Common Mistakes in Ecommerce Data Analysis

Shopify analytics reports misleading data. Pay attention to these: 1/ AOV That number will be wrong if you process a lot of zero-dollar orders (samples, warranties, 100% discounts, etc.) Exclude those from your AOV analysis to have a better understanding of your real AOV. Plus, it only considers Gross Revenue - Discounts (no Shipping Collected). That matters if you want to test Free Shipping vs. Shipping Threshold -> AOVs will not be comparable. 2/ Conversion Rate Firstly, CR includes both returning and new customers. If your retention rate goes up in a particular month (a really lucrative promo or offer that pulls customers from your email flows), your CR will not be meaningful. Secondly, CR displayed in Shopify only relates to SITE conversions. Any orders processed through the live chat, drafts, or other integrated channels, will not be included. Take those conversions into account as they go through a customer journey too. Lastly, if you drive bot traffic to your site, that will skew your sessions and misstate CR. Ashburn is one of the biggest sources of bot traffic (Amazon servers). Filter random "small" cities with high session counts out of your total figure for a more accurate CR. 3/ Total Sales I mean, this one is outrageous -> this figure reports net sales + SALES TAX COLLECTED... No idea who at Shopify thought that it was a smart idea to calculate it that way AND place it on the main dashboard, but they haven't fixed it. Always go to Shopify reporting and exclude collected taxes to understand how much sales you actually generated. --- Understanding these shortcomings will help you make better financial decisions and look behind the data. Ideally, build your own financial model and paste Shopify data into it to generate more accurate metrics and insights. P.S. Need help with understanding financial metrics reported in Shopify and acting upon them? Let's talk! #ecommerce #d2c

We’ve worked with 8–9 figure brands and even the most advanced teams still fall for these 3 measurement mistakes: 1) Over-reliance on click-stream data It’s 2025 and some teams still judge performance based on what happened right before the conversion. First party data is valuable but often skews towards lower funnel channels until you light that data up with survey data (zero party data). The irony is that the channels that actually drive awareness (like influencer, podcast, or CTV) get cut because they don’t “convert.” If you’re trying to scale, this mindset will cap your growth. 2) Being dogmatic about attribution models Attribution isn’t religion. There is no single source of truth. Some teams swear by multi-touch. Others believe in incrementality or MMM. Some default to survey data. But real insight comes from triangulation, not loyalty to a model. Each method has blind spots: - MMM smooths over granularity - Survey data can have recall bias - Incrementality is hard to isolate and often misunderstood - Platform ROAS is self-reported and biased toward lower funnel The best teams don’t chase perfect data. They build frameworks to compare directional signals and make smarter, faster bets even when the data is messy. 3) Underinvesting in measurement infrastructure This one’s subtle and expensive but bad data costs way more than good tooling. If you're spending $1M+ per year on paid media, but don’t have a measurement strategy that scales with your channels, you're flying blind. Good measurement isn’t a “nice-to-have.” It’s part of the CAC equation. Whether it’s hiring a data lead, implementing survey tools like Fairing, or running incrementality tests, there’s a cost to seeing clearly. The biggest mistake of all? Thinking better measurement is optional. It’s not. In a world where CACs are rising, privacy is tightening, and attribution is only getting messier, the brands who understand why things work will be the ones who win. Not the loudest. Not the most funded. The ones who learn faster and reallocate smarter. P.S what other measurement mistakes do you commonly see?

You’re not growing ...because your analytics are lying Quietly. Repeatedly. So we found 266 reasons why. Not theory. Real mistakes in tracking, calculation, and interpretation. Use this to: - Audit reports - Train your team - Avoid wrong calls - Debug dashboards Covers 8 key areas: 1/ Product Performance 2/ Conversion Funnel 3/ Traffic Attribution 4/ Revenue Metrics 5/ Tech Accuracy 6/ Segmentation 7/ Retention 8/ Email Each mistake includes: - Category - Type - Description - Impact level - Prevalence - Checklist how to fix - How It Looks in Reality - Misleading Outcome - Related Metrics - Sources Built as a searchable, filterable Airtable database Perfect for audits, onboarding, or daily use. 𝗪𝗵𝗼 𝘄𝗮𝗻𝘁𝘀 𝗮 𝗹𝗶𝗻𝗸? 💎 Available publicly until April 10 only. No exceptions #analytics #marketing #ecommerce