Cover Your Pergola: Meta Spend vs HubSpot Revenue, 2025 vs 2026

Year to date (Jan 1 to Nov 3) Meta spend against revenue from all Meta contacts, 2026 vs 2025, overall and by campaign (matched by campaign ID first, then campaign name).

Headline: YTD 2026 vs YTD 2025

Spend grew 22% in 2026, but revenue from Meta contacts fell 42%, so ROAS fell from 3.39x to 1.62x. Spend is all Meta spend (three ad accounts); contacts, customers and revenue are all Meta contacts created Jan 1 to Nov 3, with closed-won deals to Nov 3, including the Unattributed Meta contacts that could not be tied to a campaign. Late 2026 leads have had less time to close; the Matured cohort tab controls for that.

2026 YTD spend already exceeds full year 2025 spend ($821,646).

Two years side by side

Light: 2025 YTD Dark: 2026 YTD

Campaign table

Spend and Meta CPA are from Meta Ads Manager exports (Jan 1 to Nov 3). Contacts, customers, revenue and SQLs are Meta contacts created Jan 1 to Nov 3 matched to that campaign by campaign ID, or by exact campaign name when no ID is available, with closed-won deals to Nov 3. Unattributed Meta holds the contacts no method could place; it has no spend, so ROAS and cost columns show n/a.

Sort by Account Search

Campaign	Status	Spend	Contacts	Cust.	Revenue	ROAS	Meta CPA	Cost/SQL	Cost/cust

Meta CPA is Meta cost per website lead from Meta Ads Manager (Jan 1 to Nov 2). Compare it with Cost/cust, which is HubSpot cost per closed customer. n/a means no Meta match or 0 Meta leads.

Top campaigns by revenue

2026 YTD

2025 YTD

Top campaigns by ROAS

Campaigns with at least $5,000 spend in the period.

2026 YTD

2025 YTD

KNB National_2 - 8/26

The 8/26 ads carried the older 7/11 campaign's hsa_cam ID as a fixed number but the correct utm_id, so matching by utm_id first credits the 8/26 campaign with its own leads. The URL parameters were fixed in Meta on Nov 3, 2026 and now use dynamic macros.

Spend

$101,890

2026 YTD

Contacts

2,819

Customers

Revenue

$99,962

ROAS

0.98x

vs 1.62x all Meta

Cost per SQL

$525

194 SQLs

Cost per customer

$2,997

vs $2,633 all Meta

Installs & Builds vs Sheets Only, YTD 2025 vs YTD 2026

Closed-won deals of Meta contacts created Jan 1 to Nov 3 of each year, by deal pipeline, from the HubSpot CRM (deal counts stand in for customers).

Pipeline	2025 won	2025 revenue	2025 avg deal	2025 share (deals / rev)	2026 won	2026 revenue	2026 avg deal	2026 share (deals / rev)
Installs & Builds	287	$1,744,752	$6,079	59% / 74%	174	$1,115,968	$6,414	52% / 82%
Sheets Only	114	$364,242	$3,195	23% / 16%	72	$180,688	$2,510	21% / 13%
Miami Install	18	$130,731	$7,263	4% / 6%	1	$5,725	$5,725	0% / 0%
Miami Build	2	$37,857	$18,928	0% / 2%	0	$0	n/a	0% / 0%
Ecommerce Pipeline	66	$64,617	$979	14% / 3%	76	$57,083	$751	23% / 4%
Passing Lead to Partner	0	$0	n/a	0% / 0%	14	$0	n/a	4% / 0%
Total	487	$2,342,200	$4,809	100%	337	$1,359,464	$4,034	100%

What it showsInstalls & Builds is the revenue engine both years (75% of linked 2025 revenue, 82% in 2026). Its average deal held steady ($6,079 in 2025, $6,414 in 2026), but won deals fell from 287 to 174 so far. Sheets Only fell from 114 deals and $364,242 to 72 deals and $180,688, with a lower average deal ($3,195 to $2,510). Miami Install and Miami Build, worth $168,588 in 2025, are nearly gone in 2026 ($5,725). Ecommerce added more small orders (76 vs 66, about $750 each).

Findings

Growth story2026 brought 20% more Meta contacts but 25% fewer customers on 22% more spend, so cost per customer rose from $1,607 to $2,633. Revenue per customer also fell, from $5,451 to $4,270, so Meta revenue is down 42%. Part of the gap is time to close; through Sep 3, revenue is down 38%.

Best 2026 by campaign: Colorado - Builds - 10/29/25 ($169,004 on $27,025, 6.25x), Miami - Materials & Installs - 8/26 ($144,019 on $75,096, 1.92x), Houston - Install & Materials ($100,334 on $72,232, 1.39x), BYP Page Colorado Builds ($95,252 on $14,483, 6.58x), Colorado - Installs & Materials - 3/23/26 ($79,066 on $17,203, 4.60x).
KNB National_2 - 8/26 is the biggest spender and returns below average: 2,819 contacts, 34 customers and $99,962 revenue on $101,890 (0.98x).
Worst 2026 by campaign: CTX - Builds - 11/7/25 ($26,117, 3 customers, $0 revenue), BYP Page CTX Builds ($16,909, $0), Chicago - Installs & Materials ($13,679, $22), Utah ($9,709, 0 customers) and CTX - Install & Materials ($41,696 for $3,744, 0.09x).
Best 2025: KNB - Colorado - 10/30 ($344,350 on $34,850, 9.88x), KNB - National_2 - 7/11 ($222,226 on $78,935, 2.82x), KNB - 2/24 - Houston - Build Campaign ($197,532 on $18,983, 10.41x), KNB - Houston - 11/18 - Combined ($188,734 on $60,901, 3.10x).
Colorado is the standout market both years; Dallas and CTX consistently return less than they cost.

Caveats

Recent 2026 revenue is understated: with a sales cycle of about 6 weeks, leads from roughly mid September onward have not had time to close.
25 Meta ads have HubSpot tracking errors (19 from the old 1.91:1 image crop), which can hide contacts by campaign. KNB - National_2 - 8/26 ads carried a hardcoded hsa_cam ID of the 7/11 campaign (fixed Nov 3, 2026); matching uses utm_id first, which corrects this.
Some won deals have $0 amounts, which lowers revenue and ROAS.
Each contact counts once. The campaign comes from the first landing URL when it carries a valid ID, otherwise from later sessions or the campaign name. 850 2026 YTD contacts ($246,773) and 1,098 2025 YTD contacts ($321,802) could not be placed and sit in Unattributed Meta. They are in the headline totals but not in any campaign row, so campaign revenue is a floor.
Pipeline split: deals are tied to Meta through their associated contacts in the CRM (Facebook click ID, or original or latest source Paid Social), counted per deal rather than per contact, so pipeline totals ($1.36M for 2026, $2.34M for 2025) differ slightly from the headline revenue. Deal counts stand in for customers.
No monthly breakdown yet, so the trend inside each year is not shown.

Campaigns grouped by branch and type, YTD (Jan 1 to Nov 3)

Same metrics as the campaign table, summed across similar campaigns (date-suffixed copies and KNB variants for the same market and type roll into one row). Campaigns with no clear market are their own group. Grouping rules are in Method.

Sort by

Group	Camps	Spend	Contacts	Cust.	Revenue	ROAS	Meta CPA	Cost/SQL	Cost/cust

Matured cohort: Jan 1 to Sep 3, 2026 vs 2025

Contacts created Jan 1 to Sep 3 of each year, with revenue from their deals closed to date (Nov 3). Two months of buffer lets most of these leads close, so the comparison is fairer than YTD.

Spend sourceMeta Ads Manager exports for Jan 1 to Sep 3 of each year, all three ad accounts (Cover Your Pergola, Build Pergola, Budget Build). 2026: $623,939.90 + $33,719.43 + $2,143.19. 2025: $474,593.35 + $62,968.11 + $0.

Spend

$659,803

+22.7% vs $537,561

Revenue (all Meta)

$1,258,344

-37.6% vs $2,015,764

ROAS

1.91x

vs 3.75x

Contacts

11,764

+20.2% vs 9,785

Customers

285

-19.9% vs 356

Cost per contact

$56.09

+2.1% vs $54.94

Cost per customer

$2,315

+53.3% vs $1,510

Revenue per customer

$4,415

-22.0% vs $5,662

Meta CPL

$41.42

-5.3% vs $43.76 (15,930 vs 12,285 leads)

Contact to customer

2.4%

2025: 3.6%

Pipeline split, matured cohort

Pipeline	2025 won	2025 revenue	2025 avg deal	2025 share (deals / rev)	2026 won	2026 revenue	2026 avg deal	2026 share (deals / rev)
Installs & Builds	249	$1,549,641	$6,223	61% / 76%	164	$1,056,848	$6,444	54% / 84%
Sheets Only	93	$292,139	$3,141	23% / 14%	66	$163,414	$2,476	22% / 13%
Miami Install	16	$122,564	$7,660	4% / 6%	1	$5,725	$5,725	0% / 0%
Miami Build	2	$37,857	$18,928	0% / 2%	0	$0	n/a	0% / 0%
Ecommerce Pipeline	50	$38,890	$778	12% / 2%	62	$38,082	$614	20% / 3%
Passing Lead to Partner	0	$0	n/a	0% / 0%	10	$0	n/a	3% / 0%
Total	410	$2,041,091	$4,978	100%	303	$1,264,069	$4,172	100%

What it showsEven with two months to close, the 2026 cohort trails 2025: 20% more Meta contacts produced 20% fewer customers and 38% less revenue. Installs & Builds fell from 249 deals and $1,549,641 to 164 deals and $1,056,848 (average deal $6,223 to $6,444). Sheets Only fell from 93 deals and $292,139 to 66 deals and $163,414. So the gap is fewer closes, not smaller jobs, in the main pipeline.

Campaign table, through Sep 3

Meta spend and Meta CPA from Ads Manager for Jan 1 to Sep 3. Contacts, customers and revenue are Meta contacts created Jan 1 to Sep 3 matched to a campaign by campaign ID first, then by exact campaign name, with closed-won deals to Nov 3. Matching places 11,023 of 11,764 2026 contacts (231 of 285 customers, $1,019,507 of $1,258,344 revenue) and 8,778 of 9,785 2025 contacts (305 of 356 customers, $1,719,883 of $2,015,764). The rest are in the Unattributed Meta row.

Sort by Account Search

Campaign	Status	Spend	Contacts	Cust.	Revenue	ROAS	Meta CPA	Cost/SQL	Cost/cust

Meta CPA is Meta cost per website lead from Meta Ads Manager (Jan 1 to Sep 3), defined as on the YTD tab. Status is the Meta delivery status at export. n/a means no Meta leads or no matched SQLs.

Method

Meta contact: HubSpot contact with a Facebook click ID, or original or latest source Paid Social. Cohort by contact create date. Revenue is the amount on associated closed-won deals, closed any time up to Nov 3, 2026. Customers are contacts with at least one closed-won deal.
Revenue definition (headline KPIs, campaign tables, groups and charts on both tabs): closed-won deal amounts, closed up to Nov 3, 2026, of all Meta contacts created in the window. Campaign rows hold the contacts matched to a campaign (see the campaign matching point); the rest are in the Unattributed Meta row, which is included in the headline totals. Spend is all Meta spend from Ads Manager exports, all three ad accounts. ROAS is that revenue divided by that spend. Pipeline split tables also use all Meta contacts.
Grouping (bottom sections, both tabs): market comes from the campaign name (Colorado or Denver; Miami; Caribbean; Dallas; Houston; CTX or Central Texas; Austin; San Antonio; OKC or Oklahoma; Kansas City; Georgia; Carolina(s); CA or California; Chicago; Oregon; Utah; NW Arkansas; El Paso). Type: Installs & Materials (any install wording), Materials (materials only), Sheets, Builds (builds, build campaign, Build Your Pergola page, Budget Build), Mixed (Miami Spanish builds, installs and materials), and KNB general for KNB market campaigns with no type in the name. National, Nationwide, MOF and Video Views campaigns are not guessed: each is its own group. Miami - Caribbean sits under Caribbean. Group Meta CPA is group spend over group Meta leads; Cost/SQL and Cost/cust are group spend over group SQLs and customers.
SQL: a Meta contact (same definition as above) that entered Sales Qualified Lead (hs_v2_date_entered_salesqualifiedlead set) or is at SQL, opportunity or customer stage, counted by contact create date in each window. Campaign comes from campaign matching (next point). Cost/SQL is campaign Meta spend divided by SQLs. 1,243 of 1,370 2026 YTD SQLs and 1,046 of 1,157 through Sep 3 match a campaign row; the rest are in Unattributed Meta, so campaign Cost/SQL is a slight upper bound. 2025 shows far more SQLs (7,705 YTD) than 2026 (1,370), which points to a change in how lifecycle stages were set, so compare Cost/SQL within a year, not across years.
Campaign matching (campaign tables, groups and the Lead quality campaign table): each contact's first landing URL is read for utm_id and hsa_cam, and each is checked against the campaign IDs in the Meta Ads Manager exports for that year (Jan 1 to Sep 3 and Jan 1 to Nov 3, all three accounts). If both are valid and differ, utm_id wins (61 contacts through Sep 3 2026, 70 YTD 2026, 143 and 1,531 in 2025). If only one is valid it is used: hsa_cam was a stale hardcoded ID while utm_id was right for 2,260 contacts through Sep 3 2026 (2,826 YTD). Only when neither ID is valid is the source drill-down name matched to a campaign name (1 contact through Sep 3 2026, 3 YTD, none in 2025). If the first landing URL has no valid ID, three more steps run in order: (1) the utm_id or hsa_cam in the contact's latest landing URL or referrer (later session, not first touch), which placed 1,548 contacts through Sep 3 2026 (1,578 YTD; 318 and 433 in 2025); (2) the utm_campaign name in that latest URL (0 extra contacts); (3) the HubSpot latest source drill-down name, exact match to a Meta campaign name in that year's export, taking the highest-spend campaign if a name repeats (66 contacts through Sep 3 2026, 110 YTD; 5 and 6 in 2025). Lead form name could not be used, because the main forms (KNB Landing Page Form, Pergola Cover KNB Facebook Form) are shared by many campaigns, and numeric ad set or ad IDs in the drill-down could not be resolved without an ad set export. Contacts no step could place go to Unattributed Meta: 741 of 11,764 through Sep 3 2026 (54 customers, $238,837), 850 of 14,488 YTD 2026 (58, $246,773), 1,007 of 9,785 (51, $295,881) and 1,098 of 12,089 (59, $321,802) in 2025.
Deals with several Meta contacts count once per deal in the pipeline tables. Passing Lead to Partner deals carry no amount.

Campaigns grouped by branch and type, through Sep 3

Sort by

Group	Camps	Spend	Contacts	Cust.	Revenue	ROAS	Meta CPA	Cost/SQL	Cost/cust

Lead quality: which Meta leads turn into revenue

Meta contacts created Jan 1 to Sep 3, 2026 (11,764) and 2025 (9,785), with revenue from their closed-won deals to date (Nov 3). Revenue per lead is the number to compare: it combines close rate and deal size. For reference, Meta spend per contact in this window was $56.09 in 2026.

In service area, revenue / lead

$192

4,665 leads (40%), 71% of revenue. 2025: $365

Out of service area, revenue / lead

$46

6,336 leads (54%), 23% of revenue. 2025: $102

In service area + need an installer

$276

2,228 leads (19%), 49% of revenue ($614,222)

In service area + install myself

$33

1,242 leads, avg customer $1,237. 2025: $220

Short answerYes, the data can tell. Two form answers separate high and low value leads clearly: service area (lead_location IS vs OOS) and the install question. An in service area lead who needs an installer is worth about $276 in revenue, versus $33 to $56 for in service area DIY and out of service area leads, which is below the $56 Meta cost per contact. What the data cannot tell is why a lead went nowhere: lead status is empty on every contact and only 1 of 3,087 Meta deals has a closed lost reason.

What service area means hereIn service area and out of service area in this tab come from the HubSpot contact field Lead Location (lead_location), which holds IS (in service area) or OOS (out of service area). It is not the area targeted in Meta. The lead form does not ask it; the form asks for a zip code, and Lead Location follows the lead's state: about 99% of leads with a state have IS for TX, FL, CO or OK and OOS for every other state. It is state level, not metro level, so an El Paso lead from the Houston campaign counts as in service area. Of the 6,336 out of service area 2026 leads, 96% list a state outside TX, FL, CO and OK, 1% are inside those states, and 3% have no state. 63% came from campaigns aimed outside the service area on purpose (California, Chicago, the Carolinas, Georgia, Kansas, Utah, Oregon, NW Arkansas, plus KNB National_2, MOF and Video Views), 19% have no campaign recorded, 13% came from Dallas, Houston, Miami, Central Texas, OKC and Colorado campaigns, and 5% from Caribbean and other campaigns.

The four groups, 2026 cohort

Group (2026 cohort)	Contacts	Revenue	Avg / median per customer	In service area	Need an installer	Install myself	Build leads	2025 contacts / revenue
High-value customers (Installs & Builds won)	158	$1,059,414	$9,901 / $6,656	70%	60%	22%	28%	227 / $1,690,704
Low-value customers (Sheets, ecommerce or partner only)	128	$204,656	$3,010 / $2,072	30%	20%	51%	16%	130 / $350,387
Got a deal, not closed	859	$0	n/a	43%	44%	34%	33%	900 / $0
Never qualified (no deal)	10,619	$0	n/a	39%	45%	35%	36%	8,528 / $0

High-value customers are 70% in service area (Lead Location IS) and 60% need an installer. Low-value customers are 30% in service area and 51% install themselves. Leads that got a deal but did not close look almost the same as leads that never qualified, so the form answers predict deal size better than they predict who sales will quote.

Service area and install answer combined

Service area and install answer	Leads	Got a deal	Customers (I&B)	Close rate	Avg revenue / customer	Revenue / lead	2025 revenue / lead (leads)
Out of service area, Need an installer	2,718	7.2%	33 (17)	1.2%	$4,015	$49	$104 (1,699)
Out of service area, Install myself	2,627	9.9%	64 (22)	2.4%	$2,289	$56	$93 (3,641)
In service area, Need an installer	2,228	12.7%	80 (72)	3.6%	$7,678	$276	$430 (1,576)
In service area, Install myself	1,242	9.1%	33 (13)	2.7%	$1,237	$33	$220 (1,847)
In service area, no answer	947	10.1%	22 (19)	2.3%	$9,954	$231	$749 (438)
Out of service area, Already have an installer	547	12.1%	9 (1)	1.6%	$1,085	$18	$243 (258)
Out of service area, no answer	444	8.1%	6 (0)	1.4%	$4	$0	$0 (68)
Lead Location blank, Need an installer	332	6.6%	7 (6)	2.1%	$5,583	$118	$0 (2)
In service area, Already have an installer	248	11.3%	13 (6)	5.2%	$1,701	$89	$345 (143)
Lead Location blank, Install myself (sparse)	226	7.5%	3 (0)	1.3%	$6,175	$82	$0 (4)
Lead Location blank, no answer	128	13.3%	13 (1)	10.2%	$1,109	$113	$0 (109)
All Meta leads	11,764	9.7%	286 (158)	2.4%	$4,420	$107	$209 (9,785)

Customers column shows all customers, with the Installs & Builds (incl. Miami) customers in brackets. Revenue / lead = revenue / leads in that row. Rows marked sparse have fewer than 5 customers.

Service area (HubSpot Lead Location field)

Service area (lead_location)	Leads	Got a deal	Customers (I&B)	Close rate	Avg revenue / customer	Revenue / lead	2025 revenue / lead (leads)
Out of service area (OOS)	6,336	8.8%	112 (40)	1.8%	$2,578	$46	$102 (5,666)
In service area (IS)	4,665	11.1%	148 (110)	3.2%	$6,055	$192	$365 (4,004)
Lead Location blank	763	8.7%	26 (8)	3.4%	$3,044	$104	$0 (115)
All Meta leads	11,764	9.7%	286 (158)	2.4%	$4,420	$107	$209 (9,785)

Need a pergola built?

Need a pergola built?	Leads	Got a deal	Customers (I&B)	Close rate	Avg revenue / customer	Revenue / lead	2025 revenue / lead (leads)
No, I already have one	5,649	11.4%	177 (97)	3.1%	$3,046	$95	$192 (5,863)
Yes, I'd like a quote	3,786	7.3%	50 (34)	1.3%	$9,413	$124	$169 (2,801)
No answer	1,693	9.0%	42 (20)	2.5%	$5,268	$131	$522 (758)
Yes, but I have a contractor	636	11.6%	17 (7)	2.7%	$1,948	$52	$135 (363)
All Meta leads	11,764	9.7%	286 (158)	2.4%	$4,420	$107	$209 (9,785)

Lead type

Lead type	Leads	Got a deal	Customers (I&B)	Close rate	Avg revenue / customer	Revenue / lead	2025 revenue / lead (leads)
Build	4,209	8.3%	66 (45)	1.6%	$9,085	$142	$242 (3,073)
Product Only	4,075	10.1%	110 (42)	2.7%	$2,438	$66	$128 (4,789)
Install	3,348	10.8%	92 (69)	2.7%	$4,065	$112	$390 (1,748)
No answer	106	17.9%	15 (2)	14.2%	$1,455	$206	$0 (137)
All Meta leads	11,764	9.7%	286 (158)	2.4%	$4,420	$107	$209 (9,785)

Build leads close least often (1.6%) but bring the biggest deals ($9,085 average), so their revenue per lead ($142) beats Product Only ($66).

State

State	Leads	Got a deal	Customers (I&B)	Close rate	Avg revenue / customer	Revenue / lead	2025 revenue / lead (leads)
TX	2,451	8.3%	67 (51)	2.7%	$2,723	$74	$317 (2,236)
CA	1,585	8.8%	31 (18)	2.0%	$2,024	$40	$130 (1,340)
FL	1,348	16.9%	36 (19)	2.7%	$5,116	$137	$176 (1,071)
No answer	929	8.3%	15 (10)	1.6%	$4,619	$75	$19 (459)
CO	474	14.8%	34 (29)	7.2%	$12,707	$911	$1,071 (424)
IL	436	8.0%	6 (2)	1.4%	$819	$11	$0 (198)
OK	344	5.5%	10 (8)	2.9%	$6,380	$185	$376 (199)
GA	324	31.8%	5 (2)	1.5%	$5,792	$89	$176 (404)
NY	300	4.7%	6 (0)	2.0%	$3,780	$76	$112 (248)
NC	294	4.8%	7 (1)	2.4%	$882	$21	$106 (339)
SC (sparse)	212	2.8%	4 (1)	1.9%	$6,384	$120	$134 (187)
MO	178	15.7%	5 (0)	2.8%	$2,335	$66	$41 (119)
All Meta leads	11,764	9.7%	286 (158)	2.4%	$4,420	$107	$209 (9,785)

Colorado is the standout in both years ($911 per lead in 2026, $1,071 in 2025). California, Illinois, North Carolina and Ohio bring volume with little revenue.

Placement and platform

Placement (from ad URL)	Leads	Got a deal	Customers (I&B)	Close rate	Avg revenue / customer	Revenue / lead	2025 revenue / lead (leads)
Facebook_Mobile_Feed	1,574	10.2%	37 (29)	2.4%	$5,953	$140	$438 (229)
Facebook_Mobile_Reels	1,263	6.9%	18 (12)	1.4%	$7,505	$107	$202 (76)
Instagram_Reels	674	10.1%	13 (5)	1.9%	$4,574	$88	$324 (31)
Instagram_Feed	453	12.4%	10 (5)	2.2%	$3,454	$76	$97 (106)
Instagram_Stories	214	8.9%	5 (4)	2.3%	$10,025	$234	$5,666 (25)
Facebook_Stories (sparse)	129	4.7%	1 (1)	0.8%	$3,568	$28	$0 (17)
Facebook_Desktop_Feed	103	14.6%	9 (5)	8.7%	$5,095	$445	$0 (5)
Facebook_Marketplace (sparse)	73	16.4%	3 (2)	4.1%	$8,493	$349	$0 (9)
Facebook_Instream_Video (sparse)	47	8.5%	0 (0)	0.0%	n/a	$0	$0 (3)
Not tracked in URL	7,168	9.8%	187 (93)	2.6%	$3,579	$93	$190 (9,268)
All Meta leads	11,764	9.7%	286 (158)	2.4%	$4,420	$107	$209 (9,785)

Only 39% of 2026 leads (and 5% of 2025 leads) carry a placement in the ad URL, so read this as directional. Facebook Mobile Feed has the most volume and the best revenue per lead of the large placements ($140). Facebook Stories (129 leads, 1 customer) and Instream Video (47 leads, 0 customers) are the weakest.

Platform	Leads	Got a deal	Customers (I&B)	Close rate	Avg revenue / customer	Revenue / lead	2025 revenue / lead (leads)
Facebook	7,737	9.0%	166 (100)	2.1%	$4,559	$98	$163 (6,895)
Instagram	2,889	10.2%	62 (33)	2.1%	$5,059	$109	$261 (2,614)
Not tracked	1,138	13.3%	58 (25)	5.1%	$3,339	$170	$856 (276)
All Meta leads	11,764	9.7%	286 (158)	2.4%	$4,420	$107	$209 (9,785)

Campaign

Campaign (Meta campaign ID)	Leads	Got a deal	Customers (I&B)	Close rate	Avg revenue / customer	Revenue / lead	2025 revenue / lead (leads)
KNB - National_2 - 8/26	2,301	7.0%	31 (13)	1.3%	$2,970	$40	$39 (76)
CA - Materials Only - 12/12/25	1,191	8.8%	20 (10)	1.7%	$2,014	$34	n/a (0)
Miami - Materials & Installs - 8/26	894	15.9%	25 (12)	2.8%	$5,028	$141	$0 (26)
Unattributed Meta	741	14.8%	54 (24)	7.3%	$4,423	$322	$308 (1,001)
Dallas - Materials & Installs	734	7.1%	16 (12)	2.2%	$1,362	$30	n/a (0)
Houston - Install & Materials	687	8.7%	20 (15)	2.9%	$4,821	$140	$859 (26)
Miami - Caribbean - Install & Materials - 8/26	594	9.8%	6 (1)	1.0%	$4,696	$47	$0 (12)
Carolinas - Materials Only - 12/10	471	5.7%	9 (1)	1.9%	$3,115	$60	n/a (0)
Dallas - Builds	465	6.7%	13 (8)	2.8%	$3,412	$95	$0 (26)
OKC - Installs & Materials - 11/10/25	358	7.8%	9 (5)	2.5%	$2,724	$68	n/a (0)
Kansas City - Materials & Installs - 12/10	318	11.6%	5 (1)	1.6%	$2,135	$34	n/a (0)
CTX - Install & Materials - 2/21/26	291	7.2%	12 (8)	4.1%	$312	$13	n/a (0)
Colorado - Builds - 10/29/25	290	11.4%	14 (12)	4.8%	$12,072	$583	$6,554 (2)
Chicago - Installs & Materials (sparse)	274	7.7%	3 (0)	1.1%	$7	$0	n/a (0)
All Meta leads	11,764	9.7%	286 (158)	2.4%	$4,420	$107	$209 (9,785)

Campaign comes from campaign ID matching, then campaign name (see Method on the Matured cohort tab). Unattributed Meta holds the 741 leads no method could place; they return $322 per lead, so campaign rows understate revenue a little. KNB National_2, CA materials only, Dallas materials & installs, Kansas City and the Caribbean campaign return $30 to $47 per lead, under the $56 cost per contact.

Form and contact data

Form	Leads	Got a deal	Customers (I&B)	Close rate	Avg revenue / customer	Revenue / lead	2025 revenue / lead (leads)
KNB Landing Page Form	10,533	9.4%	237 (134)	2.3%	$4,317	$97	$400 (164)
Build Form - Miami	328	10.4%	6 (6)	1.8%	$9,939	$182	$0 (47)
Build Form - CTX (sparse)	298	7.4%	4 (4)	1.3%	$8,865	$119	$0 (125)
Build Form (sparse)	228	8.8%	4 (4)	1.8%	$22,106	$388	$832 (319)
Metform Elementor Form Builder	97	39.2%	9 (3)	9.3%	$2,310	$214	$684 (28)
All Meta leads	11,764	9.7%	286 (158)	2.4%	$4,420	$107	$209 (9,785)

Phone

Phone on the contact	Leads	Got a deal	Customers (I&B)	Close rate	Avg revenue / customer	Revenue / lead	2025 revenue / lead (leads)
Valid phone	11,611	9.8%	285 (157)	2.5%	$4,435	$109	$210 (9,439)
Missing or invalid phone (sparse)	153	4.6%	1 (1)	0.7%	$0	$0	$159 (346)
All Meta leads	11,764	9.7%	286 (158)	2.4%	$4,420	$107	$209 (9,785)

Budget

Budget answer (only about 8% answer)	Leads	Got a deal	Customers (I&B)	Close rate	Avg revenue / customer	Revenue / lead	2025 revenue / lead (leads)
Not answered	10,843	9.8%	267 (142)	2.5%	$4,035	$99	$189 (9,242)
$5,000+	586	7.2%	9 (7)	1.5%	$4,232	$65	$307 (305)
$10,000+	250	13.2%	7 (7)	2.8%	$14,474	$405	$197 (157)
$20,000+ (sparse)	85	10.6%	3 (2)	3.5%	$15,776	$557	$2,402 (71)
All Meta leads	11,764	9.7%	286 (158)	2.4%	$4,420	$107	$209 (9,785)

Bad phone numbers are rare (153 leads in 2026) and only 1 of them bought, so junk contact data is not the main quality problem. Budget and timeline are answered by only about 8% of leads; $10,000+ and $20,000+ budgets show higher revenue per lead but on 3 to 7 customers.

What to change in Meta

Targeting and exclusions: put most budget behind the zips and metros you install in (zip or radius targeting around the install markets), and exclude other regions from the install and build campaigns. Out of service area leads are 54% of volume but return $46 per lead. Dallas, Houston, Miami, Central Texas, OKC and Colorado campaigns still drew 766 leads from other states in 2026 (Houston: 157 of 312), so check that location targeting is limited to people living in those areas. If product-only sales outside the area are still wanted, run them as a separate, capped campaign judged on Sheets and ecommerce revenue.
Cut or cap weak campaigns: KNB National_2 ($40 per lead on 2,301 leads), CA materials only ($34), Dallas materials & installs ($30), Kansas City ($34) and Caribbean install & materials ($47). Keep Houston install & materials ($140) and Miami materials & installs ($141), and grow Colorado builds ($583 per lead).
Placements: favor Facebook Mobile Feed; test removing Facebook Stories and Instream Video. Add utm_placement={{placement}} to every ad so all leads carry a placement.
Form qualifying questions: zip is already required on the main landing page form; make the install question required on every form, and add budget and timeline as required questions with the Higher intent form type. Route in service area DIY answers to a product flow rather than the install sales team.
Value-based events: send HubSpot stages back to Meta through the Conversions API: a Qualified Lead event when a deal is created and a Purchase event with the deal amount when it closes won. Then optimize campaigns for qualified leads or value instead of raw leads.
Lookalikes: build a value-based lookalike from the 385 Installs & Builds customers (227 in 2025, 158 in 2026, $2.75M combined), and exclude existing customers and out of service area contacts from the seed.
CRM hygiene so quality can be measured next time: start filling Lead Status (junk, unreachable, not a fit) and Closed Lost Reason. Today they are empty, so we cannot say why 10,619 of 11,764 2026 leads never got a deal.

Method and caveats

Meta contact and revenue definitions match the Matured cohort tab (customers 286 in 2026 and 357 in 2025; revenue $1,264,069 and $2,041,091). High-value customer: at least one closed-won Installs & Builds or Miami Install/Build deal. Low-value customer: closed-won Sheets Only, ecommerce or partner deals only. Got a deal: any associated deal in any pipeline.
Groups use "got a deal" instead of the SQL stage, because 2025 lifecycle stages were set differently (6,168 2025 leads carry an SQL date but sit at Lead with no deal).
Revenue is deal amount, not margin. Averages are pulled up by a few large jobs (2026 high-value average $9,901 vs median $6,656). Small rows (under 5 customers) are flagged sparse.
Service area is the HubSpot Lead Location field (lead_location, IS or OOS, filled on 94% of 2026 leads), which tracks the lead's state, not the Meta targeting area; placement and campaign come from the first page URL and source drill-down, and are partly missing. This shows correlation, not proof that a change will move revenue by the same amount.

Sources: Meta Ads Manager campaign exports with campaign IDs (Jan 1 to Sep 3 and Jan 1 to Nov 3, both years) joined to HubSpot contacts and closed-won deals by campaign ID, then campaign name. Headline KPIs cover all Meta contacts, including Unattributed Meta. Meta CPL and CPA from Meta Ads Manager.