The pipeline: September 1-17 vs August

Google Ads-attributed only, Local Services excluded, all other campaign types included so both columns sit on the same basis. Leads = ServiceTitan inbound calls (extension + website) + lead forms tied to an ad click. Booked = ServiceTitan booked jobs from those clicks. September covers the 1st to the 17th, including Labor Day weekend and the holiday pauses, and recent booked jobs still upload for days after the call, so the most recent days read high before they settle.

*Power Plumbing's numbers are the known August tracking gap, not performance; its rebuild is deliberately throttled until the conversion feed completes (budget stepped up Sep 7). September standouts: Williams Sewer & Drain at $54 per booked job on a small budget, and Forsyth at $112. 2 Sons has come down to $310 while carrying the most volume. Pride sits high at $915, which is the geography cut and the promotion both landing mid-month, so it reads properly at the end of the month rather than now.

Sep 28: Local Services areas matched to where the jobs are, Universal trimmed, Forsyth cleaned up

• Local Services service areas now cover the zips where the trucks already go. Every zip added had 10 or more ServiceTitan jobs in the last six months, sat outside the Local Services area, and was already a PPC target. Forsyth Rooter: Clemmons, Advance, Lewisville and six more (577 jobs). Rolland Reash: the close-in zips around Orange Park, Fleming Island, Middleburg, Green Cove, Ponte Vedra and Saint Johns (391 jobs); St. Augustine, Fernandina, Yulee and Callahan stay excluded, in line with the June and July refinement. 2 Sons: Tukwila, Kenmore and Woodinville on Seattle (166 jobs); Gig Harbor, Sumner and Tacoma on Auburn (101 jobs). Anchor: Mico on Helotes. The full list is on the LSA geo analysis tab. If any of these were left out on purpose, say so and we put them back.
• Universal's far zips are out, per Shalanda and Lauren: six zips removed from the Local Services area and all eight excluded, with the same zips cut from PPC.
• Forsyth non-brand cleaned up. Thirteen new negative keywords (competitor names, installers, soil testing, Thomasville) and four keywords paused that had spent without booking a job in 90 days. Non-brand runs Maximize Conversions without a target.
• Three PPC zip adds: Mico and Canyon Lake at Anchor, Starke at Rolland.
• The online scheduler was test-booked end to end at Rolland, Pride and Anchor today. All three created the booking in ServiceTitan on the right campaign and business unit; the tests were dismissed.
• Schrader's call-only campaigns in the second account are paused. They used the main ServiceTitan number, so nothing they drove could be credited.
• Local Services ads now run business hours at Anchor Helotes, Anchor New Braunfels (Monday to Friday, 7 to 6) and 2 Sons Auburn (Monday to Saturday, 7 to 5). Overnight leads there booked at 11 to 14% against 32 to 45% in the day. Universal and E.R. keep overnight coverage; Pride, Forsyth, Power and Sewer Lines Only already ran business hours. 2 Sons Seattle stays as is for now.
• Missed calls from Oct 1. Google will charge a missed Local Services call during business hours when the caller stays on the line past 20 seconds; after-hours calls are not covered. It looks like the direct cost is small, about $300 to $700 a month across all accounts, because most unanswered callers hang up inside 20 seconds and the longer ones are mostly charged already as voicemails. Most exposed: Anchor Helotes (misses 17% of calls, about $230 a month), Pride Mount Gilead (14%, about $220, mostly 7 to 8am and Saturdays) and E.R. Plumbing (about $160, an estimate until we see its call log). We check the first two weeks of October against these numbers.
• Answering the phone lines up with rank. The accounts that answer 91% or more of Local Services calls (Sewer Lines Only, Universal, 2 Sons Seattle) take the first ad spot about 58% of the time; the ones under 85% sit at 9 to 22%. Across the accounts, the share of calls answered tracks first-spot rate and the share of leads marked booked more closely than review score does. Where it looks like answering is holding volume back: Anchor Helotes (82 to 84% of calls answered, 44 leads never opened), 2 Sons Auburn (41% of message leads replied to; its first spot fell to 21% while Seattle holds 60%) and Pride (misses cluster 7 to 8am, after 5pm and Saturdays; Southern Pines replies to 38% of messages). We'll re-read the scorecard monthly and flag changes here.
• Two Local Services accounts need a new form of payment: E.R. Plumbing and Pride Southern Pines, per Google's account alerts. Until it is updated, the ads can stop. Rolland's billing email also needs verifying so Google's billing notices reach someone.

Sep 21: Anchor's scheduler live, all 16 Local Services accounts open, and a new Local Services tab

• Anchor's online scheduler went live Thursday. Eight real bookings across the live pages since Sep 11, six of them outside business hours. Two came in at 2 Sons over the weekend; one was booked by hand from a call the customer made herself on Sunday, so it sits under the phone line instead of the campaign that produced it. The fix is process: work the Bookings inbox and book from the booking record, not from the call.
• Every Local Services account is now linked, 16 of 16. The last five invitations were accepted over the weekend, thank you. There is a new Local Services tab and a certifications and inventory page: every licence and policy with days left, Google's pass or fail per service area, and what each account is set up to sell, where and when.
• Pride Southern Pines' insurance is past date with a replacement in review (expired Jul 28, new policy submitted Sep 16, and the ad stopped serving the week of Sep 14). Rolland Reash checked out clean in the account. Universal's licence is due Nov 28; E.R. Plumbing's three NC licences and Pride Mount Gilead's are due Dec 29 to 30. Renewals go in 60 days ahead from here.
• On lead ratings, a correction to what we thought. About 45% of charged leads in the last 30 days already carry a rating from P3, most of them on declined leads, and that earned 3 credits. Rating is happening; it is not paying. We submit a small set of well-formed ratings this week and read Google's decision on each before deciding how much of a weekly pass is worth anyone's time.
• Six accounts switch Local Services off nights and weekends (Power, Pride Mount Gilead, Forsyth, Fischer, Sewer Lines Only, Sewer Lines Plus). Same after-hours demand the scheduler is catching. Worth a decision per opco.
• Power Plumbing is paused as of Thursday, per Natalie.
• Five accounts carry call extensions at the ad-group level as well as the campaign level; ServiceTitan's documentation says the ad-group ones stop its campaign-level call tracking from working. To check with Natalie before we change it.

Sep 17: Williams and Universal connected, the scheduler starts paying, Schrader stepped up

• Williams is fully connected. Reporting, dynamic call tracking on both sites, a dedicated map-listing number with FranchiSEO, and the online booking connection ready. It went from the least measurable account to the same footing as the rest in a day.
• Universal's tracking is restored. Marketing Pro is on, the Google Ads integration is reconnected after sitting in an error state, and call tracking is live on the landing page. The one remaining piece is the tracking script on their main website, which is with Jane.
• The online scheduler has produced six real bookings in nine days, all of which became jobs. Five have completed and the invoiced total is about $8,200. Four of the six arrived outside business hours. Schrader and Anchor are next.
• Schrader's non-brand budget stepped to $600/day. It was holding a third of its available searches with another third behind the cap, at the cheapest clicks in the portfolio.
• Pride's new-customer promotion is live on the page and in the ads. The promotion line is out-performing the lines it replaced on click-through by a wide margin; whether it moves booked jobs needs another two weeks.
• Map-listing numbers are ringing. After FranchiSEO corrected the phone fields, 2 Sons, Power Plumbing and Schrader have all started taking calls on their tracking numbers. The remaining brands get re-checked Sep 22.
• 93 new negative keywords went in across five accounts after reviewing what the new general-intent keywords were actually matching.

Sep 8: Changeover live, wave-4 pages shipped, September scoreboard up

• The GBP changeover is done. FranchiSEO added all eight tracking numbers this morning and confirmed. Map-listing calls now credit to the ads; a daily watch covers every number.
• All four wave-4 landing pages are updated and deployed with every note from Friday, and the ad copy revisions are marked on their cards. Cottrell's copy is deliberately on hold (Josh runs his own).
• The Scoreboard tab now shows the September pipeline against August: leads, booked jobs, and cost per booked job per account. Early standouts: 2 Sons and Schrader both at $259 per booked job.
• The holiday pauses ran as requested: Pride, 2 Sons, and Power Plumbing down through the weekend, back on Tuesday morning.
• Power Plumbing's tracking wait is explained and healthy: calls are confirmed flowing into ServiceTitan, and the final conversion linkage completes automatically with the first ad call on the new number now that its budget stepped up.

Sep 3: Numbers to FranchiSEO, September budgets, the negatives wave

• The GBP tracking numbers are done and with FranchiSEO. All eight are live in ServiceTitan, one per brand, each on its own attribution campaign, and the full list went to Mackenzie with the placement instructions and a request for one changeover date. Multi-listing brands get the same number on every listing (Pride 3, 2 Sons 4, Anchor 2). Once swapped, calls from paid map listings credit to the ads instead of the profiles.
• September budgets are live per Natalie's approvals: Forsyth $120/day, Schrader $350, Piedmont $25, Anchor $700, Rolland holding at $800. Verified in-account the same afternoon.
• A new competitor-name negatives wave went in across 2 Sons, Forsyth, Piedmont, Anchor, and Schrader, pulled from the competitor report: Roto-Rooter, Beacon, ARS Rescue Rooter, Benjamin Franklin and the rest of each market's names no longer trigger your ads.
• Scheduler groundwork is finished. The reporting connection now carries booking and scheduling-capacity permissions in all six tenants. One end-to-end test booking runs with Nikki once the version-two pages are staged.
• Power Plumbing's bidding config re-verified clean: both enabled campaigns bid on Book Appointments and nothing else. The open item is on ServiceTitan's side (the call-extension booked-job action still is not created); it is with their support and we check it automatically four times a day.

Sep 2: Ad copy for every brand, plus the competitor deals report

• The Ad copy tab now covers the whole portfolio. The standard Natalie approved for 2 Sons, written for all twelve remaining brands: 10 headlines and 4 descriptions each, every claim sourced from that brand's own site and landing page, every line pre-checked against Google's character limits, and unconfirmed promos held out and flagged per brand. Review brand by brand; we implement each account as it's approved.
• New tab: Competitor intel. Fresh auction insights on 2 Sons, Pride, and Rolland Reash, screengrabs of the actual ads competitors are running, and every published deal, membership plan, and financing offer in each market. Highlights: 2 Sons is the only major Seattle player without an offer, Pride's market has almost no deal pressure at all, and Jacksonville competes on flat entry prices and financing. Once the copy cleanup lands and the data cleans up, this feeds the per-market promo strategy.

Aug 31: August close: the surgery month

• August was the month we rebuilt the machine while it was running, most of it inside a three-week window: landing pages live with verified call tracking, the ServiceTitan connection covering every brand, bidding moved to booked jobs only, click-price governors on every campaign, roughly $10,000/month of non-booking search spend cut, ad copy rebuilt and staged for approval, and the GBP tracking numbers created. That much change mid-flight makes August a messy month to read, on purpose.
• Where it already shows: Anchor booked a third more jobs in August at 27% lower cost per job on flat spend, and Forsyth held its cost per booked job exactly steady while we cut about a fifth of its spend, which says the trimmed spend was pure waste.
• September is the first clean read. Every account now measures the same thing (booked jobs), bids on the same thing, and blocks the same waste, so efficiencies should start showing, and anywhere that stays soft will stand out quickly instead of hiding in noise. We read it weekly and adjust as it settles.
• Landing in September: the new ad copy is already live on 2 Sons and rolls out portfolio-wide on your approval, FranchiSEO adds the tracking numbers to the profiles so paid map calls get credited correctly, the Power Plumbing non-brand rebuild goes live with its conversion feed completed, the version-two landing pages add live booking, and the new accounts get the same treatment the core five got in August.

Aug 26 morning: Power Plumbing live on its new landing page

• Power Plumbing's search campaigns now point at help.powerplumbingtx.com. Manager access was accepted this morning (thank you, Farrukh) and the swap went in right away: brand and all four non-brand service groups send clicks to the new page with verified dynamic call tracking, and the sitelinks on those campaigns were swapped to the new page at the same time.
• Schrader's manager access was accepted too. Its landing page stays staged for now and goes live with its onboarding wave.

Aug 24 evening: Power Plumbing first wave in, ServiceTitan connected everywhere

• Power Plumbing's first wave of fixes went in this afternoon, per our alignment: the Performance Max campaign is paused (its spend was going to Display and YouTube placements whose "conversions" were call-button taps, not connected calls), conversion counting now reports only real outcomes (reported numbers will look lower before they look better, because they are now honest), brand is on the same manual caps as the rest of the portfolio, and non-brand has a brake on its cost per click with the full negative keyword stack attached. Watching daily; the non-brand rebuild is next.
• ServiceTitan reporting now covers every brand. The Texas tenant (Power Plumbing and Anchor) and Schrader connected today on top of the original four, so same-day booked-job reads run portfolio-wide, with brands split correctly where accounts share a call center (Pride's family, and the Texas pair). This is the foundation the live dashboard builds on this week.
• Geography cleanup from a portfolio-wide sweep: two stray international postal codes that predate our access came out of the Pride and Anchor targeting, and three leftover mobile bid boosts (Piedmont, Anchor) were cleared so the manual caps stay flat everywhere.
• The five new help domains are with Jane (Forsyth, Piedmont, Power Plumbing, Anchor, Schrader). Forsyth and Piedmont launch on your approval from this afternoon as soon as the records land; your landing page notes (2 Sons logo, answering-hours wording) are being applied now.

Aug 24: first read on the new landing pages, straight from ServiceTitan

• Calls straight from the ads have always been the workhorse in these accounts, and they stay fully tracked. The new landing pages open a second front. Leads arriving with the full ad-click ID attached nearly tripled in the first four days (11 of 36 Google Ads leads versus 4 of 38 across the same four weekdays before; Rolland went from zero to 38%, Pride from 4% to 21%). Those click-tagged leads are the ones that tie jobs, and soon job values, back to the exact keyword and page, which is what the optimize-toward-revenue phase runs on as budgets scale.
• Booked jobs held and edged up through the switch (16 to 19 across the same matched windows), so the cutover cost nothing while the tracking gain landed.
• The 2 Sons investment is doing what it was approved to do: since the budget step-up on the 18th, ServiceTitan booked jobs are running about 2.8 per day against a 1.8 per day baseline before it, with the month landing on the agreed number.
• These reads come straight from ServiceTitan same-day, not from ad-platform imports, which is the new reporting standard here: what the CRM says, when it says it.

Aug 20-22

• Forsyth and Piedmont audited and standardized (Aug 22). Both accounts are now on the program: brand on manual caps (Forsyth's brand was uncapped on value bidding at 98% impression share, so expect its brand spend to drop meaningfully, which is the point) and bidding steering on booked jobs only, the same standard as the rest.
• Five more landing pages built and staged: Forsyth and Piedmont (awaiting your go-ahead), and Power Plumbing, Anchor, and Schrader, ready to launch with onboarding the day account access lands. Previews on the Landing pages tab.
• A larger negatives cleanup pass is staged behind the routing decisions on the Waiting on / blocked tab (Pride septic, 2 Sons commercial boilers, the converting competitor-name terms). Everything is drafted and releases the day the answers land.
• Rolland picked up a dedicated slab-leak page at /leak (Aug 20), and its per-session number swap is verified live on both pages.

Aug 19

• Rolland's bidding basis tightened to booked jobs only, the same standard as Pride and 2 Sons. Early-August volume was thin enough that supplemental lead signal helped the algorithm learn; booked-job volume has roughly doubled since, so bidding and reporting now run on the single strictest outcome. Reported conversions and CPA shift to that stricter basis from today.
• Local Services Ads are out of our reporting everywhere. The LSA campaigns at Pride and Rolland stay live and untouched, but their spend and leads are not part of the program we manage, so they no longer appear in any number we report, on any account.
• Full conversion reconciliation completed across all three accounts using the new ServiceTitan API connection: brand is on manual bidding everywhere at 93-95% impression share, non-brand is on Maximize Conversions with no limiting targets, and 2 Sons + Pride count booked jobs only (verified down to the per-campaign level).
• Negatives wave 3 live in all three accounts: 28 new phrase negatives from a fresh scan of the last ten days, covering roughly $1,100 of spend with zero booked jobs behind it. Named competitors (including the window's single biggest waste term, a competitor name at $327 in 2 Sons), equipment and cost-shopper queries, out-of-area geographies, and a brand-spelling leak routed back to the brand campaign. In-area city searches stay untouched, as always: they are demand, not waste.
• We are working directly inside ServiceTitan now. With the API connection live on all four tenants, an initial reporting view built straight from ServiceTitan data is in progress. That shortens the feedback loop meaningfully: we read jobs and outcomes at the source, same day, instead of waiting for conversions to sync into Google Ads.
• Landing pages are LIVE, call-first, on three brands: help.callprideplumbing.com, help.plumbingjacksonville.com, and help.2sonsplumbing.com. Every button dials the company's real office line today, the ServiceTitan tracking snippet ships on every page, and per-session call tracking switches on with one setting per tenant (adding the new addresses to the call-tracking config). Cottrell's two pages (brand + /tankless) are built and deployed and launch as soon as Josh's photos land. This is the primary focus in action: drive cost per lead down while quality holds. The book-a-job form is phase two, tomorrow's work.
• Target CPAs on non-brand are the next control layer, deliberately not yet applied. With every account now bidding on the strictest outcome, targets are what pull click prices down. We are holding until the landing pages are live, because they should bring cost per lead down and a target set from today's numbers would be calibrated to economics we intend to change. The priority order: pages live, cost per lead down while quality holds, then targets that lock in a falling cost per booked job.

Aug 17-18

• 2 Sons is funded to $25K for August (approved Aug 18), and the non-brand search budget stepped up from $450 to $800/day the same morning. That campaign had been losing half or more of its weekday impressions to budget while brand sits capped near $100/day, so the increase flows entirely to non-brand demand. The measuring stick, in Natalie's words: leads and booked jobs rise with the spend. Pre-increase baseline is about 1.8 booked jobs a day (ServiceTitan-imported); we check it with every pacing look.
• 2 Sons PMax retargeting paused (Aug 17): roughly $1,250 over two weeks with near-zero booked jobs. Those dollars now back non-brand search instead.
• Heating service-intent negatives live at 2 Sons (Aug 17): 13 terms (hvac, furnace, AC and family) per Natalie's note that 2 Sons no longer offers heating and air. The negatives shared set now holds 431 terms.
• Upward bid modifiers removed in every account (Aug 17): mobile boosts of +10 to +25% and 13 Seattle-area geo boosts were quietly inflating the manual brand caps. Caps are now flat: exactly $8 / $10 / $10 on every device and geo.

Aug 12

• Brand bidding is now on manual caps in every account, checked daily. Buying brand impression share as cheaply as possible is half the program; this locks it in.
• Pride campaign goals corrected: website-originated booked jobs now count toward bidding (they previously did not on this account).
• Cottrell's two landing pages added to the preview row and sent to Josh: the brand page and the water heater page (his top ramp-up category). His photos and edits fold in as they arrive.
• The held service-line questions resolved against the OpCos' own websites instead of a back-and-forth: one negative added (Spanish-language searches at 2 Sons), everything else confirmed as real demand and left serving.
• DNS is in for all three landing page subdomains (Jane, Aug 11), and the duplicate call-tracking plugin on Pride's site is deactivated. Landing page launch now waits only on the ServiceTitan tracking wiring.
• The hub itself reorganized: tabs, the portfolio goal up top, and a Waiting on / blocked view so dependencies are visible at a glance.

Aug 10-11

• Second negatives wave live across all three accounts, every term checked against the OpCo overview first; the service-line questions were later resolved against the OpCos’ own websites (row 6).
• Pride and 2 Sons brand campaigns moved to manual caps ($8 and $10).
• Pride and Rolland landing pages updated with Natalie's feedback.

In motion

• A small set of well-formed lead ratings on Universal, with Google's credit decision read on each, before any weekly pass is sized.
• Renewal notes to the opcos with documents in review or due inside 120 days.
• A bid and target clean-up list per account for Natalie's approval, ahead of the move into Google Ads (manual bidding ends; one target per campaign).
• First monthly Local Services table per brand, Oct 5: charged, credited, declined share, message share, ServiceTitan jobs, spend per job.