No DMARC record
19%
of 963 stores publish no DMARC record at all, so anyone can send email that claims to come from them.
95% interval: 17% to 22%
Research
The State of Ecommerce SEO 2026
A random sample of Shopify and WooCommerce stores, read the way a crawler reads them, finds the same avoidable failures again and again.
75%do not enforce DMARCof 963 stores checked
48%have images with no alt textof 963 stores checked
46%have a fully out-of-stock productof 928 stores checked
963 stores, 19,127 pages, one crawl. These are the highest shares of the 963 stores audited.
Email trust
Email authentication decides whether order and newsletter mail reaches an inbox. DMARC tells a receiving mail server what to do when someone else sends as your domain.
DMARC not enforced
75%
of 963 stores do not enforce DMARC (no record, or a monitoring-only p=none policy).
95% interval: 72% to 77%
No SPF record
12%
of 963 stores have no SPF record, so their order and newsletter email is more likely to land in spam.
95% interval: 9.9% to 14%
AI readiness
How the store treats crawlers that feed AI assistants, and whether it publishes a machine-readable summary for them.
An AI crawler is blocked
2.9%
of 963 stores block at least one AI crawler in robots.txt.
95% interval: 2.0% to 4.2%
GPTBot is blocked
1.7%
of 963 stores block OpenAI's GPTBot.
95% interval: 1.0% to 2.7%
llms.txt published
79%
of 963 stores publish an llms.txt file for AI assistants.
95% interval: 77% to 82%
Product data
What a crawler can learn about the products from the pages and from the store's public catalogue.
Invalid JSON-LD
8.0%
of 963 stores serve at least one page with a JSON-LD block that is not valid JSON (for example a trailing comma or a raw line break in a text value).
95% interval: 6.4% to 9.9%
No valid price
13%
of 928 stores list at least one product with no valid price in its public catalogue.
95% interval: 11% to 15%
Fully out of stock
46%
of 928 stores have at least one product that is fully out of stock in every variant.
95% interval: 43% to 49%
Crawl health
Signals that waste crawl effort and split indexation across a site.
Dead internal links
6.5%
of 963 stores have dead internal links among the links checked.
95% interval: 5.1% to 8.3%
Soft 404
8.3%
of 963 stores serve a page that answers 200 but reads like "not found" (a soft 404).
95% interval: 6.7% to 10%
Noindex page in the sitemap
6.1%
of 963 stores list a noindex page in their sitemap, sending Google contradictory signals.
95% interval: 4.8% to 7.8%
Speed and images
Page weight and image delivery, measured on the pages the audit read.
A heavy image
45%
of 963 stores serve at least one image over 250 KB on the pages checked.
95% interval: 41% to 48%
Legacy image formats only
41%
of 963 stores serve images only in legacy formats, with no WebP or AVIF.
95% interval: 38% to 44%
Images with no alt text
48%
of 963 stores have images with no alt text on the pages checked.
95% interval: 45% to 51%
Trust pages
The policy pages shoppers and search systems look for, and whether they are linked from the store.
No shipping policy
36%
of 948 stores link no shipping policy page that we could find.
95% interval: 33% to 39%
No returns policy
31%
of 948 stores link no returns or refund policy page that we could find.
95% interval: 28% to 34%
Shopify vs WooCommerce
The same checks, side by side. The two groups are not the same size, so a gap between them is a lead to investigate, not a verdict on the platform.
The platform split
Same audit. Different patterns.
Three gaps worth a closer look, then every check in one place.
Shopify · 713WooCommerce · 250
AI file publishedShopify 99%WooCommerce 24%Noindex page in sitemapShopify 1.8%WooCommerce 18%No shipping policy linkedShopify 26%WooCommerce 65%
| Check | Shopify | WooCommerce | Difference |
|---|---|---|---|
| Email trust | |||
| No DMARC record | 18%n = 713 | 23%n = 250 | Woo +5.1 pp |
| DMARC not enforced | 74%n = 713 | 75%n = 250 | Woo +1.0 pp |
| No SPF record | 13%n = 713 | 8.0%n = 250 | Shopify +5.2 pp |
| AI readiness | |||
| An AI crawler is blocked | 1.3%n = 713 | 7.6%n = 250 | Woo +6.3 pp |
| GPTBot is blocked | 0.7%n = 713 | 4.4%n = 250 | Woo +3.7 pp |
| llms.txt published | 99%n = 713 | 24%n = 250 | Shopify +74 pp |
| Product data | |||
| Invalid JSON-LD | 11%n = 713 | 0.8%n = 250 | Shopify +9.7 pp |
| No valid price | 12%n = 702 | 16%n = 226 | Woo +3.7 pp |
| Fully out of stock | 56%n = 702 | 14%n = 226 | Shopify +42 pp |
| Crawl health | |||
| Dead internal links | 6.0%n = 713 | 8.0%n = 250 | Woo +2.0 pp |
| Soft 404 | 5.2%n = 713 | 17%n = 250 | Woo +12 pp |
| Noindex page in the sitemap | 1.8%n = 713 | 18%n = 250 | Woo +17 pp |
| Speed and images | |||
| A heavy image | 50%n = 713 | 29%n = 250 | Shopify +21 pp |
| Legacy image formats only | 39%n = 713 | 48%n = 250 | Woo +9.2 pp |
| Images with no alt text | 49%n = 713 | 46%n = 250 | Shopify +2.3 pp |
| Trust pages | |||
| No shipping policy | 26%n = 713 | 65%n = 235 | Woo +38 pp |
| No returns policy | 22%n = 713 | 56%n = 235 | Woo +34 pp |
Email trust
Shopify n = 71318%
WooCommerce n = 25023%
Shopify n = 71374%
WooCommerce n = 25075%
Shopify n = 71313%
WooCommerce n = 2508.0%
AI readiness
Shopify n = 7131.3%
WooCommerce n = 2507.6%
Shopify n = 7130.7%
WooCommerce n = 2504.4%
Shopify n = 71399%
WooCommerce n = 25024%
Product data
Shopify n = 71311%
WooCommerce n = 2500.8%
Shopify n = 70212%
WooCommerce n = 22616%
Shopify n = 70256%
WooCommerce n = 22614%
Crawl health
Shopify n = 7136.0%
WooCommerce n = 2508.0%
Shopify n = 7135.2%
WooCommerce n = 25017%
Shopify n = 7131.8%
WooCommerce n = 25018%
Speed and images
Shopify n = 71350%
WooCommerce n = 25029%
Shopify n = 71339%
WooCommerce n = 25048%
Shopify n = 71349%
WooCommerce n = 25046%
Trust pages
Shopify n = 71326%
WooCommerce n = 23565%
Shopify n = 71322%
WooCommerce n = 23556%
Benchmark your store
Every store went through the same 20-page audit, so the counts compare like with like. Errors and warnings per store, at the median: 43 for all 963 stores, 40 across 713 Shopify stores and 63 across 250 WooCommerce stores.
Run the same audit on your store and compare its count against this sample.
Methodology
We took a random sample of 1000 stores from the Tranco L5PZ4 (created 2026-09-25) list (seed 2026), a public ranking of the most visited domains, and kept the ones that sell products on Shopify or WooCommerce. From 22,389 domains probed, 1000 entered the sample. 963 were audited, 24 were blocked and 13 were unreachable. The platform split was 713 Shopify stores and 250 WooCommerce stores.
Platform was detected by calling each store's public product API and parsing the response, not by matching a fingerprint in the HTML. Detection is limited to Shopify and WooCommerce; a store on another platform is not counted.
Each audited store was read with the same 20-page audit the site runs: a sample of public pages, fetched from outside the network that serves them, the way a crawler sees them. Across the sample the audit read 19,127 pages. The audit honours robots.txt: 538 requests were skipped because robots.txt disallowed them. Failed fetches were retried before a page was treated as unavailable.
A blocked store is one whose home page refused an outside crawler, for example with a bot filter, so nothing further could be read. This happened to 24 stores (4 Shopify, 20 WooCommerce). An unreachable store gave no answer at all. Neither is averaged in as a store with one finding. The crawl started on 2026-09-26.
The limits: this is one crawl on one date, so it is a snapshot and not a trend. It is an outside view, which is what a crawler gets, and it can differ from what the owner sees on a signed-in session. It reads a sample of pages per store, not every page a store has. Platform detection covers only Shopify and WooCommerce. The intervals are 95 percent, so overlapping intervals should not be read as a difference.
What we corrected before publishing: running the study exposed errors in our own audit, and we fixed them before counting anything (platform detection fooled by HTML error pages, stores that redirect to www flagged for their canonicals, files and sign-in pages sampled as if they were pages, footer links past our read limit, policy pages missed in languages other than English, sitemap entries past the first 400 ignored). Two measures are left out of this edition because a spot check showed they overstated the problem: product pages without Product structured data (the markup often sits past the part of a large page we read) and content that appears only after JavaScript runs (flagged pages usually had no price at all rather than a hidden one). The out-of-stock figure compares platforms only loosely: WooCommerce stores can hide sold-out products from their public catalogue, so their share is likely understated.
The sample method and seed are described above. The full domain list is below; the sampling, crawl and aggregation scripts are available to researchers on request.
Inside the study
A closer look at the measures and the sample behind them. The report presents the findings; full research files are available on request.
01 / Aggregate measures
What we measured
Each figure has its own sample size and 95% interval in the report.
DMARC not enforced75%
A heavy image45%
Images with no alt text48%
Explore the platform comparison 02 / Sample register
Who was counted
1,000 confirmed stores. The directory shows domain and platform only.
20percent.co.zaShopify
360cookware.comShopify
3dprintcart.comWooCommerce
View the clickable preview The stores sampled
The domains behind the sample, with their detected platform. You can visit each store; the study does not publish findings for individual stores.
Sample register1,000 stores
Domain Platform- 20percent.co.zaShopify
- 360cookware.comShopify
- 3dprintcart.comWooCommerce
- 4frnt.comShopify
- 5startuning.comWooCommerce
- 7pandas.com.auShopify
Browse all 1,000 stores
Scroll to browse the register. Links open in a new tab without sending a referrer. No store-level results are published.
Open the directory to browse.
Need your store removed?
If you manage a store in this list, send its domain and an email we can use to verify your request. We review requests by hand.
How to cite
Shipwork. The State of Ecommerce SEO 2026. 2026. https://shipwork.io/research/state-of-ecommerce-seo-2026
Published figures may be reused under CC BY 4.0 with credit to Shipwork and a link to the study. More numbers from the same sample are on the statistics page.