Breaking down the ROI mystery: which creators actually drive sales, not just vanity metrics?

Стратегии инфлюенс-маркетинга (RU)
1

I’ve been in this space long enough to know that engagement metrics are a trap, but I realized recently that I’m still falling for them. I’d look at a creator’s post that got 50k likes, 2k comments, and think “wow, great engagement.” Then I’d check actual conversions, and it was basically nothing.

So I dug in and tried to understand: why is a post with massive engagement not converting? And more importantly, how do I identify creators who actually drive sales instead of just vanity metrics?

Here’s what I found:

High engagement doesn’t mean high intent. A post on a macro-influencer’s account with 50k likes might include 40k likes from people who just double-tap while scrolling, had no intention of buying, and won’t remember the post exists in 30 seconds. Whereas a micro-creator’s post with 2k likes might have 1.8k likes from people who actually clicked to the product page.

Engagement type matters way more than engagement volume. Likes are almost worthless. Comments are better—they indicate someone’s willing to interact meaningfully. Clicks to the product page are worth exponentially more. And comments asking about the product or pricing are worth the most.

Audience composition matters more than audience size. A creator with 500k followers where only 5% of the audience is in your target demographic is worse than a creator with 50k followers where 60% of the audience matches and has shown purchase intent.

So I switched my approach. Instead of looking at vanity metrics, I started tracking:

Micro-conversions: Did they click to the product page? Did they add to cart? Did they complete purchase? What was the purchase value?

Repeat behavior: Did that person who converted come back and buy again? (This is a killer metric—repeat purchase rate shows you which creators actually brought good customers, not just one-time impulse buyers.)

Content resonance by creator: Which specific creators’ content styles led to higher conversion rates, even if their follower counts were smaller? This data was shocking. Some of my highest-performing creators had less than 100k followers.

Once I had this data, everything shifted. I stopped allocating budget based on who had the biggest following. I started allocating based on who actually moved revenue. And the results have been way better—lower cost per acquisition, higher repeat purchase rate, and better LTV.

But here’s the part I’m still figuring out: how do you get good conversion and repeat purchase data without having deep analytics integration? If a creator doesn’t share detailed performance reports, or if you’re running campaigns across multiple platforms and attributing is messy, it’s hard to know for sure who actually drove the sales.

How are you tracking performance and ROI across influencers? And how do you handle the creators who won’t give you detailed data?

2

You’ve hit on the most important unlock in influencer ROI: moving from vanity metrics to actual business outcomes. This is where most brands get stuck, so I’m glad you’re asking about it directly.

Here’s my methodology for tracking influencer-driven revenue:

Setup:

  • Unique UTM codes for every creator campaign (utm_source=creator_name, utm_medium=influencer, etc.)
  • Unique discount codes per creator (if you offer them) that let you track at checkout
  • Pixel-based tracking on high-intent actions (add to cart, product view from creator’s link)

Measurement:

  • Track traffic source from creator link → product view → add to cart → purchase
  • Calculate conversion rate at each step
  • Cost per visit, cost per conversion, cost per dollar of revenue, and repeat purchase rate
  • Repeat purchase rate within 90 days is my killer metric

The data I prioritize:

  1. Cost per dollar of revenue (total paid to creator ÷ total revenue attributed) — this is your true ROAS
  2. Repeat purchase rate within 30/60/90 days — shows whether they brought quality customers
  3. Average order value of referred customers — some creators bring bargain hunters, some bring high-ticket buyers
  4. Time to conversion — how long between click and purchase? (Fast conversion = high intent)

For creators who won’t share data:
I have a simple principle: no detailed tracking agreement, no significant budget allocation. I’ll do small test campaigns ($500-1000) with them to see performance, but I don’t scale without data. This actually gives creators incentive to share data because they realize they can’t access larger budgets otherwise.

Once you have 2-3 quarters of data, you can segment creators by their actual ROAS and repeat purchase rate. You’ll probably find that 20% of creators drive 80% of your revenue. Allocate accordingly.

The biggest surprise I found: micro-creators often outperform macros on repeat purchase rate because their audience is tighter and more trusting.

3

Your insight about repeat purchase rate is exactly right—that’s the metric that separates wheat from chaff in influencer ROI. A creator who brings one-time impulse buyers is low quality. A creator who brings customers with 2-3 repeat purchases is valuable.

Let me add a layer: you need to think about this as a cohort analysis problem, not just a per-campaign view.

The framework:

  • Track customers by their source creator and their cohort (week of click, month of click, etc.)
  • For each creator cohort, measure:
    • Day 1 revenue (immediate purchase)
    • Day 30 cumulative revenue (how much they’ve spent after 30 days)
    • Day 60 and 90 cumulative revenue
    • Customer lifetime value projection (usually estimated from 90-day data)

Using this framework changes allocation:
Creator A brings 100 customers, day 1 revenue $5k, day 90 revenue $12k. CLV = $120.
Creator B brings 100 customers, day 1 revenue $6k, day 90 revenue $7k. CLV = $70.

Creator A looks worse on immediate results, but the repeat purchase is so much better that over time, Creator A is vastly more valuable. If you allocate based on day 1 revenue, you’re making a big mistake.

For creators who won’t share data: you have to create attribution yourself. Use unique landing pages, unique discount codes, or pixel-based tracking on add-to-cart. The creators don’t have to cooperate; you just need to track what happens on your end.

The harder question: how are you currently separating influencer-sourced revenue from organic or other channel revenue? If your analytics isn’t clean on that, you’re flying blind.

4

I’m starting to position tracking and performance data as a core part of my creator relationships now, because I’m realizing that creators who share performance data openly are usually the ones who actually perform well. It’s a signal.

What I’m seeing: creators confidently say “let’s track it and see” are usually experienced, professional, and know they deliver results. Creators who resist tracking often know their engagement is inflated or their audience isn’t a good fit.

So I use the tracking request almost as a vetting step. If a creator is hesitant about detailed analytics, I’m less interested in partnering them with my brand clients. If they’re excited to see their performance data because they know they’ll look good, that’s a creator I want to work with.

I’m also helping creators think about their own data. Teaching them to look at their own performance patterns helps them understand their value better and advocate for themselves more effectively. Win-win.

5

From the creator side, I love when brands want to track performance because it aligns incentives. I know my content converts, so transparent data makes me look good. And honestly, it helps me understand my own audience better too.

What I notice: brands that track and share performance data back with me are also brands I’m more excited to create for in the future. Because I see the impact of my work, not just the payment. It’s motivating.

One thing worth noting: help creators understand what metrics matter to your business. Some creators think vanity metrics are what you care about, so they optimize for likes and comments instead of clicks and conversions. If you take time to explain “we actually care about people who click through and buy, not just engagement,” they’ll adjust their content strategy accordingly. That feedback loop makes everyone better.