.

Can it be an ambassador, or does it need to be a founder? It can be an ambassador, but a founder is much better. It feels genuine if the problem comes from me or someone close to me and I want to solve it. Someone in my family became ill, my mother became ill, and I want to solve the problem. It's close to my heart and becomes a natural step for me.

I can explain it. Hopefully I've solved the problem for myself or my family and now want to share the solution with others. It becomes genuine and honest. Transparency can get you quite far.

Let's take an example: a company selling gym products, like Joel at Gymkompaniet. If he wants to build more community and engagement, how would he do it?

That one's not actually so difficult. I understand Joel hates CRM and doesn't send many emails. But most people buying a home gym want tips and tricks. I've shopped with him myself and built a small home gym. I want to know more about others in the community and which exercises I can do with the equipment. Of course you can assume everyone takes the time to educate themselves, but it's helpful to get an email: Fedja, I know you've bought dumbbells. Here are three tips for using them. Try these five exercises, seventeen things or a program I've made. You can autogenerate that …

… with AI. Dumbbells are probably one of his best-selling products. Make a program, and you create the engagement we're talking about and a greater incentive to return. Another way is what he's already doing: fronting things everywhere, appearing onstage, creating relationships and building networks. That makes me want to shop at Gymkompaniet because I like Joel and want to support him. He's in industry media and articles. It becomes the little company against the big giant. I want to support the little company where Joel is at the front.

It was fun that he listened to our podcast and wrote a LinkedIn post about going out and testing the products. He'd been both at the warehouse and with customers when equipment was being assembled. I thought what he wrote was interesting.

He'd noticed that we e-commerce retailers often want to streamline things and make packages as small as possible. One insight was that it's sometimes better to have certain things preassembled. If we streamline so much that everything is reduced to its smallest components, the equipment becomes really difficult to assemble. I buy that reasoning.

It's a [unclear in the Swedish source: invisible]. Yes, but as e-commerce retailers we know we want to get the product out and make it usable as quickly as possible. If people use the product, they're more likely to recommend it and return.

And fewer returns. Lots of people buy a product, put it on the shelf and think they'll deal with it later. Been there, done that. I received a walking pad from Gymkompaniet last winter so I could walk while taking meetings. It was really convenient and already preassembled. I just had to take it out of the package, slide it under the desk, pull up the section and press a button, and it was running. Really helpful that it was so easy.

If it had arrived in lots of pieces, it would probably have sat there for several months before I dealt with it. You always prioritize other things before yourself: perhaps the children, your wife, the house and the company. Then you remember: Oh right, the walking pad I bought.

That sounds like a good insight. I haven't read the post, but it's relevant. Don't over-optimize logistics so the product has to arrive in a thousand little pieces. We're not running IKEA here.

Västerbottenssåpa also has a Facebook community. How have you created engagement there? Starting a Facebook group and making it work are two different things.

We've done this for several brands. When we worked with Switch Nails, they didn't believe in a Facebook community at all. One person on the team started the group and then it exploded. If you fundamentally have engaging marketing and a product that engages people and solves an everyday problem, starting a community is fairly easy. Then you have to make sure to [unclear in the Swedish source: spika det här communityt].

How do you do that? Invite them behind the scenes. Give them first dibs on new products or reviewing new releases. Show what's happening at the warehouse and what challenges arise. Let them help solve the challenges and shape new products through questions and polls.

Build it together. Then you want to engage people so they answer each other. It doesn't always have to be you, as the person running the group, who holds the authority. Someone asks a question and ten others answer, a bit like a Reddit thread. That's the goal you're aiming for.

How do you get people to join? Before they're in, they don't know everything they'll get. We usually include it in the email flows. If you've acquired a few thousand customers, you also have a few thousand email addresses. Then you can inform them: I know you're a customer of ours. Did you know we have this community of like-minded people? Click here. Facebook has actually worked well for our communities.

We haven't made groups work as well on Instagram. We've tried Discord and once had a Slack channel, but it became so expensive. We thought it could pay off because a customer is worth [amount missing from source]. But no, it was a terrible idea. There are different kinds of communities.

WhatsApp is another option right now. Facebook groups have worked naturally for the groups we've tested, where the audience is often women over about 25. It might work for men too. I can imagine it working even better with a niche product requiring a lot of information gathering or knowledge sharing. But here we have two successful examples.

Also look at groups you know yourself. If you listen to the podcast, hopefully you've got a few good names from us. There are several ways to create engagement. Returning to the original question: A brand can reach plateaus at different levels. Then you need to broaden the work beyond the audience we talked about.

When moving beyond the audience, above all you may need to reach the same people through different channels. You might need a TikTok account and TikTok ads, or a Pinterest account with ads, to give people more points of contact with your brand. That doesn't mean running the same ad in every format. But you need to broaden the channels so they have other touchpoints. That increases credibility.

It could be radio, which we've run with Såpa, or TV, which we're testing now. Admittedly it's IPTV, for example TV4 Play. It's working bloody well, we think. We'll evaluate soon. Many such things build the brand over time: I heard them on the radio, now I see them on Facebook and then on TV. That gives credibility. If I don't trust you, I won't buy from you. E-commerce is very much about trust. You want instant trust built in, so the customer immediately trusts the brand when she arrives on the website. I don't shop with someone I don't trust, and neither do you. People rarely shop with a company they don't trust.

Different things increase trust. One is this broadening within the funnel. Many companies get stuck targeting only the bottom of the funnel, with hard performance marketing in Google Ads or Facebook Ads. That's the small group who know us, know they have a problem and how to solve it, and buy the product shown.

What happens with the rest of the population, those 99 percent who don't even know they have the problem? That's exactly what we're discussing: broaden, reach higher up the funnel, inform and work on the brand. When they're in the market to buy a product, you want to be top of mind. The only way is to be constantly visible everywhere. By everywhere, I mean wherever the customer is.

If the company turns over a million a month, perhaps there's no budget for TV4 and radio. Then you have to choose which channels to appear in initially. Såpa doesn't turn over a million a month, but over ten million. Some months we might reach twenty million and may increase even more.

Then you want to broaden, because perhaps you're also starting to reach the limit on a platform like Facebook. Although it reaches many people, it doesn't reach everyone all the time. You need to reach beyond Facebook Ads. Google is an even clearer example: There you only reach people actively searching, unless you've activated the push elements. People are generally fairly bad at that. We should almost have a whole episode on how to do Demand Gen well on Google.

Not 99 percent, but nine out of ten, meaning 90 percent, of the companies I look at fail with Demand Gen on Google. It's an easy addition for companies already succeeding with Meta, who should therefore be able to succeed there. We can do a separate episode on that. But can you briefly say what someone can start with? If I start mentioning it, we'll be here another hour. We'll have to draw a line under this episode. We've talked intensely about scaling brands through different plateaus, with loads of practical tips. It'll be exciting to see whether I get involved in …

… this brand. Then I can introduce it to the podcast listeners, so you can follow the journey from literally zero to a hundred. We can also invite the founders to join the podcast and tell their story.

That would be interesting. One last question: If someone feels they need to broaden out and reach another part of the funnel, how do they do it? How do they access more parts of the funnel?

It can happen in different ways. You can start earlier in the purchasing journey. Often people focus on those at the very bottom of the funnel, searching for the product. They might search for camera equipment in Stockholm or an SLR camera in Stockholm. Someone who doesn't know they need camera equipment might only know they're filming a video at a wedding this weekend or in six months.

Does the person need to buy their own equipment or hire a photographer? What are the pros and cons? That's where you need to educate them about the value of buying their own equipment. You can bring in a photographer who uses it and get better control over the result. It's a really bad example, but you understand what I mean.

You have listicles for Såpa and sometimes drive traffic to an article before the product page. Yes, that's a typical way to broaden the audience and let them encounter you in several places. If we get an article in Breakit, we might drive people interested in technology there, because we know it creates instant trust: Såpa gets written about a lot in Breakit, nice. Then they might buy for that reason. For people who read Svenska Dagbladet, we can drive traffic to an article about Swedes' laundry habits where Såpa is mentioned several times. That's a good article to send traffic to. Then they click through to Såpa.

How do you do it at Såpa? Not everyone knows, for example, that you can get soft soap on your skin without any danger, compared with ordinary chemicals or other detergents.

How do you educate people on why they need soft soap instead of other washing and cleaning products? Look in our ad library. There are hundreds of variations where we try to educate people so early in the process they don't even know they need a new detergent. They don't know about the warning triangles on the back of what they already use, or that they need to act if they get it on their skin, and especially if they get it in their eyes.

We have one type of ad for the consumer who isn't aware at all. Then we have other ads for someone who's just had a baby, moved to a new house or got their first student apartment. These are people at different stages of life. We call it occasion-based shopping.

We have similar information on the landing pages, with different images, audiences and copy. That way we respond to the problems and situations people face in their lives.

So you build different landing pages. Are they different product pages, or how do you ensure that an ad with a particular message leads to a page …

… with the same message? The same product can reach different audiences, so the pages can differ. Facebook is good today at reading both the page and the ad content and matching them to the right people, at least in theory. It works fairly well in practice too.

At Såpa today, we don't drive traffic to lots of different landing pages for the same product. We try to concentrate as much traffic as possible on the same product or product page. At Twistboxes, though, we're still experimenting. Fundamentally it's the same product, a roof box, but it meets needs for different audiences: surfers, skiers, families with children, people in cities and in the countryside.

People with small cars and with big cars. They have different problems we want to solve, and we do that with different landing pages. The first image addresses the audience in the ad. For a city person with limited space in the city center, the problem might be not having a large storage area for the roof box. Then you want a foldable roof box. Another audience is electric car owners. We know range anxiety is one of the first things people think about when buying an electric car: What if the battery runs out on the way, or the range isn't enough? We also know a roof box prevents the car from …

… going as far. This roof box is designed so you can go a little further. It's kinder to you and the environment and provides longer range. That's a practical example of an audience we can address. It can also overlap with other audiences: An electric car owner might live in a big city and also be a surfer.

Then you've addressed three touchpoints and three pain points for that audience. You drive traffic to the same or different landing pages for the same product, but address different problems.

It might take three ads like that before the person makes a purchase decision. One final tip for someone wanting to learn is to go into Facebook Ad Library and look at Såpa or Twistboxes. When I first started learning advertising, I watched and transcribed. Now you can transcribe with AI. Initially I did it by hand and analyzed what was shown in the image, what they said and in what order.

You can feed the transcripts into AI and ask what categories the ad covers. When we started selling for [unclear in the Swedish source: Core by D], we looked at hoodies. They explained what function a particular pocket served and that it had room for snacks. We classified that as functionality and translated the approach to our own product, which was something completely different.

We saw that we needed to talk about functionality during those seconds. That's how we improved at advertising, by seeing how others sold and analyzing the content. That's an insanely good tip: Go into other people's ad libraries, transcribe with AI and get compilations with ideas for what you can do going forward.

Benefits over features, as I usually say. Don't just look at the function, but at its value to the end customer. We can draw a line there, otherwise the episode will get very long if we talk ad strategies too. Thanks for this week, Fredrik. It was incredibly fun to see you again. See you next Thursday as usual, and please rate the podcast on Spotify.

Rate the podcast! Now Fredrik and I are properly underway. Subscribe on YouTube too. We publish all the episodes there, so even those of you who've only listened can see us. Subscribe, comment and send feedback, or do what Joel at Gymkompaniet does: Write a post on LinkedIn and debate or discuss.

We love that sort of thing. Take care. Goodbye! Bye!