The Complete eCommerce Analytics Stack: What Tools Work Together (And What's Redundant)

Your Shopify store has three analytics subscriptions running. None of them can tell you which product to put in this week’s email.

That’s the analytics stack problem in one sentence. Not too few tools. Too many tools doing overlapping things, with a critical gap nobody filled.

Most stacks get built sideways: GA4 because someone said you needed it. A session recorder when a product page had unexplained drop-off. A marketing attribution tool when paid spend crossed a threshold where guessing felt expensive. Each addition solved a real pain at the time. Now they all run simultaneously, and you still can’t answer the questions that drive your actual merchandising decisions.

Here’s how to think about what your analytics stack needs to do, where the overlaps hide, and what’s genuinely earning its place.

The four jobs an eCommerce analytics stack has to fill

Not all analytics tools do the same thing. The clearest way to audit your stack is to assign every tool a specific job. There are four distinct jobs.

Job 1: Marketing attribution. Where did customers come from? Which channels, campaigns, and ads drove traffic that converted? This layer connects spend to acquired customers.

Job 2: Session and UX behavior. What did visitors do on your site? Which pages caused drop-off? Where do people get stuck? Session recordings, heatmaps, and on-page funnel analysis live here.

Job 3: Product-level catalog performance. Which of your products converts? Which ones appear in abandoned carts constantly? Which product drives the most repeat purchases? Which category has the highest customer retention rate 90 days post-purchase? This is not about sessions. It’s about your catalog.

Job 4: Customer health and retention. Who is at risk of churning? Which segments buy most frequently? Which cohort generates the highest lifetime value, and what triggered that loyalty?

Most tools handle one or two of these jobs at reasonable depth. None do all four well. The redundancy problem starts when merchants add a tool hoping it will cover a job it was never designed for. The eCommerce KPIs that actually drive decisions map directly to these four jobs. If a tool in your current stack can’t produce at least one of those metrics without a manual export, it probably isn’t earning its cost.

Signs your current stack has the wrong shape

A few diagnostic questions that quickly reveal structural problems.

You’re getting different revenue numbers from three different tools. GA4 shows one number, Shopify shows another, your attribution tool shows a third. The fix isn’t reconciling them manually every week. It’s cutting the tools that don’t need to measure revenue independently.

Your team uses different reports depending on who’s asking. If marketing looks at the attribution tool, operations looks at Shopify, and the founder looks at something else entirely, you don’t have an analytics stack. You have three separate views of the same business that will never agree on the same number.

You can describe your traffic but not your products. You know your session count, your ad ROAS, and your email open rates. But ask which specific SKUs are dragging your conversion rate down, and the answer is “I’d have to pull a report.” That’s the product analytics gap in plain sight.

You added a tool to solve a specific problem and never removed it. A session recorder installed during a redesign project. A custom dashboard built for a board presentation that’s still running. Tools that were right for a moment often stick around much longer than they should.

Where most stacks have overlap

Shopify native analytics and GA4 together. You’re getting the same session count and revenue data from two sources, neither of which is fully accurate. GA4 misses 30 to 60% of Shopify purchase events by default, due to checkout domain mismatches, ITP signal loss, and consent handling gaps. If both tools are giving you different purchase counts, you’ve stopped trusting either.

GA4 earns its place only if you’re actively using its path analysis, BigQuery export, or paid search integrations. If you’re not pulling those reports regularly, GA4 is adding noise.

Two attribution tools measuring the same channels. Triple Whale, Northbeam, and similar tools rebuild the attribution layer from scratch because GA4 is unreliable on Shopify. If you’re already paying for a dedicated attribution tool, GA4’s acquisition reporting is redundant. The attribution tool does it better.

Two session recording tools. Microsoft Clarity is free. Hotjar costs money. Running both simultaneously is common, usually because someone installed one before the other existed. Pick one.

Manual Shopify exports as a reporting process. This isn’t a tool overlap. It’s a gap showing up as a process. If you’re pulling order exports into Excel to figure out which products are performing month over month, you have a missing layer in your stack, not a reporting workflow. A spreadsheet costs analyst hours weekly and is stale the moment you finish building it.

The layer most eCommerce stacks are missing entirely

Most stacks handle jobs 1 and 2 reasonably well. Job 4 gets partial coverage from a CRM or email tool. Job 3, product-level catalog performance, is almost always missing.

Shopify Analytics gives you total revenue, session count, and overall conversion rate. It does not tell you which of your 400 products has a 9.1% conversion rate vs. which ones are pulling your average down to 2.4%. It doesn’t show you which products appear in abandoned carts 60% of the time, or which product category drives customers with the highest 90-day retention rate. What Shopify Analytics doesn’t tell you about your products documents the specific gaps in detail.

GA4 doesn’t fill this gap either. It tracks sessions and events. Your catalog has products and SKUs. The translation between those two data models doesn’t happen automatically in GA4, and even when configured manually, you’re getting item-level revenue reporting, not product-level customer behavior data.

This is the missing layer: a product analytics tool that treats your catalog as the primary unit of analysis. Not an event tracker you’ve configured to watch product pages, but a tool built natively for eCommerce product decisions.

Stormly fills this job. The core reports cover SKU-level cart abandonment (broken out by product, brand, and category), product-level conversion rate, new arrivals performance tracking, cohort retention split by first-purchase product category, repeat purchase rate by product, and customer lifetime value analysis by the product a customer first bought. The AI-powered anomaly detection flags unusual product-metric shifts before you’d catch them in a weekly review. The native Shopify integration captures full purchase event data without the gaps that affect GA4.

See your product-level analytics in Stormly. Start a free trial.

The confusion between what marketing analytics and product analytics do is exactly why this layer gets skipped. Merchants assume their attribution tool covers product performance. It covers traffic and spend. The product layer is a different job with different questions. How product analytics differs from marketing analytics explains this distinction and is worth reading before you restructure your stack.

How the layers work together

The best analytics stacks don’t run in silos. They connect.

Attribution plus product analytics. Your attribution tool tells you which campaign brought a customer. Your product analytics tells you what that customer bought, whether they returned, and what their LTV looks like at 60 and 90 days. Together, you can answer: which paid channel drives our highest-LTV customers? That’s a question neither tool answers on its own. eCommerce attribution and why product and marketing analytics need to connect covers how this changes acquisition strategy in practice.

Session recording plus product analytics. Your session recorder shows visitors dropping off on a specific product page. Your product analytics shows that same product has a 68% cart abandonment rate at the SKU level. Two signals pointing at the same problem from different angles: product analytics gives you the scale, session recording shows the specific behavior. The combination is faster to act on than either tool alone.

Shopify admin plus product analytics. Shopify admin handles orders, inventory, and fulfillment. A product analytics tool handles catalog performance and customer behavior. These don’t overlap at all. They answer entirely different questions.

What a functional stack actually looks like

For a Shopify store doing $1M to $20M in annual revenue, the core stack has three layers.

Essential: - Shopify admin for operational data and basic revenue reporting - One attribution tool for paid channel and acquisition analytics - One product analytics tool for catalog performance, customer retention, churn prediction, and anomaly detection

Add when you hit a specific problem: - A session recording tool when you’re diagnosing page-level UX issues - Email analytics when you need campaign-to-product-segment performance data

What you probably don’t need: - GA4 alongside a dedicated attribution tool - Two attribution tools measuring the same channels - Both Microsoft Clarity and Hotjar running simultaneously - Manual Shopify exports to spreadsheets for product performance reporting

The test for every tool in your current stack: which of the four jobs does this do? If two tools answer the same job, one is redundant. If job 3 has no dedicated coverage, you have a gap that compounds every week you leave it unfilled.

Before making cuts, a structured review helps. How to audit your current eCommerce analytics setup gives you a 10-question framework for evaluating each tool against your store’s actual decision needs.

Most stacks are inherited, not designed

Most eCommerce teams didn’t plan their analytics stack. They accumulated it. GA4 when the Shopify data looked questionable. A session recorder after a confusing high-traffic week. An attribution tool when the paid budget grew past a point where guessing felt expensive. Each addition solved a real problem. The subscriptions just never got rationalized afterward.

The fix is straightforward: list every analytics tool, write down which job it covers, look for where two tools answer the same job, look for which job has no coverage. Most stores doing this exercise find they can drop one or two subscriptions and add the product analytics layer for roughly the same total monthly spend.

The question to start with is simple. What’s the one thing your current stack can’t tell you that would change what you do next week?

For most Shopify stores in this revenue range, that question is some version of: which of my products is actually working, and which ones are pulling my numbers down?

If answering that requires opening three tabs and building a table manually, you’ve found your gap. That’s the layer to add first.

Stormly answers it in a dashboard. Start a free trial and run the product performance report on your catalog today.