After consent banners are deployed, GA4 and ad platforms show fewer sales than the shop’s own order records: in the example of an outdoor equipment shop, that is 876 transactions in GA4 against 1,200 paid orders. Consent Mode and missing conversions in GA4 mean a permanent loss of measurement. Before peak season it is easy to mistake this for a real drop in sales, and a decision based on such a reading can cut the budget of a campaign that is making money. This guide shows how to calculate daily GA4 coverage against shop orders, read channel differences from GA4 and ad platform figures, and derive a correction factor for adjusted ROAS, with MER as the check.
Three sources, one September, three different numbers: the shop records 1,200 orders, GA4 sees 876 transactions, and Google Ads reports 432 conversions for a channel to which GA4 attributes 360 transactions. Implementing and validating the purchase event belongs to a separate audit. This post is about interpreting the loss that remains even after a correct implementation, and how to get an accurate reading of campaign profitability from it.
A scene from the September review
Shop orders versus GA4 transactions
The manager of an outdoor sports shop opens the September results before sitting down to plan Q4. The shop panel shows 1,200 paid orders worth €114,000, GA4 reports 876 transactions worth €83,220, and Google Ads shows 432 purchase conversions. All three numbers describe the same month of the same shop, yet 324 orders and €30,780 are missing between the shop records and GA4.
September · outdoor sports shop · 30 days
Each tile is 12 orders: 73 solid tiles = 876 transactions visible in GA4, 27 hatched = 324 orders recorded only by the shop. Total 1,200 / 876. Illustrative data.
The shop, GA4 and the ad platform count the same sales by three different rules, so their results will never match exactly. How to combine these sources into one picture, and why they are compared on shared days and totals, is covered in the guide to automated e-commerce analytics. This post deals with one of these discrepancies: orders that GA4 never saw at all.
GA4 sees an order only if the tag in the buyer’s browser was able to send the purchase event. The most common causes of the gap:
- Consent refused on the banner: the buyer clicks “reject” and from that moment analytics tags record neither the visit nor the purchase.
- Ad and script blockers: browser extensions stop the GA4 tag before it sends anything.
- Safari and browsers that restrict cookies: the shortened cookie lifetime cuts off the user identifier and the traffic source, so some purchases disappear or are assigned to the wrong channel.
- Switching between devices: an ad click on a phone, a purchase on a laptop a few days later; the path breaks, and the purchasing session is often the one without consent.
Consent Mode (the consent setting in Google tags) determines what happens after a refusal. In basic mode, tags without consent send nothing, so the order disappears from GA4 and from the ad platforms. In advanced mode, cookieless signals are sent and Google’s modelling fills part of the gap; however, it does so only in some reports and only once traffic and consent volume thresholds are met, so the shortfall remains in the raw data and in many reports.
Banner configuration and the legal basis for collecting consent are legal matters and fall outside the scope of this guide.
The shop checkout is the anchor, because it records every order regardless of consent. The order is created on the shop’s server, with an amount and a payment status that neither a banner nor a blocker can stop. Every other figure (GA4, the Google Ads interface, the Meta interface) is therefore read as a percentage of the checkout.
The Google Ads interface shows more than GA4, because it adds modelled conversions and counts in its own way. The panel reports 432 conversions where GA4 attributes 360 transactions to Google Ads, i.e. 1.20 times more. Neither figure replaces the shop’s orders: the shop total of 1,200 is the reference for the whole, and per-channel numbers remain GA4 or panel readings.
The 27% gap remains even with a correctly implemented purchase event. It stems from buyers’ decisions and browser settings, so no tag fix will close it. Instead of fighting it, it needs to be measured: per day, as a stable baseline against which subsequent weeks are compared, and across channels through GA4 and ad platform readings.
Definitions and formulas
GA4 coverage and correction factor: definitions and formulas
Two figures calculated against the shop checkout are enough to correct reports: GA4 coverage and the correction factor. First, a shared vocabulary is needed, because the word “conversion” means something different in GA4, in the Google Ads interface and in the shop.
Working definitions
Consent Mode is the way Google tags behave when cookie consent is missing. In basic mode they send nothing; in advanced mode they send signals without identifiers, from which Google models some of the conversions.
GA4 coverage is the percentage of the shop’s paid orders (excluding cancelled ones) that GA4 recorded as transactions. It is calculated on both the number of orders and their value.
Platform-to-GA4 ratio is the purchase conversions reported by the ad platform divided by GA4 transactions attributed to the same channel. It shows where GA4 probably loses more than average; both numbers are readings from their own sources.
The correction factor is the inverse of coverage: shop orders divided by GA4 transactions. GA4 figures are multiplied by it to bring them back to the checkout scale.
Coverage by count and by value comes out identical in the example, because a single AOV of €95 was assumed for all channels. In a real shop these two values diverge when consent is refused more often by customers with cheaper or more expensive baskets. In that case, the number of orders is corrected using coverage by count, and revenue and ROAS using coverage by value.
Four calculation rules determine whether coverage measures tracking or an error in the calculation itself:
- The same day and the same time zone. The shop records an order according to the server clock, GA4 according to the time zone set in the property. A one-hour offset shifts orders placed around midnight to the neighbouring day and distorts daily coverage.
- On the shop side, only paid orders, excluding cancelled ones. Unpaid and cancelled orders in the denominator inflate the checkout with sales that never happened and falsely lower coverage.
- Comparison on daily aggregates. The consent gap is statistical in nature, so daily totals of shop orders and GA4 transactions are enough to measure it; per-channel GA4 and panel figures are read alongside. Matching individual orders to transactions by order_id is not needed for this.
- Period coverage calculated from total orders and transactions. For September that is 876 / 1,200. An average of thirty daily percentages would give a day with 34 orders the same weight as a day with 47.
Platform readings are often noticeably higher than GA4. For Google Ads, GA4 attributes 360 transactions, while the interface reports 432 conversions (1.20 times more), because it adds modelled conversions and counts according to its own attribution window and click date. The higher figure in the interface contains more estimation, which is why the platform does not replace the checkout as the reference point.
Every shop has its own coverage baseline, set from its history: in the example, September stays within a 72–74% band day after day. When coverage sits below the shop’s own baseline from the first day of implementation, the cause is often the implementation itself, for example the purchase event does not fire for some payment methods or after returning from the payment provider. Such a case is resolved by an audit of GA4 events and purchase coverage; this guide deals with the shortfall that remains after a correct implementation.
Two figures are enough to correct any report. Coverage shows what share of sales tracking sees; the factor shows what to multiply a GA4 reading by to bring it back to the checkout scale. Both are calculated on shop and GA4 totals, and channel differences are read from the platform-to-GA4 ratio, because the gap is distributed unevenly.
Channels
The measurement gap across channels in GA4 and ad platforms
Overall coverage of 73% describes the whole shop, but the loss is probably spread unevenly across channels. The factor of 1.37 describes the whole shop well, while budget moves between channels. Shop orders cannot show this split, so channels are read through GA4 and ad platform figures, scaled by the overall factor.
Per-channel figures come from GA4 and the ad panels and are labelled as such. GA4 transactions come from the source/medium report mapped to channel groups; the estimate scales them by the overall factor of 1.37, and panel conversions show where GA4 loses more than average. September in a sports and outdoor shop (illustrative data) looks like this:
| Channel | GA4 transactions | Ad platform conversions (panel) | Estimate: GA4 × 1.37 | Share of GA4 transactions | Ratio to GA4 (panel; total: shop orders) |
|---|---|---|---|---|---|
| Google Ads | 360 | 432 | 493 | 41.1% | 1.20 |
| Meta | 130 | 200 | 178 | 14.8% | 1.54 |
| Organic (search) | 238 | – | 326 | 27.2% | – |
| Direct and unassigned | 104 | – | 143 | 11.9% | – |
| Other sources | 44 | – | 60 | 5.0% | – |
| Total | 876 | – | 1,200 | 100% | 1.37 |
The panel-to-GA4 ratio suggests that social loses the most. The Meta panel reports 200 purchases where GA4 sees 130 (1.54 times more), Google Ads 432 against 360 (1.20 times): traffic from in-app ads lands mainly on phones, where the consent banner more often ends in refusal, and in-app browsers lose cookies and entry parameters. Visits from search more often happen in a regular browser and complete the purchase in a single session, so GA4 probably loses less there.
The overall factor is therefore a conservative estimate for Meta and a generous one for organic. Scaling Meta’s GA4 transactions by 1.37 gives an estimated 130 × 1.37 = 178 purchases, while the Meta panel reports 200 purchases. The same multiplier for organic gives 238 × 1.37 = 326, with no panel to check against and a real loss that is probably smaller; each channel is read as a range between the GA4 × 1.37 estimate and the panel reading, and the whole is checked by MER.
Direct in GA4 collects some of the orders that lost their source. When a session loses cookies or parameters along the way (consent refused on the first visit, a change of device, a link opened from an app), GA4 assigns the transaction to direct traffic instead of the campaign. That is why direct channel coverage should be read with caution: its 104 transactions mix genuine typed-in visits with purchases whose origin GA4 has lost.
The smallest channel has the noisiest reading. Other sources account for 44 GA4 transactions a month, i.e. 1–2 a day, so their share can jump from week to week. A figure like this is calculated from several full weeks and treated as indicative.
Revenue from ad platforms is not summed as a substitute for shop records: the same purchase is often attributed by both Google Ads and Meta, so the sum of the platforms overstates sales, and the reference remains the shop’s orders in total, i.e. MER, while per-channel numbers stay labelled as GA4 or panel estimates.
Methodological note: the figures are illustrative. Shop orders carry no traffic source, so per-channel figures are GA4 and panel readings, and the estimate scales GA4 by the overall factor. GA4’s data-driven model and the platforms’ attribution windows split the same purchases differently, so the per-channel estimate is an approximation: it works correctly when GA4 and the panels use the same channel mapping and the same period.
Each channel is read as a range between the GA4 estimate and the panel. Each channel gets an estimate (GA4 × 1.37) and, where a panel exists, a second reading; the sum of the estimates comes back to the 1,200 orders in the shop records, and the panel-to-GA4 ratio (1.20 for Google Ads, 1.54 for Meta) shows where the estimate is conservative.
Budget decision
Corrected ROAS before a budget decision
The measurement gap costs the most at the moment of a budget decision. ROAS calculated from GA4 revenue inherits the 27% shortfall directly: the spend in the denominator is complete, because the ad platform knows every euro, while the revenue in the numerator is cut by sales without consent. Such a reading systematically understates the profitability of every campaign, and unevenly so, because each channel loses a different share of measurement.
Working definitions
ROAS (return on ad spend) = revenue attributed to the campaign / campaign spend. The result depends on whose revenue goes into the numerator: GA4, the ad platform or, for the whole portfolio, the shop’s records.
MER (marketing efficiency ratio) = total shop revenue / total ad spend, without attributing sales to channels. Break-even ROAS = 100 / gross margin in %: below this value, advertising eats the entire margin.
The break-even point and the target come from the shop’s margin and stay fixed. The outdoor shop has a 40% gross margin, so the break-even point is 100 / 40 = 2.5 (full calculation in the guide what is a good ROAS). The owner set a target of 4.0 so that advertising leaves profit to cover fixed costs; the only thing that changes is the reading you compare with these two numbers.
Two campaigns, three readings of the same sales (September, illustrative data; scale 0–6):
Google Ads · spend €9,120
Meta · spend €4,750
The Google Ads interface shows 4.5, a result in the middle. Its 432 conversions, including modelled ones, are 1.20 times GA4’s 360, while the overall gap is 1.37, so here the panel lands below the estimate; on top of that, it counts using its own attribution windows and rules. That is why the adjustment starts from GA4 and the shop total, is checked against MER, and the ad interface remains a second checkpoint.
The portfolio MER of 8.22 is a check that consent cannot distort. Revenue of €114,000 comes from checkout and spend of €13,870 from the ad platforms, so the consent banner has no effect on this figure at all. If campaign ROAS in GA4 falls while MER holds steady, the drop lies in measurement; if MER falls as well, the problem concerns sales or spend.
The most expensive mistake after coverage drops
Cutting budget based on GA4 ROAS: Meta, with a reading of 2.6, looks like a campaign “barely breaking even” just above the 2.5 threshold and is the first to be cut. After adjustment by the overall factor the same campaign reaches 3.6, and the Meta panel shows 4.0: both readings sit well above the threshold and close to the owner’s target. Switching it off cuts profitable sales that GA4 simply does not see.
Procedure before every campaign budget decision:
Calculate GA4 coverage over 4 full weeks
Set paid checkout orders against GA4 transactions from the same days, and note per-channel GA4 transactions and panel conversions alongside. Four full weeks even out the weekday pattern and smooth over one-off deviations.
Recalculate GA4 ROAS with an adjustment factor
Factor = shop orders / GA4 transactions = 1.37: Google Ads 3.75 × 1.37 = 5.1, Meta 2.6 × 1.37 = 3.6. The panel-to-GA4 ratio (1.20 and 1.54) shows for which channel the estimate is conservative.
Compare with the panel reading and with MER
Adjusted ROAS should land within reach of the panel reading (4.5 and 4.0), and the MER of 8.22 confirms that the portfolio is profitable. A clear gap points to an error in the channel mapping or an outdated factor.
Make the decision on the adjusted value
The 2.5 threshold and the 4.0 target still come from margin; you compare 5.1 and 3.6 (with panel readings of 4.5 and 4.0) against them instead of 3.75 and 2.6. Google Ads above target is a candidate for a budget increase, Meta between threshold and target should be maintained.
Record the factors for the monthly review
The overall factor and the panel-to-GA4 ratios per channel, with their calculation date, go into the review table. After any change to the banner or consent settings they are recalculated, because old ones would skew the adjustment up or down.
The adjustment factor changes the reading; the target and threshold still come from margin. The same calculation turns “Google Ads below target” into Google Ads above target (5.1 against 4.0), and turns Meta from a candidate for switching off into a campaign well above break-even (3.6, panel 4.0). The budget did not change by a single euro; only the figure the decision rests on changed.
Diagnosing a drop
Lasting measurement loss or a real fall in sales
A drop in GA4 has three possible sources, and each requires a different response: a lasting change in measurement, a tag failure or genuinely lower sales. One figure, calculated before any interpretation, settles which it is: daily GA4 coverage against its own baseline. When coverage holds, a drop in GA4 is a drop at checkout; when coverage shifts, part of the “drop” comes from measurement.
The coverage baseline is built from full weeks after the last banner change. In an outdoor equipment shop, the week of 8–14 September gives 280 checkout orders and 204 GA4 transactions, i.e. 72.9%, and individual days fall within a narrow range of 72.2–73.7%. This stability sets the baseline: 72–74%.
8–14 September · checkout and GA4 day by day
Daily coverage on a 55–80% axis
Grey band: baseline 72–74%. Solid dots: 8–14 September (total 280/204 = 72.9%). Dashed dot: 18 September.
Bar heights change from day to day; coverage barely moves. Sunday has the most orders (47), Saturday the fewest (34), yet the share visible in GA4 stays within the same band. That is why 18 September, with 61% coverage, stands out on the chart immediately, even though 41 orders on their own look entirely normal.
Three patterns are enough to identify the cause of any drop visible in GA4. The key is always how checkout data and coverage behave, viewed together:
GA4 falls, checkout holds, coverage jumps and stays
A change in consent, the banner or browser behaviour. August → September: checkout 1,215 → 1,200 (−1.2%), GA4 972 → 876 (−9.9%), coverage from 80% to 73%. The new level becomes the new baseline.
Coverage drops out of the baseline, a second day confirms
The purchase tag, a new banner version, a change to the checkout template. 18 September: 61% against a Thursday baseline of 72.5%, and the following day also outside the band. This needs urgent checking.
Checkout and GA4 fall together with stable coverage
Measurement works correctly, but sales have dropped. Diagnosis by decomposition: sessions × conversion rate × average basket, with the channel split taken from GA4.
The August–September case shows how misleading a drop in GA4 alone can be. Had coverage stayed at 80%, September’s 1,200 orders would have produced 960 transactions in GA4; 876 were recorded. Of the 96 transactions lost in GA4, as many as 84 result from lower coverage, and only 12 correspond to 15 fewer orders in the till (15 × 0.8 = 12).
A failure is caught in the morning, on the complete data for the previous day. Comparing yesterday’s coverage with the norm for the same weekday (Thursday against Thursdays) is a fixed part of the morning review described in the guide to monitoring and anomalies in e-commerce; the exact time of the fault will only show up in the publishing history of the tag manager or the change log of the consent tool. A soft deviation is confirmed by a second day, so a single weaker reading does not yet trigger an alert.
Reaction thresholds are calculated from the norm for the specific weekday. The signal is a coverage deviation of more than 3 points on the coverage scale (for a Thursday with a norm of 72.5%: a result below 69.5% or above 75.5%), and the reaction is two such days in a row:
| Coverage reading | Example from an outdoor shop | Status | Action |
|---|---|---|---|
| Within the weekday norm (deviation up to 3%) | Thursday: 69.5–75.5% | within norm | read a drop in GA4 as a drop in the till; diagnose sales |
| Deviation over 3%, one day | 18.09: 61% (41/25) | signal | note, quick test purchase on the site, check the next day’s reading |
| Deviation over 3%, two days in a row | 18.09 and 19.09 outside the band | reaction | review the purchase tag, banner version and template; budget decisions based on MER from total till orders and panel readings, no GA4 ROAS until fixed |
| New stable level after a consent change | September 73% after 80% in August | new norm | norm from the date of the change, recalculate the correction factor and panel-to-GA4 ratios |
| Coverage stable, till falling | till and GA4 down by a similar percentage | real decline | decomposition: sessions × conversion × basket, channel split from GA4 |
Banner change
After every change to the banner, the consent tool version or the message wording, the coverage norm is recalculated from scratch, starting from the date of the change. The date goes into the shop’s event calendar as an annotation, so that in a quarter’s time nobody compares September’s 73% with August’s 80% as if they were one measurement. Until full weeks after the change have been collected (ideally four), the new norm is provisional.
Coverage first, then interpretation of the drop. Stable coverage means that GA4 faithfully reflects sales proportions and a drop in the report is a drop in the till. Shifted coverage means the correction factor must be recalculated before drawing any conclusion about campaigns.
DataOrganizer
How DataOrganizer calculates GA4 coverage on daily data
The coverage calculation requires three sources in one daily table: the shop’s till, GA4 and the Google Ads panel. Done manually, this means three exports, aligning dates and time zones, and stitching a spreadsheet together at every review. In DataOrganizer this data goes into a single data model and is refreshed every morning with the complete previous day.
- Shop (WooCommerce, Shopify, IdoSell, PrestaShop, Shoper and other connected systems): number and value of orders, status and payment method. This is the denominator of every coverage calculation.
- GA4: transactions and revenue per source/medium and campaign, i.e. the numerator of coverage and the per-channel figures for the estimate.
- Google Ads: conversions and all_conversions with values per campaign, needed for the panel-to-GA4 comparison, plus spend for ROAS.
- Meta: campaign spend and conversions with values, used to calculate corrected ROAS and the panel reading next to it.
The comparison is performed by the AI assistant on aggregates. It sets daily shop order totals against daily GA4 transaction totals and reads channels from GA4 and panel figures; the calculation works on totals, without matching individual orders. Connecting sources is self-service, with no developer required, and you pay for the connected sources.
The monthly review before a budget decision looks like this:
September: GA4 sees 876 of 1,200 paid orders, i.e. 73%. The overall correction factor is 1.37; for paid channels, GA4 and panel readings look like this:
| Channel | GA4 | Panel (total: shop) | Share of GA4 (total: coverage) | Ratio to GA4 | ROAS from GA4 | Corrected ×1.37 |
|---|---|---|---|---|---|---|
| Google Ads | 360 | 432 | 41.1% | 1.20 | 3.75 | 5.1 |
| Meta | 130 | 200 | 14.8% | 1.54 | 2.6 | 3.6 |
| Whole shop | 876 | 1,200 | 73% | 1.37 | – | MER 8.22 |
Per day: coverage holds the 72–74% norm. The exception is 18.09: 41 orders, 25 transactions, 61% after a banner version update; the following day confirmed the deviation, so this day goes for checking as a tag failure, separately from the permanent loss.
August to September: till 1,215 → 1,200 (−1.2%), GA4 972 → 876 (−9.9%), coverage from 80% to 73%. The drop in GA4 is mostly a measurement loss, while real sales are almost flat. The Google Ads panel shows 432 conversions against 360 GA4 transactions for this channel (1.20 times more), because it adds modelled conversions.
Budget conclusion: Meta, with a GA4 reading of 2.6, gives 3.6 after correction and 4.0 in the Meta panel (€19,000 / €4,750): well above the 2.5 threshold, so the campaign stays on. Google Ads corrected at 5.1 (panel 4.5) sits above the 4.0 target. The portfolio MER of 8.22 from shop orders confirms that the whole is profitable.
The answer can be repeated every morning. The same reports on the complete previous day show whether coverage stayed within the weekday norm, and at month end they provide fresh factors for the budget review. A shift in coverage shows up in the daily data before it reaches campaign decisions.
Methodological note: the figures in the example are illustrative. For simplicity, a single AOV of €95 was assumed across all channels, so coverage by count and by value comes out identical; in a real shop it is worth calculating both. Per-channel figures come from GA4 and the ad platforms, the estimate scales GA4 by the overall factor, and the way Google models conversions changes over time, which is why the correction factor is recalculated every month.
Key takeaways
Budgets are set on total shop orders (MER) and adjusted ROAS. GA4 remains the source of channel proportions and customer journeys; the correction factor from shop and GA4 totals converts its reading to the scale of orders, and daily coverage checks whether that scale has changed.
Methodological note: the figures in the examples are illustrative (a sports and outdoor shop, September), and a single average basket of €95 across all channels is a simplification. Shop orders carry no traffic source, so per-channel figures come from GA4 and the ad platforms. Coverage is calculated on daily aggregates without matching by order_id, and Google’s conversion modelling changes over time, so the coverage norm is worth refreshing every month.
Key takeaways
Make budget decisions on total shop orders (MER) and ROAS adjusted by the correction factor; GA4 remains the source of proportions and journeys. GA4 coverage measured against its own norm makes it possible to tell a permanent consent-related loss apart from a tag failure and from a real drop in sales, before anyone cuts a campaign that is making money.
FAQ
FAQ: Consent Mode and missing GA4 conversions
How many conversions does GA4 lose because of Consent Mode?
There is no single universal figure: the loss depends on the consent refusal rate, the device and browser mix, and the channel structure. In the outdoor shop example, GA4 sees 876 of 1,200 paid orders, i.e. 73%. Your own gap is calculated as GA4 transactions divided by paid shop orders from the same day.
Why does GA4 show fewer transactions than the shop?
GA4 loses orders from customers who refused consent, use blockers or cookie-restricting browsers, or switched devices during the purchase; the shop checkout records every order regardless of consent. Only paid, non-cancelled orders count in the comparison, because unpaid ones falsely lower coverage. If coverage has been low since the first day of implementation, the issue calls for an audit of the purchase event.
Does Google Ads model missing conversions after consent is refused?
In Consent Mode’s advanced mode, tags send cookieless signals and Google fills part of the gap with modelling, provided traffic thresholds are met; in basic mode nothing is sent without consent. That is why the ad platform often reports more than GA4: in the example, the Google Ads panel shows 432 conversions where GA4 attributes 360 transactions to this channel. Modelled conversions are an estimate, so the profitability check for decisions is MER calculated on total shop orders.
How do you calculate a GA4 conversion correction factor?
The correction factor is shop orders divided by GA4 transactions from the same days, calculated over the last four full weeks: 1,200 / 876 = 1.37. Multiplying channel ROAS from GA4 by it gives an estimate: Meta 2.6 × 1.37 = 3.6, Google Ads 3.75 × 1.37 = 5.1. Because the gap is uneven (the Meta panel reports 1.54 times more purchases than GA4, Google Ads 1.20 times), the result is read together with the panel reading and checked against MER from shop orders.
How do you tell a sales drop from a loss of measurement in GA4?
Check coverage first: when shop sales are flat, GA4 falls and coverage drops sharply and stays at the new level (August 80%, September 73%, shop −1.2%, GA4 −9.9%), the decline is mainly a measurement issue. Coverage falling outside the norm for that weekday for two days in a row points to a tag or banner failure, while a joint drop in shop sales and GA4 with stable coverage is a real fall in sales. DataOrganizer keeps shop orders, GA4 and Google Ads in one model refreshed each morning with the complete previous day, so the AI assistant calculates daily coverage, the correction factor and MER from a single question.
Cheat sheet for the whole guide
The Consent Mode gap in three calculations
1. 1. Shop checkout as the reference point. GA4 coverage is GA4 transactions divided by paid shop orders from the same day; your own norm (72–74% in the example) shows how much of your sales disappears from measurement even with a correct implementation.
2. 2. Adjustment before budget decisions. GA4 ROAS multiplied by the correction factor and read next to the panel figure replaces the raw GA4 reading, while the break-even and target still come from the margin (2.5 and 4.0 at a 40% margin). MER calculated from checkout data serves as a consent-proof check.
3. 3. Coverage calculated every morning in DataOrganizer. Shop orders, GA4 and Google Ads campaigns flow into a single data model with the previous day complete, and the AI assistant calculates coverage, factors and adjusted ROAS on request. A deviation from the weekday norm confirmed on a second day separates a failure from a permanent loss.
DataOrganizer · consent measurement gap
Connect your shop, GA4 and ad platforms: GA4 coverage per day, the correction factor, GA4 and panel readings per channel, corrected ROAS and MER on shop orders are calculated from the data. Self-service setup, no developer needed; you pay only for connected sources.