Wati With Causality Engine: Click or View Attribution?

The short answer: the gap shows the moment you switch Wati's Shopify dashboard from Click to View and the order count climbs while the store's own total does not move. Wati, the WhatsApp Business API platform with broadcasts, chatbots and Shopify automations, documents both rules: Click credits a purchase after the customer clicks a message, View credits it after the customer only received one, and the page adds that the figure "does not represent your total Shopify store orders" (Shopify Analytics Dashboard article). It mentions an attribution window and never prints its length. The other side of the comparison sits in GA4, for every tapped link that carried tags. Export the Attribution paths report and drop it on the start page of Causality Engine, a causal attribution tool that estimates what each channel caused from that file; the first finding costs nothing.

Disclosure: I founded Causality Engine and own it, and this page recommends it. Every claim about it links to Causality Engine's own pages.

Three readings before Wati's revenue goes into a board deck:

1. Count who received, not who tapped. For example, a shipment update reaches 1,000 buyers and 90 of them order again inside the window. View can credit all 90 to Wati; Click credits only the ones who opened the link. The order confirmation, shipment and fulfilment messages go to every buyer, so View touches nearly every repeat order.
2. Read both settings on the screen you already pay for. Attribution View switches between Click and View and recalculates orders, revenue, ROI and conversion rate (dashboard article). Write down both for the same month. The difference is orders that followed a message nobody tapped.
3. Tap one of your own links. Send a campaign to yourself and read the address bar on the store. If no utm_ parameters arrive, GA4 cannot name WhatsApp at all.

Wati and the read, row by row

Two rules, one window, no stated length

The dashboard article defines its numbers plainly. Orders are the count of unique orders attributed to Wati automations. Revenue is their summed order value. ROI is total revenue divided by total spend, and spend is estimated from messages sent and the applicable prices. It covers Abandoned Checkout, Abandoned Cart, Cash on Delivery, Order Confirmation, Follow-up After Purchase, Shipment Tracking, Order Fulfillment and Return and Exchange. When several automations touch one order, "attribution priority rules may be used", and the article does not say which.

The older Shopify app screen is stricter. Its Order Value Recovered counts revenue from customers who clicked the abandoned cart message and then bought (Shopify app analytics article). So the same cart reminder earns a click-only figure on one Wati screen and, under View, a received-only figure on the other.

Broadcasts sit apart. Campaign Overview reports spend, audience size, delivered, read and engagements, where engagements are replies plus link clicks, updated hourly (campaign analytics article). No order or revenue column appears there.

From a WhatsApp tap to a row in GA4

Wati's click tracking swaps the button's address for "a structured Wati tracking URL", by default on whatsapp.wati.io or on a branded subdomain you submit at setup (click tracking article). The article names no UTM option and does not say whether a query string you typed survives the hop. Type the tags into the Visit Website URL yourself, then test on a live send.

GA4 decides the rest. Google's default channel group page files a visit as Organic Social when its source matches its list of social sites, and the source list it publishes puts both whatsapp and whatsapp.com in the social category. Paid Social needs, in addition, a medium matching cp, ppc, retargeting or paid. Values are case sensitive, so WhatsApp and whatsapp are two sources (URL builder page). A link with no tags at all loses its source, and Google lists missing UTM parameters first among the causes of (direct) / (none) (direct traffic page).

One Wati article sounds relevant and points the other way. Capture and store UTM parameters reads tags from your website's chat widget into a Wati contact through a Collect UTM chatbot node. That tells Wati which campaign started a chat. It sends nothing to GA4.

A worked example: one October, four claims

For example, with figures made up to show the arithmetic and no store behind them: the shop books 1,480 orders in October. Wati's dashboard shows 118 orders under Click and 301 under View. Meta Ads Manager claims 742 and Google Ads 655. In GA4's last-click view, Organic Social, where the tagged Wati links land beside Instagram posts, holds 64 orders and Direct holds 311.

Under View: 301 + 742 + 655 = 1,698 orders claimed, 1,698 - 1,480 = 218 more than the shop took, before organic search or email say a word. Under Click: 118 + 742 + 655 = 1,515, still 35 over. And 301 - 118 = 183 orders went to Wati from customers who never tapped a link, so not one of them can appear in GA4 as a WhatsApp visit.

A causal read does not pick between Click and View. Its how-it-works page describes what it does with the export: split Direct back to the channels that sent those 311 buyers, set each channel's caused share beside a last-click rebuild from the same file, grade each with a data-health score from 0 to 100, and give a next step to raise, cut or hold, tested in one region first. Organic Social is one row there, WhatsApp and Instagram together. If that row's caused share sits far below 301 orders, hold the broadcast budget and test before you add volume.

Hold back a tenth of the list first

The free check needs no new tool. On Pro and Business, a static custom segment captures a fixed snapshot of contacts and can be chosen as a campaign audience (segments article). Build two from a random split. On Growth, split the CSV before uploading.

For example, 4,000 opted-in contacts: 3,600 receive the offer and 400 do not. After 7 days, 144 of the 3,600 have ordered, which is 144 ÷ 3,600 = 4.0%, and 12 of the 400, which is 12 ÷ 400 = 3.0%. The broadcast caused about (0.040 - 0.030) x 3,600 = 36 orders. That figure stands without any other product, and it is the one to set beside Wati's View count for the same send.

For every channel at once, Causality Engine, which I founded, takes the GA4 export on the start page: a free finding worked out in your browser, then the full read at €99 once, with a refund inside 30 days (pricing, checked on 11 November 2026).

Where the read stops

There is no Wati connector, and none is needed. Causality Engine's integrations page lists eight platform pages, Wati not among them, and says it measures every channel visible in your GA4 data. The consequences:

• Untapped messages are invisible to it. The 183 View-only orders above leave no WhatsApp trace in GA4.
• One row, not one trigger. Which automation earned its keep stays a Wati question.
• History and scale. Export 12 months or more; a channel under 1% of credit on both methods is called too small to act on, and Unassigned traffic is flagged as a tagging problem first (pricing for agents). Where spend is too small for a signal, the homepage says it declines.

Wati's pricing page sells "ROI tracking" as part of Business, listed at $279 a month billed annually (checked on 11 November 2026), and the View setting returns the larger ROI from the very same orders, so the switch that flatters the dashboard is also the strongest case for the top tier.

See also

• Wati: plans, the message rate card, complaints weighed
• How to use Wati: a first broadcast with a holdout group
• Wati on the marketplace: Wati's affiliate program terms
• Causality Engine: what the read takes and returns
• How to use Causality Engine: from GA4 export to regional test
• Causality Engine on the marketplace: its affiliate terms, stated as the owner's
• respond.io vs Wati: sales inbox against broadcast engine
• How to find what direct traffic is hiding: untagged visits traced in GA4

FAQ

Can Wati push its attributed orders into Causality Engine?

No. Nothing passes between the two products. The read works only from the Attribution paths file you export from GA4, where Wati shows up solely as the tagged WhatsApp visits that reached the store (integrations page).

Which Wati setting should go in a report to finance?

My recommendation: Click, labelled as Wati's own count, with the holdout result beside it. View includes customers who received a message and never opened the link, which is the reason the dashboard article separates the two.

Why do tagged WhatsApp visits land in Organic Social?

Because Google's source list classes whatsapp as a social site, and the Organic Social rule matches on source alone (default channel group). A medium starting with paid moves the visit to Paid Social; there is no WhatsApp channel among the defaults.