Theme this issueThe last few issues focused on what happens on the truck and in the sales conversation — pricing discipline, answering the phone, keeping the financing close clean. This week the stories move up a level, to the structural forces quietly setting the ceiling on what any well-run shop can earn: the refrigerant your installs now depend on, the Google pipe most of your leads come through, the capital deciding what your business is worth, the labor market deciding whether you can staff the work, and the grid deciding how long storm demand lasts. Three of these carry hard deadlines. Google is folding Local Services Ads into Google Ads right now — and your historical performance data disappears when it does, unless you download it first. The R-454B transition has settled into a permanent 10–20% cost layer on system replacements, with installs still running weeks long in tight markets. And private equity keeps buying the vertical at multiples that can be double what an owner-dependent shop fetches — which is worth knowing before you get the call, not after. Wrapped around all of it: a skilled-trades shortage that isn't easing, and an Indiana derecho whose lights still weren't fully back on nearly two weeks later — a reminder that the restoration and standby-power money runs long after the storm leaves the news.
Story 1
Google is folding Local Services Ads into Google Ads — download your history before it vanishes
Local Services Ads → Google Ads / Performance Max migration · PPC.land, Search Engine Journal, MediaPost
The single most important administrative task on your desk this month has a countdown on it. Google is retiring the standalone Local Services Ads dashboard and converting every LSA campaign into a specialized Performance Max variant inside Google Ads, built around the same pay-per-lead billing and Search/Maps placement — but stripping out the manual controls contractors rely on. Phase 1 began in August 2026 for select US home and storefront service categories: plumbing, HVAC, electrical, roofing, appliance repair, house cleaning, lawn care, pest control, and moving. Service-area businesses and custom-bidding accounts follow in late 2026, non-US accounts in 2027. Account admins get an email notice 14 days before their individual migration date. The catch that will hurt shops who aren't paying attention: campaign-level performance history does not carry over — impressions, clicks, weekly spend, and your historical reports vanish once the account transitions, and Google is explicit that advertisers must manually download their historical data ahead of the switch. Lead history (contact details and message threads) does transfer. Also going away: manual max-cost-per-lead bidding, weekly budgets (they convert to daily), and BBB callouts.
Why it matters: LSA is the most cost-efficient lead source most home-services shops have, so a forced change to how it's bid and measured is not a "marketing" footnote — it's a revenue-pipe change. Two moves, one urgent and one strategic. Urgent: export your full LSA performance history now — cost-per-lead by category, seasonal booking patterns, spend-to-revenue — because once you're migrated it's gone, and that baseline is exactly what you need to tell whether the new Performance Max setup is helping or quietly costing you more per booked job. Strategic: losing manual per-lead bid caps means Google's automation now decides what you pay per lead, so you'll need to split campaigns by service category to keep any control over where budget flows (an emergency AC call and a tune-up are not worth the same lead price), and watch your blended cost-per-booked-job weekly through the transition rather than trusting the platform to optimize for your margin. If you run LSA through an agency or your marketing partner, confirm this week who is responsible for the data export before someone's 14-day clock runs out.
Sources (3)
Story 2
The R-454B transition is now a permanent cost layer — and installs still run long
A2L refrigerant transition, supply and pricing · ACCA HVAC Blog, Facilities Dive, Contracting Business
The refrigerant changeover has stopped being a headline and become a fixed condition of doing business. Since the AIM Act forced the industry off R-410A to lower-GWP A2L refrigerants (primarily R-454B, with R-32 gaining ground as an alternative), the transition has done two durable things to your cost structure. First, supply stayed tight: demand for R-454B ran an estimated 5x–10x above manufacturer forecasts in the first months, cylinders had to be redesigned for higher pressures (a full batch reportedly failed testing), and at the worst point roughly half of contractors were improvising to finish jobs. Manufacturers responded — Honeywell pledged to double capacity and lifted blending/down-packing ~44%, Arkema entered the market, Hudson built larger cylinders — and availability has improved, but in tight markets new-system installs are still quoted with multi-week lead times. Second, and permanently, the new equipment and refrigerant have added roughly 10–20% to the cost of a system replacement versus the R-410A era. R-454B's high glide (it separates when moved and loses its composition) also makes the old shortcut — topping off or mixing refrigerants — a warranty and reliability trap that industry leaders are explicitly warning against.
Why it matters: This reframes two conversations. With customers, "the AC costs more than it used to" is now a true, external explanation you can give without touching your markup: federal refrigerant regulation changed the equipment and the chemistry, and that ~10–20% is industry-wide, not you. Lead with that honestly and it does the same work the antitrust story did last issue — it moves sticker shock off your shoulders and back onto value and financing. Operationally, protect yourself against the lingering supply tightness: quote replacement jobs with realistic lead times rather than promising next-day and eating the reschedule, keep a working buffer of R-454B through peak without panic-buying (panic-buying is what perpetuated the shortage), and make sure every tech understands that mixing or topping off with R-454B is off the table — one glide-related callback erases the margin on the job. If you haven't already, price and stock for the possibility that some of your line will shift toward R-32 systems, since the market hasn't fully settled on a single A2L winner.
Story 3
Private equity is still buying the vertical — and what your shop is worth depends on how it's built
Home services M&A multiples and PE roll-ups · CT Acquisitions 2026 M&A Multiples Report
If you own the business, this is the story that decides your number. Private equity's share of home-services deals keeps climbing — PE-backed HVAC transactions rose from ~8% of deals in 2023 to ~23% in 2024, and some trades are already majority PE-owned (pest control is ~60%) — because buyers want fragmented markets with recurring or replacement-cycle revenue. The result is a wide, well-defined valuation range in 2026: HVAC businesses trade from about 3.0x EBITDA at the low end to 10x at the high end (mid-market ~5.5x–7.0x), with plumbing ~2.4x–6.5x, roofing ~2.5x–7.0x, and electrical ~3.2x–8.0x. The spread inside each trade is not mostly about size — it's about how the business is built. A $750K-EBITDA owner-operated HVAC shop typically clears 3x–4.5x; the same trade at $2M+ EBITDA with strong service-agreement revenue, management depth, and multiple locations regularly reaches 7x–10x. Buyers pay up for de-risked, recurring cash flow — one cited example: a $3M-EBITDA HVAC business with 45% service-agreement revenue and a real management team drawing 8x–10x bids. Geography stacks on top: Sun Belt operators (TX, FL, AZ, GA, NC) command 1.5x–2.5x premiums over comparable Midwest shops, and no-income-tax states improve the seller's after-tax take.
Why it matters: Whether or not you ever plan to sell, the same attributes that raise your multiple also make the business easier and more profitable to run — so building toward them is not just exit prep. The three levers that move you from the 3–4x tier toward 7–10x are the ones inside your control: grow recurring service-agreement revenue (it's the single biggest premium driver — a maintenance base converts one-time customers into a subscription that survives you), reduce owner-dependence by putting real managers and documented processes between you and daily operations (buyers discount a business that stops working when the owner is on vacation), and clean up your books now — trailing-twelve-month financials a buyer can trust, add-backs documented, working capital normalized. Even if you're a decade from selling, know your rough number before a PE platform calls, because roll-up acquirers are disciplined and a well-prepared seller negotiates from data instead of hope. Close timelines run 90–120 days from signed LOI, with 10–30% rollover equity common on platform deals — meaning you often stay in and get a second bite, which makes the quality of the acquirer as important as the price.
Story 4
The trades shortage isn't easing — staffing is now a competitive moat
Skilled-trades labor and wages · ABC, BLS, NCCER, JobStack
The labor math under every one of the stories above is getting tighter, not looser. Industry estimates put the construction workforce short roughly 439,000 workers in 2025 on top of normal hiring, with the BLS projecting about 649,000 average annual openings across construction and extraction trades through 2034. The demographic pressure is the real problem: NCCER estimates ~41% of the current construction workforce will near retirement by 2031, and the pipeline of new entrants isn't backfilling it fast enough. Wages reflect the squeeze — as of May 2025, plumbers earn a ~$63,800 median (highest of the major trades), electricians ~$63,190 (top 10% over $108,510), and HVAC techs ~$61,010, with wide geographic spread (electricians clear ~$101,000 in Oregon vs. ~$49,000 in the lowest-paying states). The through-line: skilled labor is the genuine constraint on how much of 2026's available demand — storm restoration, refrigerant-driven replacements, LSA leads you're now paying more per — you can actually convert into completed, invoiced jobs.
Why it matters: In a market where demand is available but labor is scarce, the shops that win aren't the ones with the most leads — they're the ones that can staff the work those leads create. That makes recruiting and retention a margin issue, not an HR chore. Practically: treat your best techs as your scarcest asset — a departing senior tech takes callbacks-avoided expertise and often customer relationships with them, and replacing one at today's wages plus ramp time costs far more than retaining them, so audit your top performers' pay against the numbers above before a competitor does it for you. Build a pipeline instead of reacting to vacancies — apprenticeships, trade-school relationships, and a real path from helper to lead-tech are how you beat wage-only bidding wars, because the strongest candidates weigh growth and culture, not just the hourly rate. And recognize that every other efficiency play in this issue — AI phone answering, tighter scheduling, service agreements — partly exists to stretch the techs you already have across more revenue. In 2026, the constraint isn't finding work; it's having the hands to do it.
Story 5
Ten days without power: the Indiana derecho shows how long restoration demand really runs
August 11 derecho aftermath · AccuWeather, NWS Chicago
The storm we flagged last issue is the story that keeps teaching. The August 11 derecho that raked northwest Indiana (99 mph gust at Gary) knocked out power to about 374,500 customers — but the headline number undersold it. Nearly two weeks later, more than 28,000 were still without electricity, with full restoration in Gary and Portage not expected until August 25. The reason it dragged: the storm snapped 1,230 distribution poles — against roughly 127 poles broken in previous major storms combined — plus dozens of transmission poles, and a transmission-equipment failure days into the recovery set crews back further. Residents rode out 80-degree heat with no A/C for over a week, some sleeping in vehicles for air conditioning, while schools closed pending structural safety checks. This is the same NWS Chicago footprint that already logged a June 10 derecho and a July 27 severe event — a relentless inland convective season while the Atlantic stays quiet.
Why it matters: The lesson for operators is that restoration demand has a long tail — the money doesn't leave when the news crews do. Two weeks after a derecho, the work isn't just roofs and downed trees; it's prolonged-outage demand: standby generators, whole-home surge protection, refrigerant lines and condensers damaged by debris, and water intrusion from a week of exposure that shows up as mold and drywall work later. Three moves if you serve or can reach the corridor (or any storm belt this season): market standby-power now, because a homeowner who just spent ten days in the dark is the most motivated generator prospect there is and that window closes as the memory fades; document everything to insurance-grade (dated photos, moisture readings, measurements), since carriers in a heavy 2026 loss year are scrutinizing scopes and the well-documented supplement is the one that gets paid; and compete on local trust against the out-of-town storm-chasers who flood these zones — your license, reviews, and local standing are the edge. For everyone outside the footprint, note the pattern that's held all summer: severe convective storms, not hurricanes, are where the restoration demand and the insurance friction both live this year — and the demand outlasts the storm by weeks.
Sources (3)
This issue's to-do list
Do this before month end
- Export your Local Services Ads history this week — Google is migrating LSA into Google Ads (Performance Max) with only a 14-day notice, and campaign-level performance reports do not carry over. Download your cost-per-lead, spend, and booking history now, confirm who owns that task if an agency runs your ads, and plan to split campaigns by service category since manual bid caps are going away.
- Reset customer and scheduling expectations for the refrigerant reality — the R-454B transition added ~10–20% to replacement cost permanently and installs still run long in tight markets. Use the regulation as the honest reason for higher prices, quote realistic lead times, keep a working refrigerant buffer without panic-buying, and ban topping-off/mixing on every truck.
- Know your business's rough valuation before PE calls — HVAC trades 3x–10x EBITDA and the gap is mostly recurring revenue, management depth, and clean books, not size. Grow your service-agreement base, reduce owner-dependence, and get trailing-twelve-month financials buyer-ready even if a sale is years away.
- Audit your top techs' pay and your hiring pipeline — the trades are short ~439K workers and ~41% of the workforce nears retirement by 2031. Benchmark your best people against current medians before a competitor poaches them, and build an apprentice-to-lead pipeline rather than reacting to vacancies.
- Market standby power and document storm work to insurance-grade — the Indiana derecho left customers without power for up to two weeks, which is peak generator-buying motivation. Chase the long-tail restoration demand (generators, surge protection, water/mold, damaged condensers), document scopes carefully for scrutinizing carriers, and win on local trust over storm-chasers.
Compiled August 24, 2026. Figures reflect the cited sources. Valuation multiples, wage data, and refrigerant cost/lead-time figures are market estimates that vary by source, market, and business quality — verify current Google Ads migration timing for your own account, equipment and refrigerant pricing, insurance terms, and any valuation before acting. Storm and restoration figures reflect reporting as of late August 2026.