forProgressive Windows & Doors
Google Ads Report · Q3 2026

The real cost of a new customer.

Over Q3 2026, Google Ads reported 135 conversions at $87 each. Brand search accounts for 19% of them, so the clearest read on new demand is the non-brand line, at about $99 per conversion. The mix is worth seeing: 74 website forms, 55 calls from new callers and 4 calls from returning callers.

135
Conversions
Google-reported
$11,768
Google spend
Q3 2026
$87
Blended cost / conv.
All campaigns
$99
Non-brand cost / conv.
The real acquisition number
Conversions as reported in Google Ads. The blended figure mixes brand and non-brand; the non-brand line isolates new demand. It also blends conversion actions. Hover any conversion count to see the split.
The story

Three numbers that tell it.

The headline cost per conversion is a strong start. Here is where the new customers are coming from, so we know exactly which lever to pull next.

01
$37

Brand search is efficient

25 conversions come from people already searching Progressive Windows & Doors, at about $37 each. Efficient, and it is demand that already knows you.

02
84

Non-brand drives new demand

NZ:PMax | Replacement Windows & Doors… delivered 84 conversions, the biggest source of genuinely new customers this period, and the cheapest at $80 each. Those break down as 51 website forms, 29 calls from new callers and 4 calls from returning callers.

03
$99

The number we work down

Strip out brand and acquisition runs about $99 per conversion. That is the figure we optimize against.

Google Ads

Conversions over the quarter.

Weekly conversions against weekly spend across the quarter.

Conversions & spend per week

Conversions Weekly spend
Window total 135 conv.Shaded band = the current 13 weeks, the bars before them are the previous periodLast 3 days still filling in
i
Why the weekly bars change height part-way through. What Google counts as a conversion changed inside this window: Call from a new caller started counting on July 12. The 135 total for the window mixes both definitions, so read the trend by shape, not by comparing a week before that change against a week after it.
Period over period

Against the prior 92 days.

The same length of time immediately before this one (Mar 31 to Jun 30).

MetricPrior periodThis periodChange
Google spend$8,282$11,768+42%
Conversions37135+264%
Non-brand conversions31109+252%
Non-brand cost / conv.$258$99-61%

How each action moved

Conversion actionPrior periodThis periodChangeMeasured on
Website form1374+469%counted as conversions
Call from the ad268-69%every time it happened
i
How to read it. This report counts leads differently from the earlier ones. Phone taps on the site are no longer counted as leads. A tap opens the dialer, and only some of those become a real conversation, so counting taps made the lead total look several times larger than it was. From here on a lead means a connected phone call or a completed website form, and we have set a minimum call length so hangups, wrong numbers and very short calls do not count. Both periods above are counted on this new definition, so the comparison between them is like for like. Taps are no longer counted anywhere on this page, so every figure you see is on one definition. One note on the breakdown: calls from the ad are compared on every time they happened, because Google changed how many of them it counted between the two periods. That keeps the change a real one.
Campaign split

Where the budget worked.

Spend and conversions by campaign. Brand search is separated so the true acquisition cost stays visible.

$6,726
NZ:PMax | Replacement…
84 conv · $80 CPA
$4,103
NZ - Search - Non-Brand…
25 conv · $164 CPA
$938
NZ - Search - Brand - OK
26 conv · $37 CPA
3
Active campaigns
Search + PMax
NZ:PMax | Replacement Windows… 84 conv · $80 CPA
$6,726
NZ - Search - Non-Brand - OK 25 conv · $164 CPA
$4,103
NZ - Search - Brand - OK 26 conv · $37 CPA
$938
CampaignStatusSpendConv.Cost / conv.CTR
NZ:PMax | Replacement Windows & Doors | OKActive$6,72684$804.1%
NZ - Search - Non-Brand - OKActive$4,10325$1646.4%
NZ - Search - Brand - OKActive$93826$3738.2%
Total$11,768135$87-
i
How to read the cost per conversion: brand search converts efficiently because those people already know you, so new-customer efficiency reads cleanest on the non-brand line, at about $99 per conversion. That is the number we bring down together.
Conversion actionWhat it measuresCountedTimes recorded
Website forma form completed on the website7474
Call from a new callera phone call from someone who had not called the business before. Recorded every time it happened.5573
Call from a returning callera phone call from someone who had called before. Recorded every time it happened.414
Call from the ada call placed straight from the ad. Recorded every time it happened.28
Total135169
How they reach you

Device, day and hour.

Who is arriving, when they arrive and how they choose to contact you. This is the part your team can act on without touching the account.

93%
On mobile
Share of conversions
Tuesday
Busiest day
25 conversions
9am to 12pm
Busiest window
37 conversions

By time of day

12am to 6am 5 website forms
5
6am to 9am 8 calls from new callers, 7 website forms and 3 calls from returning callers
18
9am to 12pm 22 calls from new callers, 14 website forms and 1 call from a returning caller
37
12pm to 3pm 12 calls from new callers, 11 website forms and 1 call from the ad
24
3pm to 6pm 11 calls from new callers, 11 website forms and 1 call from the ad
23
6pm to 9pm 13 website forms and 1 call from a new caller
14
9pm to 12am 13 website forms
13

By day of the week

Monday
21
Tuesday
25
Wednesday
17
Thursday
19
Friday
21
Saturday
15
Sunday
15

By device

DeviceClicksConv.ShareConv. rate
Mobile4,25412593%2.9%
Desktop10397%9.1%
i
Two different customers. The phone rings during business hours: 100% of the 2 calls come in between 6am and 6pm. After hours the traffic keeps converting, but it comes in as forms, about 31 of them in the evening and overnight. How fast those get answered the next morning is worth as much as the ad spend.
Call tracking

What happened on the phone.

Every call these ads produced and what became of it, measured on the call itself rather than on the ad platform. Google can see that a call started; this is how it ended. These counts will not tie exactly to the conversion table above, and they are not meant to: Google dates a call to the day the ad was clicked and waits a day for the data, so the same calls land on slightly different days in the two views.

136
Calls from these ads
Tracked end to end
87%
Reached someone
118 of 136
18
Never reached a person
13% of the calls
80
Ran past a minute
301 minutes on the phone
What happened to the callCallsShare
Reached someone11887%
Caller hung up while it rang1310%
Went to voicemail43%
Rang out unanswered11%
i
18 of the 136 calls never reached a person. Most of them are the caller hanging up while the phone rang. This is the one number on the page your team can move without touching the ad account.

How long the answered calls ran

Under 30 seconds
22
30 to 60 seconds
16
1 to 3 minutes
43
Over 3 minutes
37

Calls by day of the week

Monday 21 of 26 reached someone
26
Tuesday 30 of 33 reached someone
33
Wednesday 15 of 16 reached someone
16
Thursday 14 of 16 reached someone
16
Friday 21 of 24 reached someone
24
Saturday 11 of 14 reached someone
14
Sunday 6 of 7 reached someone
7
i
116 of the 136 callers had never called you before. That is what these campaigns are for: the phone ringing with someone new on the other end.
The next chapter

Now we run it.

We keep brand cheap, push non-brand acquisition, and bring the true cost down each period. Tell us how the leads land and we steer toward the deals you want to close.

Talk through the plan