When your influencer ROI tells different stories in different markets: how I finally aligned my analysis with reality

Кейсы и истории успеха
1

I want to talk about something that’s been bugging me for months, and I think it’s worth a detailed breakdown: how I was analyzing influencer campaign ROI completely wrong across my two biggest markets.

Here’s the setup: I manage influencer campaigns for a brand with significant presence in both Russia and the US. In 2024, I ran three major campaigns—two in Russia, one in the US. When I pulled the numbers, something felt off. In Russia, we had influencers with 50k followers driving 12% engagement and 4% conversion. In the US, influencers with 200k followers were hitting 8% engagement and 3.5% conversion. By pure metrics, the Russian campaigns looked better. But something in my gut said that wasn’t the real story.

So I started digging deeper. First, I mapped out the actual ROI framework I’d been using. It was simple: (Revenue - Campaign Spend) / Campaign Spend. Straightforward, but incomplete. Here’s why: that formula treats all spend the same way, and it completely ignores the anchor point. For the Russian campaigns, I’d spent $8k total across two influencers. For the US campaign, I’d spent $15k. The Russian side generated $32k in attributed revenue; the US side generated $48k.

By raw ROI percentage, Russia won: 4x ROI vs. 3.2x ROI for the US. But here’s what that metric didn’t tell me: the US campaign acquired 200 new customers. The Russian campaigns acquired 80. So which is actually better?

That’s when I realized I was comparing wrong. The question wasn’t “which campaign had the highest ROI percentage.” The question was “where did I get the best unit economics.” So I recalculated:

  • Russia: $48k revenue / 80 customers = $600 customer acquisition cost
  • US: $48k revenue / 200 customers = $240 customer acquisition cost

Sudden flip. The US was actually more efficient at acquiring new customers at scale. The Russian campaign was better at driving higher-value transactions.

Once I saw that split, everything made sense. In Russia, the influencers had deeper, more engaged audiences, so people who bought were already predisposed to trust the influencer—they bought bigger. In the US, the influencers had broader reach, so they were pulling in more first-time customers, but those customers bought smaller initial orders.

Now, what do you do with that insight? I realized I’d been asking the wrong strategic question. Instead of asking “which market should I pour more budget into,” I should ask “what’s my actual business goal?”

If we’re trying to grow customer count and build market share, the US strategy wins. More customers, more repeat-purchase potential, lower CAC. If we’re trying to maximize revenue per campaign dollar in a mature market with existing customer affinity, the Russian strategy wins. Deeper engagement, higher order value, loyalty signal.

But here’s the part that shocked the team: once I reframed the analysis that way, suddenly the influencers who’d looked “mediocre” by the old metrics looked strategic. One Russian influencer with 40k followers and 12% engagement had been given a budget cut because her absolute revenue number was lower than others. But when I looked at her CAC and repurchase rate among her audience, she was driving the highest lifetime value per customer.

The final piece: I created a different dashboard for each market. For Russia, I track: engagement rate, average order value, customer repeat rate (the real KPI). For the US, I track: reach, new customer acquisition cost, initial conversion rate (different real KPI). I stopped trying to compare them on the same axis.

Some honest stuff about the mistakes: I’d been doing this analysis for six months before someone on the team asked a basic question—“wait, are we trying to get bigger or get deeper?” That question should have been in my framework from day one. I was so focused on building a “universal” metric that worked across regions that I missed the fact that the regions actually had different business priorities.

I’d also been trusting the influencer platform’s native ROI calculations way too much. They’re built for easy reporting, not for strategic clarity. Every influencer platform will tell you ROI, but few will tell you CAC or customer lifetime value, and those are the numbers that actually matter.

Has anyone else run into this? Like, where your numbers-game metrics made one choice look right, but your real business metrics pointed somewhere completely different? And how do you decide which analysis to present to leadership without them thinking you’re just cherry-picking data to support a predetermined conclusion?

2

This is the analysis breakdown I’ve been waiting to see someone publish. You’ve identified something really important: raw ROI percentage is almost useless for strategic decision-making without unit economics context.

I want to add a layer here, though: have you built in cohort analysis? Like, did you track whether the customers acquired through the Russian influencers (higher AOV) had better retention than customers acquired through US influencers (higher CAC)? Because that actually changes the lifetime value calculation.

Also—and this is critical—did you account for attribution window? If a customer saw a Russian influencer post and didn’t buy for 30 days, does that still count as “influenced” by that campaign? Because I’ve seen analysis fail spectacularly when they don’t account for the fact that different audiences have different decision timelines.

One more: you said the Russian influencers had “existing customer affinity,” which I interpret as they were pulling from customers who already knew the brand. But did you actually measure that? New vs. returning? Because if the Russian campaign was just re-engaging existing customers, and the US campaign was doing true new-customer acquisition, then the comparison becomes even more stark.

The dashboard split you built—solid move. But I’d suggest going deeper: create attribution models per market. What’s your 30/60/90-day retention rate by influencer? That’s the metric that separates good campaigns from campaigns that just looked good on a spreadsheet.

3

Also, one tactical thing: when you present this to leadership, frame it as “we’ve moved from vanity metrics to business metrics.” Leadership loves that language because it makes them feel like you’ve had an insight rather than corrected a mistake. And honestly, you did have an insight—you just got there by first identifying that your framework was incomplete.

4

This is exactly what I’m terrified of doing wrong as we expand internationally. We’ve got decent performance in Russia, and now we’re thinking about launching in…let’s say three new markets. And I was genuinely going to use a single ROI calculation across all of them.

Your point about “what’s the business goal” is the insight I needed. For us, the goal in the mature market (Russia) is defend share and maximize margin. In new markets, it’s acquire customers. So the influencer strategy should look completely different, but we’d have run it the same way if I hadn’t read this.

One question: when you built out these separate dashboards, did you also change your influencer selection criteria? Like, in Russia you’d look for engagement + AOV signals, but in the US you’d look for reach + acquisition speed. Did that shift how you approach partnerships going forward?

Also—six months in before asking the right question. How did you catch that? Was it someone external pushing back, or did you just wake up one day and realize the framework was wrong?

5

This is solid analysis work. The CAC comparison reveals real business insight that raw ROI percentage obscured completely.

But I want to push on something: you’ve identified the difference in unit economics, but you haven’t answered the strategic question that actually matters for allocating capital next quarter. Given that Russia has 4:1 CAC ratio advantage (600 vs. 240), but the US offers >2x customer volume, where should the next $50k of budget go?

Here’s why that matters: if your customer lifetime value in the US is 3x the Russian LTV (plausible given they’re new customers with repeat potential), then the US might actually be the better long-term investment despite the higher CAC. Conversely, if Russian customers are buying $1k first orders and US customers are buying $150, then Russia wins despite lower customer count.

So my question: did you layer in customer lifetime value by market? That’s the metric that actually answers the capital allocation question.

Also—did you test whether the Russian influencers’ audiences were truly different, or whether you just picked different influencers? Like, if you’d given a Russian influencer with broad reach the same brief as your successful US influencers, would they have delivered different results?

The reason I ask is that cohort treatment effects matter for strategy. If the difference is truly regional audience behavior, you’d invest in different types of influencers per market. If the difference is just influencer selection, then you need to refine your vetting criteria globally.

Lastly—how are you stress-testing this against seasonality? These campaigns ran at one point in time. Did you control for whether Russian customers are inherently higher AOV, or whether you just ran your Russia campaign in a higher-purchase season?

6

Okay so reading this from the creator side is useful because it actually explains why some campaigns feel “bigger” from our perspective while delivering lower customer count. The Russian influencers in your case were probably working with audiences that were more premium, more engaged, maybe smaller but higher-intent. That’s a totally different gig than the US influencers who are fishing in a broader pond.

What I’m curious about: did you ever share this insight back with the influencers? Like, did the Russian creators get feedback that they were actually nailing the high-value customer piece, even if their absolute customer count was lower? Because honestly, that’s the kind of feedback that helps creators position themselves better.

Also—when you said one Russian influencer had a budget cut because revenue looked lower, but she was actually the best on LTV—what happened to her? Did you restore her budget? Because if you didn’t, and she didn’t know she was actually crushing it strategically, that’s a creator retention issue waiting to happen.

One more thing: did you ever present both dashboards to any of the influencers, or just keep them internal? I’m asking because if creators understood that you’re measuring them on engagement + AOV instead of just engagement + follower count, they might approach partnerships differently.

7

This is the kind of analysis that separates agencies that win repeat business from agencies that just do okay work. You’ve moved the conversation from “which campaign performed better” to “which campaign solved our business problem.”

From an agency perspective, here’s what I’d want to know: did you use this reframing to renegotiate terms with your influencers? Like, if the Russian influencers were driving higher AOV, that’s higher revenue per content piece, which actually justifies different compensation structures.

Also—did you build in a communication piece where you shared these insights with the influencers? Because in my experience, influencers who understand why they’re being chosen and measured the way they are become way more responsive to partnership terms and long-term planning.

One last thing: you said you built separate dashboards for each market. How are you presenting this to leadership without it looking like you’re hand-picking metrics to support your campaign? My guess is you’re framingit as “we’ve optimized measurement for market-specific goals,” but I’d love to know how you navigated that conversation, because I have to give leadership a single ROI number.