Google Ads · mulebox.com · account 24 · 13 Aug – 11 Sep 2026 (30 days)
Spend and clicks are up; conversions are down and cost per conversion is up 16.9%. Efficiency is moving the wrong way. The three findings below explain most of it — but read the first one before trusting any figure above it.
95% of what Smart Bidding chases is one GA4 form-submit action.
| Conversion action | Conversions | Value | Bid on? |
|---|---|---|---|
| mulebox.com GA4 (web) website form submit (JT) | 367.0 | $367.00 | Yes |
| Calls from ads | 19.0 | $37.99 | Yes |
| mulebox.com GA4 (web) website_form_submit | 270.5 | $270.53 | No — secondary |
Two GA4 imports measure the same form submit — one under its human name with a “(JT)” suffix, one under the raw GA4 event name. Only the first is primary, so bidding is not double-counting. But “all conversions” reports 699 against 386 bid-on, so anyone reading the Google Ads UI sees close to double the truth.
Every form submit carries a value of exactly $1.00 — a placeholder. That makes the ROAS on the dashboard (0.04×) meaningless. Nobody should quote ROAS on this account until conversion values are real.
Unverified
Are 367 form submits actually 367 leads?
Per the August sweep, four accounts in the book bid on GA4 funnel-step events running 5–8× over — MI-BOX First Coast had 181 real bookings behind roughly 1,380 conversions. 367 leads in a month for one DFW franchise is a large claim. Until it is reconciled against real bookings, $28.98 is not a cost per lead — it is a cost per event. Every recommendation below depends on the answer.
$1,819 of $5,708 (32%) went to search terms with zero conversions — concentrated in competitor-brand and price-shopping queries.
| Search term | Spend | Clicks | Conv. |
|---|---|---|---|
| what is cheaper than pods | $38.55 | 1 | 0 |
| pods competitors | $37.60 | 1 | 0 |
| pod rental dallas | $36.31 | 1 | 0 |
| sam storage | $30.55 | 1 | 0 |
| cheapest pod rental | $29.24 | 2 | 0 |
| moving containers | $68.45 | 4 | 0 |
That is $25–40 per click to appear against a competitor’s brand name
and against people explicitly shopping on price. It is what sits behind
search | pods | 15 miles #4: the second-largest spender in the account, a
$12.59 CPC, and the worst cost per conversion at $47.86.
A 3× spread in cost per conversion, with the most money on the worst performer.
| Campaign | Spend | CPC | Cost / conv. | |
|---|---|---|---|---|
| search | portable | 25 mile radius | $4,565 | $7.60 | $30.52 | |
| search | pods | 15 miles #4 | $2,895 | $12.59 | $47.86 | |
| pmax | ros | $2,652 | $2.25 | $22.73 | |
| pmax | ros #2 | $754 | $3.68 | $15.64 |
The cheapest source of conversions — pmax | ros #2 at $15.64 — receives $754.
The most expensive, at $47.86, receives $2,895. On the reported numbers that allocation is backwards.
Two caveats. PMax is 31% of spend with no search-term visibility, and may be harvesting the same brand queries Search is paying $38 for. And PMax cost-per-conversion flatters itself when the conversion signal is a cheap web event (see 01).
search | pods | 15 miles exists
as #1 through #5 with only #4 live. Serial cloning rather than iterating in place means no campaign
carries more than a few weeks of learning — a standing drag on Smart Bidding.