The best influencer teams are the ones with the deepest creator relationships. That same strength is probably your biggest source of wasted spend right now.
This week: a decision tree for when to keep a partnership and when to walk away, plus how SNOCKS runs influencer marketing with a 75% hit rate:
Deep Dive | The 5-Step Influencer Decision Tree
I’ve observed this back then at yfood myself, as well as now across so many brands:
Most influencer teams are great at one thing: building real relationships with creators. Not just business relationships, but personal ones.
That depth is what makes influencer marketing work in the first place.
But it can also what turns it into wasted spend.
Because two things usually happen:
The people who are best at building relationships are rarely the most numbers-driven. So they often don't even notice a creator has stopped being profitable.
Or they do notice, and keep booking anyway. Loyalty wins over the P&L.
Neither should happen. Both happen constantly.
The result is the same: partnerships running on autopilot, burning spend, long after they stopped working.
To solve this problem, here are things that you need to implement in your company:
#1 Set a target for each influencer based on the partnership maturity.
Decide up front when a partnership can continue, based on how it has performed so far. The exact thresholds are yours. Here's a version I built with one of our customers off their Klar data:
1.1 Testing Phase:
Sign for a single post to keep the risk low.
👉🏽 If the post has an NC ROAS < 0.7, don't book the next one.
1.2 Validation Phase:
Book two more postings.
If combined NC ROAS of all 3 posts <1.25 and third post not > 1.5, don't continue the partnership.
1.3 Acquisition Phase:
Now you're optimising for new customers.
👉🏽 If last post NC ROAS >1.5 or combined NC ROAS of last three posts >1.25 and last post is not < 0.75, book again.
If performance starts sliding, stretch the gap between posts from 30 days to 45 or 60.
1.4 Monetization Phase:
This is where brands that optimise for new customers get stuck. They can't bring themselves to drop big creators who no longer drive new customers but still drive repeat revenue.
Follow this:
👉🏽 If NC ROAS of last three posts <1.25 but ROAS >3, book again
👉🏽 If NC ROAS of last post three posts <1 but ROAS >3.5, book again
👉🏽 If NC ROAS of last post three posts <0.75 but ROAS >4, book again
1.5 Revalidation Phase:
If you cut a creator who once reached the Acquisition stage, come back to them after 6, 9 or 12 months.
👉🏽 If post has NC ROAS of >1.25, book again - but keep the longer gap between posts, unless NC Share is above 50%.
#2 Make data widely accessible
Influencer performance usually sits buried in a spreadsheet somewhere. It shouldn't.
You should make performance data easily accessible to everyone. In Klar, you can also easily analyse the performance of a partnership based on how frequently you have worked with that person, making it easy to judge the performance based on partnership maturity.
#3 Build safeguards and support system
Ideally, this should be enough. But in reality the influencer team might struggle to follow these rules initially since they are likely not the most data-driven people. Thus, designate someone in the team who is comfortable with that and supports them in making the right decision.
Building relationships is still the foundation of your influencer success.
So this is never something that you should remove.
But you need to establish a framework and support system for your team that leads to the best possible monetary results for your company in the long term.
Masterclass Insights | How SNOCKS runs Influencer Marketing at scale
On a related note: A while ago I sat down with Jana Suckfüll for an eCom Unity Masterclass on Influencer Marketing at SNOCKS.
Jana helped builed influencer marketing at yfood from the ground up, and, until recently, she ran it at the SNOCKS Group. 12 people for DACH alone, each managing 70 to 90 partnerships per month.
She walked through what it actually looks like at that scale: how to spot creators that move revenue, what a good negotiation looks like, how SNOCKS judges performance past the vanity metrics, and how one-off collabs turn into long-term partnerships.
👉 Watch the full session here: Influencer Masterclass Recording.
Spoiler, these insights stood out to me:
A 70-75% hit rate is possible. SNOCKS books creators with a roughly 70-75% success rate. At yfood we were happy with 30%. The difference is process, not luck.
Soft signals beat the stats. Before any outreach, the team watches a profile for 2-3 days: the quality of the comments, how the creator replies, how the community talks to each other. Regional share, growth rate and past brand collabs come after that.
Negotiate harder than feels comfortable. First offers can often be cut by half, sometimes two thirds. SNOCKS works to a target CPM of €25-30 on net reach. If a post underdelivers by more than 30%, that triggers a renegotiation. And they'd rather take an extra deliverable than a lower price.
One email, everything in it. The first message already asks for an offer on the exact format plus the account insights. Products always get tested before a deal.
The reminder matters as much as the main post. It has to be a full spoken story - link, code, CTA, the whole sequence.
Codes that can't leak. Every code is individual, only live for 24 hours after posting, and only activates on a link click. No more coupon-site leakage, and much cleaner tracking.
You can't scale influencer like Meta. Start at €5-10k a month across 10-15 creators. Expect a quarter to flop, a quarter to be rockets, and the rest to earn a second or third booking.
We also got into briefings, CPM math for Twitch and YouTube, influencer events, and how SNOCKS uses AI in their workflow.
P.S. Jana just started her own influencer marketing consulting. If you're building this function in-house, she might be worth a conversation: janasuckfuell.com
This Sunday | Free webinar on scaling to €300k days in Q4
This Sunday (23.08, 11:00), our partner Nico Frank and the ECOZA team are running a free live webinar on Q4 and how to hit €300k days profitably. They walk through the full playbook - the pre-sale offer logic, Meta and Google campaign structure for cheaper CACs, the Klaviyo flows that bring new customers back within 60 days.
If Q4 planning is already on your desk, worth the hour.
→ Register here for free (German session)
Community Events | Cooking in Munich, Bingo in Cologne, …
🧑🏼🍳🍋 Italian Cooking Night | Munich, 16.09.26: Pasta in the making, glasses filled, and some of the best people in D2C right there with you.
→ Cook with us🎡 Jägermeister x Bingo Rave | Cologne, 23.09.26: Together with Jägermeister, we're hosting a Bingo Rave on night one of DMEXCO. Come for the eCom crowd, stay for whatever a Bingo Rave turns out to be.
→ Get one of the spots💬 eCom Unity Unplugged | Berlin, 07.10.26: an elite circle of 50 top D2C brands with >20m in revenue, matched for exclusive 1:1 networking based on their interests and expertise
→ Apply for Unplugged Vol. 6
Thanks for reading.
I’ve spent the last decade figuring out what’s actually working in eCommerce, and what it’s really worth in profit.
Coming from building an 8-figure and a 9-figure eCom brand myself, plus now working with thousands of brands at Klar. And sharing the insights, frameworks, and hard-earned lessons that help you build a better and more profitable eCom business in this newsletter.
If there's something you'd like me to unpack next, just reply. I read every response.
Have a great week.
Max ✌️
PS. If you missed any past issues, you can catch up on all of them here.
Not part of the eCom Unity yet? (Brands only)
Apply below to directly exchange with hundreds of other eCom decision-makers from the greatest brands such as Kapten & Son, Purelei, KoRo, Junglück, Blackroll, and many more in our closed Slack community.
Brands only. High-class content & events. 100% free.
