Shopify traffic but no sales: find where the journey fails

The short answer

If your Shopify store gets sessions but few orders, treat conversion as a stage problem — not a single “site is broken” verdict. Confirm the traffic is human and relevant, then find which step loses the most worthwhile sessions: product engagement, add to cart, reaching checkout, or completing payment. Fix that stage before you rebuild the theme, stack apps or cut prices.

Shopify’s Conversion rate breakdown report is built for this view. It tracks sessions, sessions with cart additions, sessions that reached checkout, and sessions that completed checkout. Rates at each step are calculated against total sessions unless you use a closed-funnel view for sequential drop-off.

Why one conversion rate hides the leak

Overall online store conversion rate answers “what share of sessions bought?” It does not tell you whether paid social is sending the wrong audience, whether product pages fail to create confidence, or whether payment errors kill checkout. Those need different owners and different work.

Use consistent session (or shopper) cohorts and a documented date range. Do not mix raw event counts with unique-session rates. A spike in add-to-cart events can still show a weak add-to-cart rate if sessions rose faster than carts.

Step 0: prove the traffic is human and relevant

Before you redesign anything, validate the denominator. Shopify notes that bot sessions can make conversion rates look lower than expected, and Analytics supports filtering by human or bot sessions. Diagnose conversion on human traffic when you are trying to understand real shoppers — especially after a sudden traffic spike.

Then segment human sessions before blaming the theme:

• Source / channel (organic, paid search, paid social, email, direct, referral)
• Device (mobile vs desktop)
• Geography you can fulfil vs markets with weak delivery economics
• Landing page or product
• New vs returning visitors

If one campaign drives most sessions but almost no product engagement, you may have a targeting or message-match problem — not a storewide conversion problem. Compare the promise in the ad or email with the landing page, price and market before touching the storefront.

Map the journey in stages that match the data

Shopify’s default funnel visualisation moves through all sessions → sessions with cart additions → sessions that reached checkout → sessions that completed checkout. Open funnel views include people who enter mid-path; closed funnel views require sequential completion. Know which view you are reading before you act.

For a conversion programme that starts with funnel diagnosis, see Shopify CRO. For cart-to-checkout and payment-stage work, see Checkout Optimization. For speed bottlenecks, see Shopify Performance. For eligible recovery and lifecycle flows, see Email & SMS Marketing. For GA4 and Shopify reporting reconciliation, see Web Analytics & Data Tracking.

What to inspect at each stage

Sessions without product engagement. Check landing message match, collection navigation, search with zero results, and whether mobile pages feel usable. A performance problem can look like weak interest if the main content is late — but lab scores are not the same as field Core Web Vitals or revenue. A documented lab example on a Buildaze project moved Lighthouse Performance from 28 to 91; that remains a lab metric, not a conversion guarantee. See the Shopify performance case study for published caveats.

Product views without add to cart. Review offer clarity, imagery, variant selection, sizing guidance, stock and delivery/return visibility, and whether proof is authentic. Do not assume a theme rebuild is required until the product information problem is ruled in or out.

Add to cart without checkout start. Unexpected costs, unclear totals, distraction and weak mobile cart behaviour are common. Surface shipping expectations earlier when the first meaningful cost appears only at checkout.

Checkout start without payment. Inspect payment errors, address and shipping rate issues, market eligibility and supported checkout settings. Checkout UI extensions on the information, shipping and payment steps are available on Shopify Plus — confirm plan and targets before promising custom checkout UI on every plan.

Known shoppers who leave. Abandoned checkout recovery needs an identifiable, eligible recipient plus consent and suppression after purchase. Anonymous cart visitors cannot all be emailed. Operational recovery messages and marketing campaigns are different jobs.

Illustrative funnel (not Buildaze client data)

Illustration only: one matched reporting cohort has 10,000 human sessions, 500 sessions with add to cart, 300 with checkout started and 90 with a paid purchase. Add-to-cart rate is 5%; cart-to-checkout progression among cart sessions is 60%; checkout completion among checkout starts is 30%; session purchase conversion is 0.9%.

If checkout completion moved from 30% to 40% while the same 300 checkout starts were held constant, arithmetic would produce 120 orders rather than 90. That is a scenario, not a forecast. Real tests must account for seasonality, traffic mix, refunds and measurement changes.

When not to run an A/B test yet

A/B testing needs enough traffic, a clear hypothesis, a minimum detectable effect and a stopping rule. If volume cannot support a fair test in a sensible window, prefer qualitative research, technical fixes and clearly labelled observational comparisons. Do not describe a before-and-after change as a causal A/B win when it is not.

Something plainly broken — a dead size guide, a payment method failing in a key market — should be fixed, not parked behind an underpowered experiment.

Practical checklist before you rebuild

• Filter to human sessions for the diagnosis window
• Note open vs closed funnel definitions in Shopify Analytics
• Segment by source, device, market and landing page
• Name the first weak stage in absolute session terms, not only percentages
• Check tracking agreement between Shopify orders and your analytics events
• Decide the owner: traffic, PDP, cart, checkout, performance or recovery
• Only then scope redesign, apps, discounts or experiments