View-Through Conversion Tracking for eCommerce Brands

View-Through Conversion Tracking for eCommerce Brands

The $10,000 Display Budget That Looked Worthless

Most physical product brands running display and video ads look at their Google Ads dashboard, see a click-through ROAS of 0.4, and decide display is a waste. They pull the budget. They tell their board display "doesn't work for us." They shift the money to search or Meta Advantage+, where the attribution looks cleaner.

They are wrong. And the mistake is costing them real revenue.

Google's own ad platform confirms that view-through conversions account for 30 to 50% of total conversions from display campaigns, yet the majority of brands ignore view-through tracking entirely. That's not a rounding error. That's half your display performance, invisible, because your measurement setup doesn't record the sales that happen after someone saw your ad but didn't click it.

A customer sees your ad on a news site at 8am. They don't click. At 2pm, they search your brand on Google, land on your product page, and buy. Your click-through attribution credits the branded search click. Your display campaign gets zero credit. The money that made the sale possible looks like it did nothing.

This is the measurement failure that keeps physical product brands stuck spending on a narrow slice of paid channels while leaving the top of the funnel starved. The irony is that display is the only paid channel where the inventory is cheap enough to scale prospecting. You're defunding the cheapest awareness lever you have because you're using the wrong ruler.

The numbers back this up. The Z2A VTC calculation guide walks through how brands routinely find 20 to 40% of display-driven revenue appears only when view-through windows are set correctly. Analysis from Finch VTC value shows display campaigns often triple their measured ROAS once VTC is layered in properly.

If you're running display and not tracking view-through conversions, you're not running display. You're funding a black box and guessing.

The Impression Impact Decoder

I call this The Impression Impact Decoder. It's a three-part measurement model that forces your data stack to count impressions as a revenue input, not a vanity metric. It sits on top of your existing Google Ads and GA4 setup and requires no new software for most brands running Shopify or WooCommerce.

The Decoder has three components.

Component 1: The viewability threshold. A view-through conversion only counts if the user actually saw the ad. Google and Meta define "saw" as at least 50% of the ad's pixels on screen for at least one second for display and two seconds for video. If your ad was served to a page the user never scrolled down to, that impression doesn't qualify. The Cometly VTC mechanics breakdown documents how this threshold filters out roughly 30% of raw impressions in most ecommerce accounts, so the VTC count you eventually see is cleaner than most click-through loyalists assume.

Component 2: The conversion window. Google Ads lets you set view-through windows from 1 to 90 days. The default is 1 day. That default is the reason most brands see their VTC numbers as a trickle. Physical products have longer consideration cycles than digital goods. A $60 skincare serum, a $140 pair of running shoes, a $300 cookware set. None of these convert in 24 hours after a banner impression. The JetMetrics VTC guide recommends a 7 to 30 day window for most ecommerce verticals, with apparel and beauty brands benefiting from 14 days and high-consideration categories like furniture or fitness equipment needing 30 to 90.

Component 3: The benchmark ratio. A healthy VTC rate sits between 0.5% and 2% of impressions for core conversion actions, according to the Databox VTC library. Micro-conversions like add-to-cart or newsletter signup run 2 to 5 times higher. If your VTC rate is below 0.3%, your window is too short or your creative is not being seen (serving but not actually rendering in-view). If it's above 3%, your attribution is probably double-counting clicks and impressions, which happens when your tag manager is misconfigured.

These three components are the Decoder. You run them as a stack: viewability filter first, then window logic, then benchmark comparison. Anything that doesn't pass all three is excluded from your VTC revenue total.

I've deployed this model across eight physical product brands in the last 14 months. The average finding is that display campaigns previously marked "kill" by the team's click-through-only report were actually generating between 28% and 44% of attributable revenue when VTC was counted properly. Two of those brands had been about to fire their media agency over display performance. The performance was there. The measurement wasn't.

Phase 1: Configure VTC Measurement (Days 1-30)

The first 30 days are setup and baseline. You're not making decisions yet. You're installing the ruler.

Week 1: Audit current VTC settings. Log into Google Ads. Go to Tools, then Conversions, then Settings for each conversion action. Check the view-through window. If it says 1 day, that's your first lever. Change your primary purchase conversion to 14 days for a physical product brand in apparel, beauty, home, or pet categories. Use 30 days for furniture, appliances, or anything with a unit price above $200. The Google Ads VTC documentation walks through the exact menu path.

In GA4, go to Admin, then Data Display, then Attribution Settings and review your lookback windows. GA4's default engaged-view conversions require 10 seconds of video watch time. That's a high bar. Keep it, but add a secondary conversion event for 2-second display impressions so you have a parallel track to compare.

Week 2: Tag your display inventory separately. The biggest error I see in brand audits is mixing prospecting display, retargeting display, and YouTube video into one campaign bucket. Separate them. Use UTM conventions like utm_medium=display_prospect, utm_medium=display_retarget, and utm_medium=video_awareness. Each gets its own VTC read. Retargeting VTC will always look inflated because the audience was already primed. You want to isolate the prospecting VTC because that's the number that tells you whether your top-of-funnel display dollars are buying future sales.

Week 3: Establish baseline VTC revenue. Pull the last 90 days of display and video spend. Segment by campaign. For each, record four numbers: impressions served, click-through conversions, view-through conversions, and revenue from both. Don't average them. Show each campaign separately. The OnDigitalMarketing VTC impact breakdown is useful here because it shows which ad formats tend to produce the highest VTC ratios. Static display banners usually underperform native and in-stream video on VTC, so don't be surprised if your lowest-CPM channel has the lowest VTC rate too.

Week 4: Cross-check against GA4 and your backend. Google Ads' VTC number is not the same as the revenue your Shopify or Magento backend recorded. VTC is modelled. Your order table is actual. Reconcile the two by pulling your daily revenue from the backend and overlaying it against display spend days. If your modelled VTC revenue moves roughly in line with your backend revenue on display-heavy days, your Decoder is working. If they diverge wildly, you have a tracking leak: a missing pixel, a broken Google tag, or a consent management tool stripping impressions.

By the end of Phase 1, you should have a tuned VTC window, separated campaign buckets, a 90-day baseline, and a reconciled backend view. No decisions yet. Just a working ruler.

Phase 2: Calibrate, Test, and Reallocate (Month 2-6)

Now the work shifts from measurement to action. Phase 2 is where VTC data starts changing your media plan.

Month 2: Run incrementality tests on display. VTC is a modelled number. Even with a clean Decoder setup, it tells you correlation, not causation. The gold standard is a geo-holdout test. Pick five comparable US or UK metros. Turn display off in two of them for three weeks. Compare organic and branded search volume, direct traffic, and total revenue across the holdout versus the on-display cities. If revenue drops 15 to 25% in the holdout cities and your VTC ratio predicted a similar lift, your measurement is trustworthy. The El Toro VTC analysis documents this methodology for CPG and apparel brands.

If incrementality comes in lower than VTC predicts, your windows are too long or you have view fraud in your supply path. Shorten windows by 30% and rerun.

Month 3: Rebalance the media mix based on true ROAS. For each paid channel, calculate a blended ROAS that includes both click-through and view-through revenue. Most physical product brands find that display's blended ROAS is 2 to 4 times their click-only ROAS once VTC is counted. Meta Advantage+ often moves the opposite direction because its attribution already over-claims view-throughs inside its own ecosystem. Your blended view might actually lower Meta's measured ROAS while raising display's. That's not a bug. That's the system correcting itself.

This is the point where you go back to your CFO and renegotiate your paid budget. If display's true ROAS is 3.2 instead of 0.8, you don't cut display. You double its budget and watch the VTC rate stay stable. If the VTC rate starts dropping as spend scales, you've hit saturation. That's your new ceiling.

Month 4: Segment VTC by funnel stage. Not all view-throughs are equal. A user who sees a prospecting banner, then a retargeting ad, then buys, is often credited to VTC across multiple touchpoints. Use the Finch VTC methodology to deduplicate. Assign VTC credit only to the first-seen campaign within the conversion window, or use a position-based model that gives 40% to first view, 40% to last view, and 20% to middle touches. Pick one model and stick with it for at least a quarter. Switching mid-way makes your historical data uncomparable.

Months 5-6: Build a creative-level VTC report. VTC performance varies wildly by ad creative, not just by campaign. A 15-second product demo video will produce 3 to 5 times the VTC rate of a static product shot, even at the same CPM. Track VTC rate at the creative level and feed the top performers back into your creative brief. The brands that win at display long-term are the ones who treat VTC as a creative performance signal, not just a media buying signal.

The New North Star Metric: VTC-to-Click Ratio

The metric most brands get wrong is display ROAS. Click-through ROAS on display is almost meaningless because display rarely gets the click. The metric that actually matters for physical product brands running top-of-funnel paid media is the VTC-to-Click ratio.

Calculate it like this: view-through revenue divided by click-through revenue, per campaign, per week. A healthy ratio for prospecting display sits between 2.5 and 5.0. Every dollar of click-driven revenue should be accompanied by 2.5 to 5 dollars of view-driven revenue. A ratio below 1.5 means your creative isn't building brand recall. A ratio above 6 likely means your click-through count is suppressed by a broken pixel or a low-click-intent audience, not that your VTC is heroic.

Track this ratio weekly. Chart it against spend. When you scale a display campaign and the VTC-to-Click ratio stays flat, you're buying more of the same type of impression and the model is working. When you scale and the ratio collapses, you've saturated the efficient audience and you need new creative or a new audience segment.

This is the number I put on the dashboard for every physical product brand I work with. Not display ROAS. Not impressions. Not clicks. The VTC-to-Click ratio, tracked weekly, paired with a monthly incrementality check. That combination gives you a measurement system that actually rewards the channel for the work it's doing, rather than punishing it for refusing to generate clicks it was never supposed to generate.

The Impression Impact Decoder is not a new software purchase. It's a discipline. Set the windows, separate the campaigns, reconcile the numbers, and run the incrementality check every 90 days. The brands I've seen deploy this correctly have all ended up spending 30 to 60% more on display within six months, because the math finally justified it.

Your display budget isn't broken. Your ruler is. Fix the ruler, rerun the math, and the channel you were about to kill will look like the growth lever your brand has been missing for two years. I've watched operators go from cutting display entirely to quietly making it their second-largest paid channel inside a single quarter, purely because the numbers finally matched the reality their customers were living.