Understanding incrementality - and why it matters for Shopify SEO and PPC campaigns

In the weekly trading meetings of many Shopify brands, there is often a discrepancy. If you sum the revenue reported by your paid media agency, your SEO team, and your email flows, the total often exceeds your actual bankable revenue.

This gap exists because most brands measure channel performance, not incrementality.

For a Head of eCommerce, "incrementality" is the distinction between moving revenue around and actually generating it. It asks the fundamental question: If we turned this activity off tomorrow, would total sales actually drop?.

Understanding incrementality is critical for SEO reporting, but it is equally vital for PPC efficiency. It is the only way to ensure your ad spend is driving net profit rather than just claiming credit for customers who were going to buy anyway.

The challenge of siloed reporting

The modern marketing stack is often fractured. Brands typically hire specialists to manage specific channels: one for email, one for SEO, one for paid. To demonstrate value, these partners report on their own platform metrics in isolation.

This structure creates a natural tension:

• Paid teams often bid on branded terms because they convert well, while SEO teams claim branded traffic because it drives volume.

• SEO teams often claim branded traffic because it drives volume.

• The business may see flat total revenue, even when individual channel reports show "record" results.

×

We run PPC campaigns ourselves, so we see this dynamic from both sides. We know that the relationship between paid and organic is not a battle for credit, but a balance to be struck. When planning a PPC campaign, we have to consider that pushing hard on paid brand terms often cannibalises the high-intent traffic that SEO would have captured naturally.

Without someone to interpret how these channels interrelate, it is difficult to see the whole picture.

The industry-wide complexity

This isn't just a problem for your specific store; it is a structural issue that Google and Meta have been identifying for years.

In 2019, Google released a foundational report identifying the "Three Grand Challenges" of marketing effectiveness. They highlighted that estimating the true causal effect of marketing is difficult because observational data (like attribution) is inevitably confounded by external factors. (You can read their full report here).

At the same time, large-scale econometric research commissioned by Meta demonstrated that while digital attribution focuses on short-term sales, roughly 60% of total ROI comes from long-term effects that happen weeks or months later. (You can view that analysis here).

These papers remain the benchmark because the core problem hasn't changed: attribution models tell you what happened (the last click), not why it happened (the cause).

The GA4 problem in Shopify

This complexity is compounded because of the unique relationship between Shopify, paid platforms, and tracking. Unlike paid channels, which have their own pixels and reporting ecosystems, overall store revenue attribution often relies on Google Analytics.

However, GA4 is hugely unpopular in the Shopify ecosystem. With Universal Analytics, everyone knew where they stood. Since the migration, that confidence hasn't carried over.

Tracking issues and consent mode failures mean that a significant portion of traffic is often categorised as "Direct" or "Unassigned" rather than organic. Consequently, the resulting lack of trust means many Shopify managers simply don't bother looking at GA4 anymore.

This creates a data imbalance. When tracking is imperfect, paid channels may still claim a conversion via their own attribution windows, while organic credit is lost to the "Direct" bucket. This leads to a distorted view where paid channels appear highly efficient, potentially masking the fact that they are paying for traffic that SEO (or Direct) was already capturing.

A practical approach to "flaky" data

We are not looking for something mathematically sound in an academic sense. We are transparent about the fact that digital data is often too "flaky" to achieve perfect precision. Instead, we need a sense of what is going on.

We don't run randomised control experiments or complex cohort studies. Our premise is simpler:

1. Fix the tracking foundation

We focus on getting tracking working reasonably well. In the Shopify space, we've frequently worked with server-side tracking solutions and apps like Elevar, Addingwell, LittleData, and leaf.fm.

Crucially, this also involves strictly configuring a consent management platform (CMP). For Shopify stores, we have found Pandectes to be really good.

2. Acknowledge the "shuffle"

Quite often, once tracking is fixed using these tools, we will see a two or three hundred percent improvement in organic revenue year-on-year. It is vital to be honest here: this is not net new revenue. It is revenue that has been shuffled from the "Direct" or "Unassigned" columns back into organic. It is purely an attribution fix, not immediate growth.

3. Extrapolate from there

Once the tracking is fixed, the relationship between sessions and revenue is re-established. From there, we can look at the channel mix with clearer eyes.

If paid revenue spikes by 20% but total revenue remains flat, we can see that the growth isn't incremental - it’s likely cannibalisation.

By extrapolating from cleaner data, we can start to answer the hard questions. Are we generating new customers through generic search terms, or are we just paying for clicks from people who were already trying to find us?