How-To Guide

How do you track social referrals in Shopify?

Use Shopify's referral and marketing reports, keep the 30-day attribution window explicit, and tag social links so sessions and orders are credited.

9 min read

Start with the Sessions by referrer report for visit volume, then use the marketing reports to see which referrer and UTM values produced sessions, orders and sales. Tag every social link with one utm_source per platform, one utm_medium, one campaign name and a utm_content per creator, and read one attribution model consistently. Two limits decide what the number means: Shopify resets first interaction when no purchase happens within 30 days of a session, and an in-app browser or a cross-device journey can drop the tag before checkout.

  • Sessions by referrer counts visits only, so it cannot tell you which social post sold something.
  • The marketing reports add orders and sales, and their attribution menu changes which channel gets credit without changing the order total.
  • A tagged link is the strongest evidence; a unique promo code and a post-purchase survey cover the journeys where the tag is lost.

The short answer

Shopify tracks social referrals in three places, and each answers a different question. The Sessions by referrer report tells you how many visits arrived and whether they came from direct, search or a referral. The marketing reports attach orders and sales to a referrer source, a marketing channel or a UTM value. The attribution menu decides which touchpoint receives the credit. None of them is a post-level social dashboard by default.

So the workable method is a path, not a single report. Tag every social link so the arrival is identifiable, read the report that matches the question, and write down the attribution window you are reading. Shopify's Help Center defines the referrer values and the 30-day reset. Google's Analytics Help defines what happens when the same visitor arrives with campaign parameters. Read together, those pages explain why Shopify and Google Analytics can show different social numbers for the same week.

  • For reach and visit volume, start with Sessions by referrer. It counts sessions and visitors only.
  • For orders and sales, use the marketing reports and name the attribution model beside the date range.
  • For a specific post or creator, add a utm_content value and a unique promo code, because the stock reports stop at channel or campaign level.

This page now walks the three surfaces, the two windows that change the number, the tagging scheme that survives the trip from a post to a product page, and the journeys where a tag is lost.

Three panels compare Shopify's Sessions by referrer report, Sales attributed to marketing report and Performance by marketing channel report by the question each answers and what each counts.

Original diagram of the three Shopify report surfaces. Definitions are from the Shopify Help Center acquisition and marketing report pages, read 18 November 2026.

Superdirector

Make it work for your brand

Find it. Remake it. Post it. Superdirector turns the viral formats in your niche into your brand's carousels, X posts and video scripts, and keeps the calendar full.

Which Shopify report answers which question

Open Analytics, then Reports, and filter by category. Acquisition holds visit counts. Marketing holds money. Mixing the two is the most common way to read a social channel as better or worse than it is.

The Sessions by referrer report is a visit report. It splits arrivals into direct, search and referral, and it covers the Online Store sales channel only. Shopify is explicit that acquisition reports differ from the Total sales by referrer report because they do not show converted sales or the amount of orders. A referral here is a visit, not a sale.

Social becomes a named row in the marketing reports. Their Referrer source value includes Social, which means the customer clicked a link on social media. Add Referrer source, a UTM dimension, or Marketing channel to a report and you can separate an Instagram visit from an email visit. That Help Center page notes that UTM dimensions show raw session data while marketing channel, referring channel, referring medium and traffic type are processed and standardized values. That difference matters when one report calls the same visit Social and another calls it Organic Social.

SurfaceWhat it countsWhat it cannot tell you
Sessions by referrerSessions and visitors, split direct, search and referralWhether a visit became an order or revenue
Sales attributed to marketingSales from traffic directly attributable to marketing, including UTM campaignsThe full sales total, because unattributable sales sit in other reports
Performance by marketing channelSessions, orders, sales, conversion, spend and ROAS by channelA post-level answer unless a UTM dimension is added

Two practical details save a bad afternoon. Performance by marketing channel defaults to the last 30 days and the last non-direct click model, so its numbers can differ from a report configured another way. And a dash in a spend column means the cost is unavailable, not zero. Shopify says a dash is not the same as a zero.

For a quick check, open the Analytics dashboard. It shows sales, sessions and fulfillment metrics and updates within about a minute. Use it to see movement, then open the report that owns the metric before you make a decision.

The two windows that change the number

Shopify owns the first window. A conversion, in Shopify's definition, happens when an online store visitor becomes a paying customer. First interaction means the referrer that introduced the customer. Last interaction means the referrer used just before the order. Shopify resets the first interaction when a visitor does not buy within 30 days of a session: the next referrer becomes the new first interaction. A social post can therefore start the relationship and still lose first-click credit if the order lands after that reset.

Attribution models own the second window. Shopify's marketing reports support last non-direct click, last click, first click, any click and linear. Last non-direct click is the default when you view marketing activity data. Use first click to learn which channel created interest, and last non-direct click to see what closed the sale without handing credit to a direct visit. Choose one model for a report and record it next to the date range.

Google's GA4 and Universal Analytics comparison has its own version of the same idea. A GA4 session is not restarted when new campaign parameters appear mid-visit. It ends after more than 30 minutes of inactivity, subject to the session timeout setting. Campaign timeout sets the attribution lookback window for a session. Changing the attribution model does not change the total number of key events. It changes which rows receive credit.

That is why a report can look inflated while the total is flat. A changed model or a different window moves credit between rows. It does not create orders. If two views disagree, check the window and the model before you check the traffic.

Two timelines compare Shopify's 30-day first-interaction reset with the Google Analytics rule that a new campaign parameter does not restart a session and the Campaign timeout sets the lookback.

Original diagram of the two attribution windows. Rules are from Shopify Help Center and the Google Analytics Help comparison and traffic-source pages, read 18 November 2026.

Tag the path from post to product page

A tag is the strongest evidence you can put on a social link. Google's campaign URL guidance says that adding utm campaign parameters to the destination URLs used in referral links and ad campaigns lets you view which campaigns refer traffic. When a user clicks a referral link, the parameters are sent to Analytics and the values appear in the Traffic acquisition report. Shopify's own analytics walkthrough makes the same move on the store side: add region and UTM parameters to the Sessions by referrer report to see how visitors arrived from each source.

Use four fields and use them the same way every time. Google's manual tagging page recommends one unique utm_source for each distinct platform, with Facebook, Instagram and TikTok as the example. Keep utm_medium as social. Put one exact campaign name in utm_campaign so a single effort does not split into several rows. Put the creator or the post in utm_content, which is the only field that can narrow a result to one piece of content.

  • utm_source: instagram, tiktok or pinterest, one value per platform.
  • utm_medium: social, spelled the same way on every link.
  • utm_campaign: one exact name for the effort, not a new spelling per post.
  • utm_content: the creator handle or post identifier, kept short and consistent.

Set the full set. Google's manual tagging page says that when any UTM parameter is present, Analytics derives all cross-channel traffic source values from UTM parameters exclusively, and that missing parameters produce (not set) values in reporting. Values are case sensitive too, so Meta and meta are different sources. Set the complete set or expect a partial label.

A clean link looks like this, with the store domain replaced by yours: yourstore.com/products/example?utm_source=instagram&utm_medium=social&utm_campaign=fall-drop&utm_content=creator-a. Keep it in one place, a template or a saved link, so the next post does not start from a blank field. The influencer campaign tagging guide covers the creator side of the same scheme.

A social link tagged with utm source, medium, campaign and content reaches the store, while a second panel shows a promo code and a post-purchase survey covering in-app browser and cross-device gaps.

Original diagram of the tagging scheme and its fallback. Field guidance is from the Google Analytics URL builder page; the in-app browser and cross-device gaps are from Later's guide and Shopify's influencer measurement guide, read 18 November 2026. The promo-code and survey labels are editorial recommendations, not measured outcomes.

What the reports miss, and how to catch it

A tag only works if it survives the journey. Later's measurement guide names the failure shape: in-app browsers do not persist cookies, so tracking is lost when a user jumps from an app to the web, and a cross-device purchase breaks standard tracking. A link-in-bio path adds clicks before the product page. That guide calls for a redundant stack, not one perfect method.

Two records cover most of that gap. A unique promo code per creator catches a purchase that the tag missed, including a purchase that happened on another device. A post-purchase survey question, asking how the customer heard about the store, catches some of what neither the link nor the code records. Later describes self-reported data as directional, not precise, and useful for triangulation. That is the right weight to give it.

Shopify's influencer measurement guide adds the timing problem. Some customers convert later or return through a different channel, and the guide suggests extending the attribution window to account for it. It also lists the measurement methods it trusts for sales: affiliate links, promo codes and custom URLs. A post-purchase survey, in its account, can reveal customers who saw a product in a post and converted elsewhere.

Two data limits belong in the same note. Shopify's session-based historical data goes back only to October 1, 2022, so a year-over-year comparison before that date is not a like-for-like comparison. And spend-based metrics appear only for marketing channels that report cost data to Shopify. A channel with no cost data shows a dash, which is not a zero.

Read the reports together

Pick one source of truth per metric before you compare anything. Use Shopify's acquisition report for visit volume, the marketing reports for orders and sales that carry a tag or a known referrer, and the ad platform for spend. Later's guide makes the same point: platform metrics, GA4 and Shopify data will not match perfectly, because different attribution windows and counting methods cause discrepancies.

Then run a short check in order, on one date range and one model:

  1. Open Analytics, then Reports, then the Acquisition category, and read Sessions by referrer for the week.
  2. Open the marketing reports, add Referrer source or a UTM dimension, and filter to Social or to the campaign you are checking.
  3. Write the attribution model and date range beside the numbers before you read them.
  4. Compare the same window in GA4 under Traffic acquisition, and note whether you are reading session scope or event-scope campaign values.
  5. Tag any live link that still lacks utm_content, then wait a full reporting window before comparing the two views again.

A safe conclusion names its surface and window: for example, social referrals produced this many sessions in Shopify's acquisition report for this date range, and this many attributed sales in the marketing report under this model. A conclusion that drops the surface and window will not survive the next report somebody opens.

If you also read GA4, the organic social traffic guide explains how that platform groups the same arrival. For stores selling through short video, the TikTok for Shopify guide covers the platform side of the link.

Where Superdirector fits

Clean attribution matters most when the report drives a decision about the account itself. Superdirector's Growth plan at $199 a month runs one AI-operated social account: it researches what is working in the niche, builds posts and carousels, schedules them, and publishes to supported Instagram and X accounts under account-level standing posting permission. The Starter plan at $29 a month is the Claude or ChatGPT lane, where the same data, ideas, scripts and downloads are available but the account is not run for you.

This page's measurement advice applies either way. A tagged link, a named window and one recorded attribution model make a social report readable whether a person or a system reads it next.

Frequently asked questions

How do you track social referrals in Shopify?

Use three Shopify surfaces together. Sessions by referrer gives visit volume and splits direct, search and referral. Marketing reports give orders and sales by referrer source, marketing channel or UTM value, with an attribution model you choose. For a specific post, tag the link with utm_content for the creator and keep a unique promo code, because Shopify does not show post-level sales by default.

Does Shopify show which social post drove a sale?

Not by default. The acquisition report stops at sessions, and the marketing reports group at referrer source, channel or campaign level. A utm_content value per creator or post, or a unique promo code, is what narrows a sale to a post. Even then the credit depends on the attribution model and the 30-day first-interaction reset.

Why does social traffic show as direct in Shopify?

Direct means the customer entered the store URL or the referrer could not be determined. Shopify lists a visit whose referrer is your own store as direct, and a browser that drops the referrer leaves the visit without a source. A UTM tag on the post link and a promo code are the two records that recover the path.

What is the 30-day rule in Shopify analytics?

Shopify resets the first-interaction referrer when a visitor does not buy within 30 days of a session. Whatever referrer comes next becomes the new first interaction. A social post that started the interest can therefore lose first-click credit if the order lands more than 30 days later.

Which Shopify attribution model should I use for social?

Use first click to learn which channel created interest, and last non-direct click to see what closed the sale without giving direct visits the credit. Choose one model for a report, write it beside the date range, and compare like with like. Changing the model moves credit between rows but does not change the order total.

Should I use UTM tags or promo codes?

Both. A UTM tag is the strongest evidence when the link survives to the store. A promo code, plus a post-purchase survey question, catches the in-app browser and cross-device journeys where the tag is lost. Later and Shopify's own guide both describe this as a redundant stack, not a choice between methods.

Related practical guides

Pick one report and one window before you change a link

Open Analytics, then Reports, then the Acquisition category, and read Sessions by referrer for a week. Then open the marketing reports, filter to Social or to one campaign, and write the attribution model and date range beside the numbers. If the two views disagree, the window or the model is the first thing to check.