Ad trends 2026: lessons from 110 million NOK in ad spend

Based on more than 110 million NOK in ad spend

What you are about to read is the result of dozens of hours of discussion across the entire Curve team, plus more than 110 million NOK in ad spend for our clients in 2025. Not to mention several hundred thousand NOK spent on mentors, courses and training from the best international environments.

Norway still runs 2 to 3 years behind the bigger markets in performance marketing, but the road ahead has never been clearer. In the more sophisticated markets the shift happened about a year ago: from media buying, meaning small tactical moves inside the ad account, to creative strategy. Content became the new targeting, and the advertisers with the most content took all the sales.

The numbers make it obvious. Norway has 5.5 million inhabitants, while other markets fight for the attention of 347 million people. There are 33 million companies in the US and 700,000 in Norway. Advertisers over there have had to break through with new strategies, and there is plenty to take from them, as long as you look past how wild some of the ads can be.

We realized this in the summer of 2023 when we founded Curve.no. Today we spend around 200,000 NOK per day on ads across our clients. With us, Collagen for Hund has grown from 1,500 NOK in daily ad spend to 20,000 NOK while increasing profitability. Propr has grown its bottom line by 252%.

The paid social trends that will matter most going forward boil down to three things: creative volume, creative diversity and creative strategy. The rest of the report shows you what that means in practice.

The 5 most common mistakes we see in ad accounts

1. You test blindly without a strategy

Are you testing different concepts? Are you testing different angles? You need all of it to have predictability when scaling. It is part of a bigger strategy, and the foundation of a system that works.

2. No iteration, you do not do more of what works

Every time you have an ad that works, you want it multiplied by ten with many new variants. A small preview: You test an angle, for example "Quenches your thirst", "Tastes great" or "Healthiest on the market". If the angle works, you run it through the 12 concepts further down in the report.

3. You do not treat angles and concepts differently

An angle is the reason someone buys. A concept is how the ad communicates the angle.

Far too few know this distinction. It is valuable because it gives you an ocean of possibilities. Biggest of all: running winning angles through many different concepts. That easily multiplies your volume, with a far higher probability of success.

4. You do not know how the algorithm works in 2026

First of all, you need to understand the business model of the social platforms, especially Meta. We advertisers are the customer, but the product, which matters most, is the user. Create a bad user experience and the user leaves the platform. Then there are fewer users, and fewer who want to advertise there. That is bad business for Meta.

That is why the platform's number one focus is: keep the user on the platform, do not scare anyone away. It is also why LinkedIn does not like links in posts, you are sending users out of their platform. A secondary focus is creating sales for the advertiser, but the platforms know that is best achieved through a good user experience. The smoother a buying experience is, the more users stay, and the more advertisers get sales. It is all connected.

5. You do not know how targeting works in 2026

The content does all the targeting. If an ad says "how busy moms can make healthy food for the whole family", the ad gets shown to busy moms who care about healthy food for their family. We even run some ads for the agency that are completely broad, with no targeting. They talk about advertising for companies spending around 200,000 NOK per month on ads, and it surprised us: The ad converts well with highly qualified companies.

The algorithm knows what each of us prefers to see in the feed. Some people get almost only video ads because they stop more often, click more often and buy. That is good business for Meta. That is why you want images, videos and carousels, plus many different angles, concepts and twists on your ads, because there is no one size fits all.

If you do not have a large selection of ads, your profitability goes down. Here is why:

• You give the algorithm less to work with. Fewer ads means fewer data points and worse matching to user preferences.
• The algorithm has one goal: show what people like to see, which keeps them on the platform, but also gets them to buy from you.
• With few ads you paint the algorithm into a corner where it cannot deliver relevant ads to each user.
• With 10, 20 or 50 variants, the algorithm finds the one that matches the specific user. High relevance, low cost.

More variation = more relevance = lower ad cost = higher ROAS.

You let the algorithm optimize for you, but it cannot do that if it only has one card to play.

10 real audits we did of companies' advertising in 2025

We have been paid 5,500 NOK an hour to give away this information. Now we are giving it out for free. Below you get the main findings from 10 different analyses, with benchmarks, recommended strategies and bottlenecks, for both Meta and Google. All sensitive information has been removed, and it is not possible to tell which companies are involved. At the end we reveal the common denominator across all the analyses.

Feel free to watch the video where we break down all the analyses first: Video: walkthrough of all the audits

Audit 1: Info product, Norway

One of the most unique accounts we have ever gone through. Only about 100,000 NOK spent in the last 30 days, with 80% going to click and reach campaigns that nobody uses anymore, and that no longer work.

• Numbers (Meta): approx. 750 purchases, CPA approx. 150 NOK, reported ROAS approx. 10 (strong, but needs checking against backend numbers), frequency 4.43, CTR 0.56% (very low, the ads need a refresh).
• Creative: A mix of one person talking on video and stills of the target audience with a text box. The ads are easy to recognize, with strong brand recognition and natural images that look real. But: no creative diversity.
• Recommendation: Test UGC ads, start with a big test of angles, then run the winners through all our winning concepts (you find the full list in the scaling method further down). Iterate with tweaks and optimizations to squeeze all the juice out of every winning ad.
• Snapchat: Very messy, the best campaign is retargeting, and nothing is scaled. Too many optimizations for clicks and reach. Approx. 70,000 NOK in spend, approx. 400 purchases, CPA approx. 150 NOK.

Audit 2: Digital service, Norway

• Main problem (Meta): Left with 1 to 3 real winners that spend most of the budget and get most of the results. The algorithm now punishes low variation and lack of testing. You cannot scale one ad, you need a system that continuously produces new variants, angles and creatives. The setup stops right when it starts getting good.
• Numbers: Approx. 200,000 NOK in spend and approx. 3,000 leads. Last 30 days: 70 NOK CPL. Last 7 days: 100 NOK. Yesterday: 140 NOK. Bad momentum.
• Strategy: The message lands, now it needs to scale. With the right ads and systems they can spend 600,000 NOK per month at around 100 NOK CPL. UGC iteration system: take the existing winner and build 10 to 20 variants, like street interviews, testimonials, greenscreen reactions, humor and comparisons. Same message, new packaging. The weakest variants get phased out and new ones pushed in weekly.

Audit 3: Pet supplies, USA

• Main problem (Google): The campaign structure is very simple and does not allow scale. pMax does prospecting, nurturing and retargeting at the same time, which confuses the algorithm and skews the numbers. The creative, meaning what people actually see in the ad, can be optimized for higher CTR.
• Analysis: Search impression share is below 10% in all periods, so there is plenty of market share left to take. The account does not use all available assets: no sitelinks, no promotion extensions, no prices, no self-produced videos. Automatic asset generation is enabled, which makes A/B testing impossible.
• Strategy: Scale horizontally by giving winning product categories their own campaigns to control budget and bids. Maximize the number of assets, get custom videos made, turn off Google's automatic video generation, exclude brand from pMax and optimize the site for higher AOV.

Audit 4: Cosmetics, Norway

• Main problem (Meta): Few low-hanging fruits besides creatives. The main problem is low AOV, which can be solved with technical steps in the on-site customer journey and with angles in the ads. No deliberate testing of angles, concepts, personas or use cases, and that blocks scaling.
• Numbers (30 days): Approx. 7,000 USD spent, CPA approx. 25 USD, approx. 250 sales, ROAS approx. 1.5. Spread across 11 campaigns, far too fragmented.
• Creative: Video accounts for the majority of the content. With 12 concepts to run winners through, content volume can be multiplied by 12, often with a lift in results. Half the job is finding something that works, the other half is scaling it. They need creative volume (many ads), velocity (launch fast and test effectively) and diversity (ads that look different).

Audit 5: Anonymous, Norway

• Main problem (Meta): Lifetime budget is used, which gives Meta a max cap per campaign and the advertiser little control. All optimized accounts use daily budgets. The best campaign was shut off automatically because the lifetime budget ran out. Few evergreen campaigns, which is not sustainable.
• Numbers (30 days): Approx. 100,000 NOK in spend, approx. 1,400 leads, CPA approx. 80 NOK, CPM approx. 130 NOK.
• Creative: Image ads taken as stills from video, almost like a screenshot while the person is talking. We like that. Consumers love these images because they are real, alive and in-the-moment. But: no hook testing, no dedicated image ads, no testing of different headlines and ad copy.
• Google: The account gives a skewed picture of the real impact on the bottom line. The top keywords are brand searches, meaning leads that might have come anyway. Strategy: exclude brand from all non-brand campaigns, set all targeting to "Presence" instead of "Presence or interest", turn off automatic assets at account level, and set up custom conversion events with different values per conversion.

Audit 6: Health product, USA

• Main problem (Meta): Loads of campaigns optimized for clicks and reach. Sales-optimized prospecting campaigns consistently deliver better results, including far stronger email signups and other soft metrics. Engaged audience and Existing customers are not defined, so it is impossible to split results between customers and non-customers. No budget adjustments, missing creative velocity, angles and volume, and no clear iteration on winners.
• Analysis: Less than half the budget goes to performance. In reality only 36% goes to sales-optimized prospecting. The sales come from one single catalog ad, which has big scaling potential.
• Strategy: Increase ROAS from 2.5 to 3.0+ through better budget allocation and creative strategy. Reduce retargeting spend and move budget to sales-optimized prospecting. Remove prospecting toward soft goals. Set up a dedicated testing campaign, test 3 new concepts weekly and iterate on winners along the way.
• Google: 65% of the budget went to brand keywords in the last 7 days, and less than 10% of impression share is captured. The account uses only a small part of Google's placements, and Shopping is the lowest-hanging fruit. Strategy: exclude brand from everything that is not brand-focused, become more competitive, expand to new campaign types, and install scripts that monitor placements.

Audit 7: Home, Norway

• Main problem (Meta): 58% of the budget accounts for 33% of the return. Lifetime budget is used, which is inflexible. 13% of the budget goes to sales-optimized campaigns and 87% to traffic and reach. The setup requires minimal management, and results are dropping because of it. No activity in the account for more than 14 days.
• Creative: Running a 25% sale, but only using video ads for it. Poor rig for scaling, no testing campaign and no signs of scaling. Should increase ad volume and show different use cases in separate ads.
• Google: Weaker signals than possible, plenty of potential for higher CTR, no videos (a big part of Google's inventory is lost), little active use of negative keywords, and errors in the customer match lists: all have an audience of 0 and get the "Too small to serve" error.
• Strategy: Rework the signals and include only buyers. Rework the ads with USPs, more numbers, CTAs and more unique headlines and descriptions. Build a comprehensive negative keyword list and set up custom conversion events.

Audit 8: Health products, USA

A very strange and unique case: a big brand that has barely touched paid ads.

• Main problem (Meta): In the last 7 days, 104 USD was spent on sales campaigns, out of 1,608 USD total. That is 6.5% of the budget, and this share accounts for 100% of the sales from the ads. 40% of the budget goes to an engagement campaign with 0.18% CTR and 0 sales, it has no value. 5 campaigns and very fragmented, the budget is spread so thin that no ads can scale effectively.
• Creative: Runs the majority of the ads through influencers' pages as partnership ads. An effective approach to test is whitelisted ads, which often reduce CPA by approx. 20%. No testing campaign, meaning no content is being tested, a big low-hanging fruit. We want to see more "Us vs Them" ads.
• Google: The setup does not maximize the market in either Norway or the USA. Campaigns with up to 20 target countries, and as a rule it is never recommended to run a campaign that targets more than one country. Heavy focus on video that generates very few purchases, and brand searches that cost a lot per click.
• Strategy: Rework the strategy to exploit Google's inventory, cut down to one target country per campaign, scale down video, rework the ads from scratch with as many relevant elements as possible, and upload a customer match list.

Audit 9: Housing, Norway

• Main problem (Meta): The budget was scaled 301% through new campaigns without the account being ready for it. They are in peak season, but nothing is happening on testing to sell even more. Missing naming in ads manager shows there is no deliberate testing of angles, concepts, personas or use cases.
• Numbers: Approx. 230,000 NOK spent, ROAS approx. 2.15, approx. 50 orders, CPA approx. 4,500 NOK. ROAS is down 47% compared with the previous 30 days.
• The framework: Advertising has 4 parts. Understand how the algorithm works. Understand what it takes to scale: creative diversity. Understand which angles, concepts, use cases and personas you need to measure. And when you have something that works, iterate on the specific thing that made the ad work. The top ad here is a simple demo video of an employee showing the product in use, without any form of iteration.
• Google: The account often goes outside defined targeting because "Presence OR interest" is selected, so people who are merely interested in Norway see the ads. The ads themselves (copy and image) are not optimized, and CTR is hugely important on Google. 96%+ of the budget goes to search, and Standard Shopping was barely tested without getting a real chance. The naming makes the account hard to navigate.
• Strategy: Clean up the keyword lists with negative keywords, change location to "Presence" only, rewrite or replace low-performing assets, avoid pinned headlines where not absolutely necessary, and test Standard Shopping and feed-only pMax.

Audit 10: Tech gadget, USA

• Main problem (Google): Shopping should be the first priority to capture demand, since the format only shows on keywords with purchase intent. But Merchant Center is misconfigured, so the account gets 0 impressions from Shopping. Several suboptimal technical settings. Brand searches get 42.5% of the account's total spend, while the standard is around 5%. That is waste.
• Analysis: The account is limited by the bidding strategy and only spends approx. 1,200 USD per week. Asset optimization is enabled, which makes A/B testing impossible. No customer match list. Brand is not excluded from pMax, which in practice turns the whole account into one big brand campaign.
• Strategy: Fix the Merchant Center issues and launch Shopping in pMax plus Standard Shopping. Move from tCPA to tROAS. Turn off asset optimization. Upload a customer match list and exclude it from TOF campaigns. Use manual CPC or target impression share to push down costs and the share going to brand searches.

The common denominator: 13 bottlenecks across all the analyses

1. Missing creative diversity: A set-and-forget mentality with a limited selection of ads that worked in the past, without a system for continuously developing, testing and rotating new creatives. That creates ad fatigue where the audience sees the same ads too often.
2. Static testing and scaling: Widespread lack of systematic testing of concepts and angles. Even when a winner is identified, it is rarely iterated further with changes in message, expression or format.
3. Wrong campaign optimization: Campaigns are optimized for soft goals like clicks and reach instead of sales and leads. That attracts a low-quality audience that clicks but does not buy.
4. Inefficient budget allocation: A significant share of the funds goes to underperforming campaigns, while successful sales-driven campaigns are underfunded.
5. Unstructured account setup: Fragmented accounts with too many campaigns and ad sets create noise, make it hard for the algorithms to learn, and hard to track what actually works.
6. Skewed reporting and data: Excessive focus on brand searches gives a distorted picture of performance. Much of the traffic would most likely have come anyway, and ROAS becomes misleading.
7. Technical errors and gaps: Deficient tracking and wrong settings. In some cases automatic features are enabled without consent, which makes accurate A/B testing impossible.
8. Passive scaling: The budget is increased without an established system for testing creatives and audiences, which often gives declining ROAS against an ever broader and less relevant audience.
9. Incomplete platform utilization: Only a small part of the platform's potential is used, for example only search campaigns while high-converting formats like Shopping and Display are ignored.
10. Inefficient retargeting: Unnecessarily high frequency causes overspending on existing audiences and eats into the prospecting budget.
11. Missing targeting signals: Key data like customer lists is either not uploaded or has technical errors, which weakens the algorithm's ability to find profitable new customers.
12. Fragmented geographic targeting: Multiple countries in the same campaign gives the algorithm mixed data and makes it hard to optimize per market.
13. Lack of an overarching growth strategy: The biggest bottleneck. Many companies react to short-term swings instead of following a systematic, long-term plan for growth and profitability.

How to scale in 2026: from angle to volume

Creative strategy is perhaps the most important tool a modern marketer has. Without good ads it does not matter how good you are at buying media, the ads actually have to hit the audience. Meta's algorithm is extremely good and shows the ads that fit each individual best: some like simple image ads that get straight to the point, others prefer long, explanatory videos or carousels. If you want to succeed as an advertiser, you need something that hits every type of person. You cannot have five good ads and think you are done. One size never fits all, not in creative strategy either.

Step 1: Test angles, and think A/Z, not A/B

The first thing you need to find out is which angles resonate. Find ideas and angles from TikTok, Reddit and reviews, and make simple image ads for each angle. Make sure the angles are completely different: test A against Z, not A against B. A/B testing is about small differences like headlines, colors and buttons. A/Z testing is about testing things that are actually different enough to give clear answers.

Step 2: Once you have a winning angle, make more ads with different concepts

If you know, for example, that football coaches dream of free time on evenings and weekends, you can build many concepts around exactly that.

Step 3: Exploit all the different concepts

These are the winning concepts we use at Curve:

• Before / after
• Storytelling
• Greenscreen
• Tutorial
• Social proof mashup
• Reaction video
• Review
• Unboxing
• Listicle
• POV
• Humor / sketch
• Meme
• Trend
• ASMR
• Street interview
• Podcast format
• Replying to comments

You test the concepts after you have found a good angle. The order is:

1. Test angles
2. Test concepts
3. Test audiences

3 ad frameworks we have scaled on Snapchat

Example 1: UGC with a snap hook

• Part 1, hook: Have the creator record a video exactly as if she were sending a snap to a friend. This hook has an incredible ability to stop viewers from scrolling, measured in thumbstop, meaning the share of the audience that watches the first three seconds of the video.
• Part 2, problem section: If you have a large audience that is problem aware, a winning formula is letting the next part dig into the problem stopping them from reaching the dream shown in the hook. The audience knows they have the problem, but needs more education on why it occurs. The best messaging comes from deep understanding of the audience: you need to know their pain points, what they have tried before (failed solutions), and how they talk about it. Agitate deeper and deeper as far as it goes to increase the probability of conversion.
• Part 3, solution: This is where we strike. When you have drawn the bow far in the problem section without losing the viewer, you have to deliver well enough in the solution section to convince them the product will help with their pain point.

Example 2: Organic funny hook

• Part 1, hook: If possible, find an organic clip that can be tied to your product, a funny clip that makes people stop. Say you sell chairs: an organic clip of an angry man throwing a chair, followed by a video of the creator catching your brand's chair. Or a person tipping their chair and falling, followed by the creator lying on the ground next to your chair. Here, only creativity sets the limits.
• Part 2, main section: Connect the hook to a main section that pitches the product. In our example the phone got water on it and "ordered the product by itself", relatable when the phone lives its own life.
• Part 3, product showoff: You have hooked the right audience, use part 3 adapted to TOF viewers who do not know the product and need convincing to convert.
• Part 4, CTA: End with a simple call to action.

Example 3: Shocking statistic (UGC)

• Part 1, hook: Open with a shocking statistic with clear numbers. Example from Propr: We found out how much real estate agents pulled in through fees in 2024, between 6 and 8 billion NOK, and used that as the hook.
• Part 2, intro solution: Go straight into the solution after the statistic: "That is when I understood there was a lot to save, and I found Propr."
• Part 3, feature/highlight: Explain your product or service in more detail.

Bonus, how we film: We always ask "if this person filmed the video organically and posted it on their own channel, what would the setting look like?" Would they set up a camera in a nice studio, with perfect lighting and a nice shirt? Or would they film while walking outside in beautiful nature, in the car on the way home from the store, with a handheld phone straight to camera? The answer is usually the latter.

Want to see the templates in practice? Here you find 30 ad templates we have spent more than 10,000,000 NOK on profitably: Winning Ads Templates (Figma)