Skip to main content
Card class: HeroCategory: Ecommerce Platform
creation_source breakdown showing where each customer first entered the Directory: POS terminal vs Square Online vs Invoices.

At a glance

A breakdown of new customer records by where they were first captured, the Customers Directory creation_source field. TERMINAL means the customer was first added at a POS register, ONLINE_STORE means they checked out on Square Online, INVOICE means they were created when an invoice was sent. This is the channel-of-first-touch view: it answers “is my retail counter or my website doing the customer acquisition?” which is the foundation for cross-channel attribution on a Square account.
What it countsNew customer records in the Customers Directory created in the window, grouped by creation_source. The card surfaces the TERMINAL (POS) vs ONLINE_STORE vs INVOICE split as a donut.
creation_source valuesSquare’s Customers Directory tags each record with a creation_source. The relevant retail-vs-web values are TERMINAL (added at a POS terminal), ONLINE_STORE (Square Online checkout), and INVOICE (created via Square Invoices). Other sources exist (DIRECTORY for manual adds, IMPORT, THIRD_PARTY, INSTANT_PROFILE, LOYALTY) and are grouped as “other”.
VAT / tax treatmentn/a, this is a customer count, not money.
Refunds / ordersn/a. This counts customer records, not orders. A customer who places ten orders still counts once, at their first-touch source.
Channels / sourcesThis card is the channel split. It is the customer-acquisition mirror of the order-level channel cards. Where Revenue by Channel splits money, this splits people.
Currency / unitnumber (count of customer records per source)
Time window30D (rolling 30-day window of new records)
Alert triggernone. This is an attribution and mix card, not a threshold alarm.
Rolesowner, marketing

Calculation

Calculated automatically from your Square Online data. See the At a glance summary above for what the metric tracks and the worked example below for a typical reading.

Worked example

A US lifestyle brand on Square: a flagship retail store, a Square Online storefront, and Square Invoices for wholesale and custom orders. The 30-day window covers 14 Mar 26 to 12 Apr 26. The donut shows where the period’s new customers first entered the Directory.
creation_sourceMeaningNew customersShare
TERMINALFirst captured at the POS register1,84061%
ONLINE_STOREFirst captured at Square Online checkout94031%
INVOICECreated when an invoice was sent1505%
other (DIRECTORY, IMPORT, LOYALTY)Manual adds, imports, loyalty sign-ups903%
Total new customers (this card)3,020100%
Three things to notice:
  1. Retail is the acquisition engine, web is the second channel. 61% of new customers are first captured at the POS terminal. That is typical for a brand with strong foot traffic, but it reframes the website’s job: the web channel is acquiring 31% of new customers, not just converting existing ones. A “the website is just for our regulars” assumption is wrong here.
  2. First-touch is not where they spend. A customer captured as TERMINAL may later become a heavy web buyer, and vice versa. This card is channel-of-first-touch only. To see where customers actually transact over time, pair it with Revenue by Channel. A large gap between first-touch share and revenue share is the cross-channel migration story.
  3. POS capture depends on the cashier asking. TERMINAL customer creation only happens when a register associate collects an email or phone at checkout. A drop in the TERMINAL slice often means staff stopped asking, not that fewer customers walked in. That is an operational coaching signal, not a demand signal. Square’s INSTANT_PROFILE source (auto-created from a card) can fill some of the gap but yields a thinner record.

Sibling cards merchants should reference together

CardWhy pair it with Customer Capture Source
New CustomersThe headline new-customer count this card breaks down by source. New Customers is the “how many”, this card is the “from where”.
Revenue by Channel (POS / Online / Invoices)The money mirror. First-touch capture share vs revenue share reveals where customers are acquired vs where they spend.
Email-Opted-In Customers (Directory)Capture is only useful if the record is marketable. This card finds the source; the opt-in card tells you how much of it you can email.
Repeat Purchase RateFirst-touch source often predicts retention. Web-acquired and POS-acquired customers can repeat at very different rates.
Total Customers (Directory)The full Directory size this card’s new records flow into.
Square Orders Email AttributionOnce you know first-touch source, attribution links those customers back to the orders they drove.

Reconciling against Square

Where to look in the Square Dashboard: Square Dashboard, Customers, Directory. Square does not expose a single “by creation source” donut in the UI, but you can build the comparison with filters: the Directory supports filtering by Creation source (Point of Sale, Online, Invoices, Imported, and so on) and by Date added. Set the same date range you’ve selected here and read each source count off the filtered totals. Other Square Dashboard views that look like the same number but aren’t:
  • Customers, Directory, total count: the full Directory, not just new records in the window. Filter by Date added to match this card.
  • Customers, Insights: aggregates lifetime behaviour (visit frequency, spend) rather than capture source.
  • Reports, Sales by customer: order-level, not customer-creation-level. A customer who bought ten times appears once here, but ten orders there.
Why our number may legitimately differ from Square Dashboard:
ReasonDirection of divergence
Source bucketing. We map TERMINAL to POS, ONLINE_STORE to Online, INVOICE to Invoices, and group the rest as “other”. If you read Square’s raw filter labels they may split or combine these differently.Bucket labels differ, totals match
Date-added vs first-order date. This card uses the record’s creation date (created_at). A customer can be created before they place an order, or via an invoice that predates checkout.Edge cases at boundaries
Instant profiles. Square auto-creates INSTANT_PROFILE records from card data at the register. Depending on settings these may or may not be visible in the Directory filter.”other” bucket can shift
Merged / deduplicated records. Square merges duplicate customer records; a merge can change the surviving record’s source.Small counts move
Sync lag. The most recent ~5 minutes of new records may not be in our index.Self-resolves within minutes
Cross-connector reconciliation:
CardExpected relationshipWhat causes legitimate divergence
klaviyo / email platform list growthSquare ONLINE_STORE and TERMINAL capture feeds your email list growthOnly opted-in customers (see Email-Opted-In Customers) sync to an email platform. Capture count is always higher than syncable list growth.
google_analytics.new_usersGA4 only sees Square Online (web) new usersGA4 cannot see POS or invoice customer capture. Expect GA4 new users to track only the ONLINE_STORE slice, minus the usual tracking gap.
The Square one-Directory advantage: because POS, Square Online, and Invoices all write customer records into a single Customers Directory with a creation_source tag, Square can tell you channel-of-first-touch natively. On platforms where POS is a separate product, retail and web customer lists live apart and this cross-channel first-touch view has to be stitched together by hand.

Known limitations / merchant FAQs

What exactly does creation_source mean? It is the Square Customers Directory field recording where a customer record was first created. TERMINAL means a POS register, ONLINE_STORE means a Square Online checkout, INVOICE means an invoice was sent. Other values (DIRECTORY for manual adds, IMPORT, LOYALTY, INSTANT_PROFILE, THIRD_PARTY) are grouped as “other”. It is first-touch only and never changes once set. My POS slice dropped. Did foot traffic fall? Not necessarily, and this is the most important caveat on the card. TERMINAL records are only created when a register associate captures an email or phone at checkout. If the slice falls while order volume holds, the likeliest cause is that staff stopped asking for contact details, a coaching and process issue, not a demand drop. Check order counts before assuming traffic fell. Does a customer captured in-store but who later buys online move to the Online slice? No. This card is channel-of-first-touch. Once a record is created with creation_source = TERMINAL, it stays counted as POS-captured forever, even if that customer becomes a pure web buyer. To see where customers actually spend over time, use Revenue by Channel. Why is there no alert on this card? It is an attribution and mix card, not a threshold alarm. There is no universally “bad” split, a 90% POS account and a 90% web account can both be perfectly healthy. The value is in watching the mix shift over time and pairing it with revenue and retention cards. If you want movement alerts on the underlying counts, watch New Customers. What about customers with no source, or imported lists? Imported customers carry creation_source = IMPORT and fall in the “other” bucket, they were not captured at a live touchpoint, so counting them as POS or Online would distort the acquisition read. Manually added records (DIRECTORY) are likewise “other”. A large “other” share usually means a recent list import or heavy manual data entry. Can a single checkout create more than one record? Square deduplicates against existing records by email and phone, so a returning customer does not generate a new record. Occasionally a typo or a different email creates a duplicate; Square’s merge tool consolidates these, and a merge can change which creation_source survives. This is rare and affects small counts only.

Tracked live in Vortex IQ Nerve Centre

Customer Capture Source (POS vs Online) is one of hundreds of KPI pulses Vortex IQ tracks across Square Online and 70+ other ecommerce connectors. Nerve Centre runs the detection layer; Vortex Mind investigates the cause when something moves; Ask Viq lets you interrogate any number in plain English. Start for free or book a demo to see this metric running on your own data.