Problems with Email Attributed Revenue Metrics

Blindly aiming for a high Revenue Per Recipient (RPR) will damage the profitability and long-term thinking of your email marketing strategy - here's why 👇 RPR is held up as a 'benchmark' by many ESPs such as Klaviyo for success. Yet it's probably the most easily manipulated revenue in your account. Here are some reasons I ignore RPR: 🧲 Discounting Want a high RPR? Cannibalise your margin and throw everybody on the list 30% off. Great for RPR, usually bad for LTV. 🧲 Database size As you acquire more customers, your RPR will usually drop as your segmentation strategy expands to bring in more dormant customers for winback opportunities. This isn't necessarily a bad thing as you're looking to improve total revenue contribution from the channel. You can easily have a high RPR just by sending 1 campaign per month to the most engaged customers in your database. Good for the business? No - you miss out on a lot of opportunities to sell. 🧲 Frequency Sending more emails will lower your RPR. It may also lead to a lot of incremental revenue. Just like the last point, if you've got a chance to generate more sales, you're not going to refrain from messaging somebody if there's a chance of generating more profit for the sake of RPR. 🧲 Sacrificing content for sales-specific messaging Your audience needs more than just sales to keep them engaged - so make sure to give them that. Product education, tutorials, blog posts, competitions and other forms of content often don't yield massive sales but they are still important in nurturing your subscribers. If you abstain from distributing this type of content over email for the sake of keeping RPR high, you're eroding your ability to add value to subscribers and probably expediting churn through repetitive sales-focused messaging. ... My opinion? RPR is only valuable when isolated in a sales capacity to judge the success of a specific campaign and even then has limitations. It's better to gauge the impact of your overall profitability rather than be sucked down a rabbit hole trying to optimise a vanity metric. #emailmarketing #ecommerce #klaviyo

Picture this: A mid-size nonprofit spends $783K across channels and raises $3.1M. Pretty straightforward, right? But here's where it gets interesting – and by interesting, I mean messy. Let's look at what each channel claimed versus what actually happened - see top table. Take a moment with those numbers. The platforms are claiming $7.27M in revenue from activities that actually generated $3.1M. That's not a rounding error – it's someone thinking the pie is twice as big if make 8 slices instead of 4. Why the massive gap? It's not because platforms are deliberately lying (well, mostly). It's because they're all claiming credit for the same donations. It's like every player on a basketball team claiming they scored all 100 points because they touched the ball during each play. Here's a simple illustration. 1. Jane Donor gets your direct mail piece 2. Sees your Facebook ad 3. Gets your email 4. Googles your organization 5. Makes a $100 donation Now watch the magic: - Direct mail claims it (matchback analysis!) - Facebook says "we influenced that!" (view-through conversion!) - Email takes credit (last touch!) - Google Search says "they clicked our ad!" (last click!) Here's a visual. Suddenly, your $100 donation becomes $400 in channel reporting. The problem this creates isn't the revenue side or even reporting, it's the spending side. ROAS should, in theory, be your guiding star. But when every channel inflates its impact are you going to argue to increase spend in paid search, mail, email and ads? The person in charge of each channel surely will. The key takeaway is this: before asking for more money, fix the current budget by shifting when spend (too much in 4th quarter likely) and where (channel). This will ruffle some silo feathers but the upside is everyone's starting from a flawed baseline. The goal isn’t perfection; it’s progress—because when everyone's wrong, nobody needs to defend their turf, we're just trying to be less wrong together. The solution isn't to stop measuring channel performance. It's to think and do differently. Here's suggestions going from easier to harder (but doable for any charity - use GPT as your analyst). Overall Revenue - Ignore Channel --Year-to-date vs. last year - % change --Growth Score   --Total Cost to Raise a Dollar - ALL costs (staff, agency, media, etc.)/total revenue Incrementality Testing -turn something off, see total rev impact -change spend for Channel X for period (up or down), see change in total rev -geo-test (add channel to one market and don't do it in control market) Poor Man's MMM (Excel Version) --drop me note if want all formula or --ask GPT, it can be your MMM analyst in Excel.

Many brands are making decisions based on fake email data (and don't know it).   Here's a simple Klaviyo setting that 90% of brands ignore:   Excluding bot interactions from attribution.   I've been implementing this across all my clients lately, and it's a best practice that most brands just... don't do.   Here's the problem:   Bots are clicking your emails and SMS messages.   Not real customers. Bots.   This inflates your attribution and makes you think email is performing better than it actually is.   Maybe it's competitors checking out your campaigns. Maybe it's automated tools crawling your content. Maybe it's spam filters testing links.   The fix is dead simple:   Go into Klaviyo attribution settings → exclude bot interactions from clicked email and SMS metrics.   That's it.   Now your revenue attribution reflects actual human behavior.   The difference might be small (1-2% in most cases, some as high as 5%), but the principle is huge:   Make decisions based on real data, not inflated metrics.   Your attribution should show you where real customers are actually engaging and buying.   If you're using Klaviyo and haven't done this yet, do it today.   It takes 30 seconds and gives you cleaner, more accurate performance data.

eCom brands. Klaviyo says your email made up 30% of your sales. Reality? It just took the credit. Every week, I chat with founders of 7-8 figure eCom brands running email campaigns. >80% of them keep making THIS same attribution mistake, which inflates their email revenue and blinds them to what’s really driving sales: Taking Klaviyo’s attribution at face value. The hard truth is, Klaviyo isn’t gospel. It’s attribution isn’t 100% accurate. When it shows a 30% sales lift, it’s not necessarily 30% growth. It’s just assigned to growth that happened anyway. Take your Welcome Series, for example. It triggers when someone signs up—often through a pop-up on your site. If that customer makes a purchase later, Klaviyo automatically credits the Welcome Series, just because they: - Opened the email - Clicked through But in reality, the email didn’t make the sale. That customer was already planning to buy. Klaviyo just assigns credit because of the interaction. Take another case. Someone abandons their cart. Your flow fires off within 5 minutes. Most shoppers, though, are still on your site, deciding whether to hit “checkout.” If they buy, Klaviyo claims the win, even if your email had nothing to do with it. See the pattern? Obviously, every software wants to justify its price. And over-attributing keeps you paying. So, how do you find the real numbers? Run holdout tests. Divide your customers into two groups: - Group A gets the email (e.g., an abandoned cart reminder). - Group B doesn’t receive the email. Then, compare the purchase rates of both groups over time, say 7 days, to identify the actual lift in sales caused by the email sequence. This shows how many additional purchases the email genuinely drove, rather than attributing sales that would have happened anyway. Look, I am not saying email attribution tools are inherently bad. They’re helpful. But they’re also designed to make themselves look good. That’s why it’s on you to question the data. So track. Test. And get to the real lift. #klaviyo #emailmarketing #ecombrands

Your incrementality test isn’t broken. Your attribution model is. I hear this all the time: “Wait, how can our MMM say [CHANNEL] drove 25% of our revenue? Our dashboard shows only 12 conversions.” It’s a fair question. But it reveals a deeper misunderstanding. Marketing teams are using multi-touch attribution (MTA) to validate incrementality tests and MMMs. That’s like using a broken compass to check if your GPS is working. The whole point of incrementality testing is to show you what your attribution model can’t see. But I keep seeing teams reject test results because they don’t match their dashboards. “We want to pause our holdout test during back-to-school. Our conversions always drop that week.” But are they dropping because your marketing stopped working? Or because those impressions are building awareness that converts in September? What if that "conversion drop" is exactly when your competitors are winning mindshare? Only the test can tell you, but only if you let it run. “Can you adjust the MMM to weight our email channel higher? Our attribution model says it drives 30% of revenue.” But email attribution is notoriously inflated. People who are already planning to buy often check their email right before converting. That doesn’t mean email caused the purchase. It means email was part of the path, but not necessarily the driver. This is what incrementality is built to reveal. Your attribution model has blind spots. Your testing framework exists to find them. If you’re using MTA to validate incrementality, you’re not testing. You’re just confirming your existing biases with more expensive math. The uncomfortable truth? Sometimes the channels you think are working aren't. Sometimes the channels you think are broken are actually your best performers. Sometimes your “high-converting” days are just when already-convinced people finally buy. Incrementality testing isn’t supposed to match your dashboard. It’s supposed to show you what your dashboard is missing. Stop trying to make incrementality fit your attribution model. Start letting incrementality show you where your attribution model is wrong. Just use a vendor like Stella to make sure your MMMs and incrementality results are trustworthy.