Few marketing failures are as quiet as a Local Services Ads slowdown. A broken Search campaign announces itself — disapprovals, budget alerts, a dashboard full of red. LSA just… goes still. The profile looks fine, the badge is green, the budget is untouched, and the phone that rang eleven times a week rings twice. Because the channel has no keywords to audit and no landing pages to test, owners are left staring at a system with almost no visible controls, guessing between explanations that range from “Google changed something” to “the market died” — while a competitor quietly absorbs the calls.
The truth is that LSA volume drops have a short, learnable list of causes, and they sort cleanly by where the failure lives: eligibility (you stopped being fully servable — paused status, lapsed verification, billing hiccups, policy holds), ranking (you’re servable but losing the carousel — responsiveness decay, review stagnation, competitors compounding), configuration (your own settings quietly shrank your surface — budgets, hours, services, areas, bidding mode), and market (demand or competition actually changed). Each layer has specific checks, and the layers are ordered: there is no point optimizing reviews on a profile that isn’t fully eligible, and no point blaming the market before ruling out the account.
This guide is the diagnostic, in the order a professional would run it: the five-minute eligibility sweep, the ranking-factor audit with the responsiveness math most owners have never seen, the configuration traps (including the budget and bidding interactions that throttle serving), the market checks that separate “you fell” from “the tide went out,” and the recovery sequencing once you’ve found the cause — because several fixes compound and a few, done impatiently, make things worse.
LSA call drops have four layers — diagnose in order. Eligibility first (5 minutes): ad status paused? license/insurance verification lapsed? billing failure? policy or background-check hold? Any of these silently stops or suppresses serving. Ranking second: LSA ranks on responsiveness (answer rate, reply speed), review velocity and rating, and proximity — a month of missed calls or a stalled review pipeline slides you down the carousel while competitors compound. Configuration third: budget caps set too low throttle serving; shrunken hours, unchecked services, trimmed service areas, and bid-mode changes all quietly cut your surface. Market last: seasonality, new badged competitors, and category demand shifts — verify with impression trends and live searches before blaming the algorithm. Recovery: fix eligibility instantly, then answer rate (fastest ranking lever), then review velocity (strongest), then budget headroom — and give ranking recovery 2–4 weeks of consistent signals.
Layer 1: The Five-Minute Eligibility Sweep
Before any analysis, confirm you’re actually fully servable — the embarrassing causes are the fastest to fix and more common than anyone admits:
- Ad status. Confirm the profile shows active/enabled — not paused (it happens: a team member “pausing for vacation” and forgetting is a classic), not limited.
- Verification currency. The Google Guaranteed/Screened status rests on documents that expire: license renewals, insurance certificates, periodic background-check refreshes. A lapsed document can suspend the badge — and with it, most of your serving — with notification emails that famously land in inboxes nobody reads. Check the profile’s verification section and your document expiry dates directly.
- Billing health. A declined card or payment hold stops serving without ceremony. Verify the payment method and any outstanding balance.
- Policy and quality holds. Review the account for notifications: customer-complaint escalations, guarantee claims, or policy reviews can limit serving while open. These come with a paper trail — read it before theorizing.
- Message-lead hygiene (if enabled). Persistently ignored or expired message leads can suppress that lead type entirely; if you accept messages, confirm they’re being answered inside the expected windows.
If everything here is green, the drop is competitive or self-inflicted — proceed to ranking.
Layer 2: The Ranking Audit — Responsiveness, Reviews, Proximity
LSA’s carousel has a handful of slots and ranks profiles primarily on responsiveness, review strength, and proximity (with hours-of-operation fit and budget adequacy as gates). Because two of the three decay silently, ranking loss is the most common non-obvious cause:
- Responsiveness decay. Answer rate is a first-class ranking input — and it degrades operationally, not strategically: a receptionist change, new lunch coverage gaps, calls routing to a voicemail nobody checks. Pull your answer rate for the last 60 days and compare it to the prior period; anything below ~90% is both a ranking drag and a paid-lead leak (missed LSA calls are still charged). The fix stack — ring-all routing, missed-call text-back, after-hours answering — is the same speed-to-lead infrastructure that lifts every channel, but on LSA it’s also literally how you rank.
- Review stagnation vs competitor velocity. Your review count didn’t fall — theirs rose. Search your category live and log the top carousel profiles’ review counts and ratings monthly; a competitor running a systematic post-job review request flow adds 15–30 reviews a month and will pass a static profile within two quarters. Review recency matters alongside count: a profile whose latest review is four months old reads as dormant. The solicitation system from our Maps-first playbook serves LSA identically.
- Proximity you can’t change — and shouldn’t fake. Distance to the searcher is a real factor; volume from the far edges of your area will always be thinner. The legitimate responses are service-area realism and (for multi-location businesses) location strategy — not address games, which risk the verification that everything else depends on.
Sixty seconds of arithmetic outranks a week of speculation: (1) Answer rate = calls answered ÷ calls received, last 60 days vs prior 60 — a five-point drop is a ranking event. (2) Review velocity gap = your new reviews per month minus the average of the top three carousel competitors’ — a negative number is a countdown timer on your position. These two metrics explain the majority of ‘mysterious’ LSA declines, they’re both fully in your control, and neither appears on any dashboard unless you build the habit of computing them.
Layer 3: Configuration Traps — the Self-Inflicted Layer
| Setting | The trap | The check |
|---|---|---|
| Budget cap | Caps set near expected spend throttle serving unevenly — the system paces you out of auctions late in periods; profiles that exhaust budget rank as less available | Was the cap lowered? Is spend hitting the cap? Headroom of 30–50% above target volume keeps pacing off your back |
| Bidding mode | Switching between maximize-leads and manual per-lead bids (where available) changes competitiveness overnight; a max-bid set below market quietly loses every auction | Any recent bid-mode or bid-amount change? Compare per-lead bid to Google’s suggested range for the category |
| Hours | Ads serve around your stated availability; trimmed weekend or evening hours delete exactly the emergency windows where LSA volume concentrates | Diff current hours vs 90 days ago; check whether the “drop” is concentrated in hours you no longer show |
| Services checked | An unchecked service (often during a dispute-hygiene cleanup that over-corrected) removes you from those queries entirely | Audit the service list against what you actually sell — the mirror image of the junk-lead audit |
| Service areas | Zip codes trimmed to cut out-of-area leads can amputate legitimate volume with them | Compare lead geography before/after any area change; re-expand where the trucks actually go |
| Photos & profile completeness | Thin profiles convert impressions to calls worse — not a serving problem, a persuasion one | Profile has current photos, accurate business info, and the review responses that signal a live operation |
Configuration causes have a signature: the drop coincides with a change-history event. Reconstruct the timeline — who touched the account in the two weeks before the slide — before accepting any external explanation. It’s the same discipline as the Search-side troubleshooting tree: sudden changes have configuration causes; gradual ones have competitive causes.
Layer 4: Market Reality Checks
Only after the first three layers clear does “the market” earn consideration — and it’s verifiable, not vibes:
- Seasonality sanity. Compare against the same period last year, not last month — HVAC shoulder seasons, holiday troughs, and weather-driven demand swings are enormous in most LSA categories. Your own booking history is the best baseline; category search-trend data corroborates.
- New badged competitors. Run the live searches across your services and zips at several times of day; log who occupies the carousel now versus your last audit. A newly verified competitor with aggressive review velocity and wide hours explains a share loss no setting will fix — the response is the ranking program, accelerated.
- Impression-versus-lead decomposition. Where reporting allows, separate “shown less” from “shown but not chosen”: falling impressions point to eligibility/ranking/serving; stable impressions with falling calls point to profile persuasion (photos, rating, review recency) or to demand quality shifts. The two have opposite fixes.
- Category volatility. LSA markets do experience platform-side fluctuation — slot counts, format tests, category changes roll out unevenly. If your diagnostics are clean and peers in owner groups report the same timing, patience plus the controllable levers is the strategy; what you shouldn’t do is thrash your configuration chasing a platform test.
The instinctive reaction to a volume drop — change five settings at once, pause and unpause, slash the service area, toggle bidding modes weekly — is itself a cause of prolonged slumps. Every configuration lurch resets what the system knows about your availability and competitiveness, and simultaneous changes make the eventual recovery unattributable: you’ll never know which change worked, so you’ll never trust the account again. The professional pattern: diagnose fully first, change the one thing the evidence indicts, give ranking-signal fixes (answer rate, reviews) 2–4 weeks of consistent data to register, and log every change with a date so the timeline stays readable. LSA rewards boring consistency — answered phones, fresh reviews, stable generous settings — more than any clever toggle.
Recovery Sequencing: What to Fix First
- Eligibility, same day. Reinstate documents, fix billing, resolve holds — nothing else matters while serving is suppressed.
- Answer rate, this week. It’s the fastest-registering ranking signal and it monetizes instantly (every recovered call is a paid lead you now actually receive). Routing, coverage, text-back — target >90% within days.
- Budget and settings headroom, this week. Restore generous caps, full honest hours, complete service lists, realistic areas — remove every self-imposed throttle in one documented pass.
- Review velocity, starting now, compounding for months. The systematic post-job request flow is the strongest durable lever; expect position gains to lag the review curve by weeks.
- Bridge the gap with Search. While LSA ranking rebuilds, the fastest volume replacement is the channel you fully control — the LSA-vs-Search allocation flexes toward Search during recovery and rebalances as the carousel position returns.
5 Common LSA-Slump Mistakes
- Skipping the eligibility sweep. Days spent theorizing about algorithms while an expired insurance certificate sat in the notifications tab.
- Not knowing the answer rate. The most powerful ranking lever, unmeasured in most accounts that complain about ranking.
- Blaming the market first. Layer 4 is last for a reason — it’s the only layer you can’t fix and therefore the most comfortable to blame.
- Thrashing settings. Five simultaneous changes, zero attribution, prolonged slump.
- Fixing volume while leaking conversions. Recovering impressions for a profile with stale photos, a four-month-old latest review, and a 70% answer rate refills a bucket with holes — run the persuasion and operations fixes with the visibility ones.
Frequently Asked Questions
My LSA leads dropped 60% in one week with no changes on my end. What's the most likely cause?
A cliff that sharp almost always lives in Layer 1 or a platform-side event — gradual competitive erosion doesn’t move 60% in a week. Run the eligibility sweep first and literally read every notification: lapsed license or insurance documents suspending the badge, a billing decline, a background-check refresh pending, or a policy hold are the classic silent killers, and each comes with a paper trail in the account or your email. If eligibility is clean, check for changes you didn’t make: bidding-mode defaults, category reshuffles, and serving-format tests roll out platform-side, and your change history plus LSA support can confirm. Then verify with live searches across several times of day — if your profile appears normally, the ‘drop’ may actually be a reporting or call-routing failure (test-call your own LSA number). What a one-week cliff almost never is: reviews or answer rate, which erode positions over weeks, or seasonality, which your last-year baseline will confirm or deny in one comparison.
What answer rate do I actually need for LSA to rank well?
Treat 90% as the working floor and 95%+ as the competitive standard in contested categories. The mechanism is twofold: responsiveness is a direct ranking input (the system prefers sending its finite carousel clicks to businesses that demonstrably pick up), and every missed call is simultaneously a charged lead you didn’t receive and a customer who called the next profile down — so a 75% answer rate loses three ways at once. Getting to 95% is operations, not heroics: simultaneous-ring routing to two or more phones, a missed-call text-back that fires within a minute (it salvages the lead even when it can’t save the ranking signal), coverage plans for lunch and shift changes where gaps concentrate, and an after-hours answering service if your hours claim evenings and weekends — or honest hours if it shouldn’t. Measure it weekly from your call platform, not from memory; accounts that start measuring routinely discover they’re ten points below what everyone in the building would have sworn.
Should I raise my LSA budget to get more calls when volume drops?
Only after confirming budget was actually the constraint — which has a specific signature: spend consistently hitting the cap, and serving that feels pace-limited (leads clustered early in periods, then silence). In that case, yes: LSA’s pay-per-lead model makes generous caps nearly riskless, and headroom of 30–50% above your target volume keeps the pacing system from rationing you out of auctions; budget-starved profiles effectively signal reduced availability. But if spend sits well below the cap — the more common situation in a slump — raising it does nothing, because budget wasn’t gating you; eligibility, ranking, or configuration was, and the diagnostic layers apply in order. One nuance for manual-bid categories: the ‘budget’ that matters may be your per-lead bid, not your cap — a max bid set below the category’s going range loses auctions invisibly, and comparing your bid to the suggested range (then testing a raise on its own, as a single documented change) is the correct version of ‘spend more.’
A new competitor with 500 reviews just entered my market. Is my LSA position gone for good?
No — but the response is a program, not a tweak, and it starts with reading what actually ranks them. Proximity you can’t take from them; review mass you can close over quarters, not weeks — which is why the systematic post-job review flow (every customer, automated timing, direct link) starts today: at 20–30 requests a month with normal conversion, a year adds 100–200 reviews, and recency counts alongside totals, so a fresh-review cadence partially offsets a static larger number. Meanwhile compete on the levers that move faster: answer rate (a 95%+ responder can outrank a bigger sleepier profile for real serving), hours coverage (their 500 reviews don’t answer at 9pm if their hours end at 5), and profile persuasion (current photos, responded-to reviews). Strategically, treat the entrant as a portfolio signal too: while the review race runs, shift incremental budget toward Search — where their review count buys them nothing — and toward the suburbs and services where their proximity or scope is weakest. Markets with a dominant LSA player still leak plenty of demand to the #2 and #3 slots; the goal is holding a compounding position, not conceding the carousel.
How long does LSA recovery take once I've fixed the cause?
It depends entirely on which layer you fixed, and setting expectations by layer prevents the mid-recovery thrash that restarts the clock. Eligibility fixes are fastest: once documents clear or billing resolves, serving typically resumes within days — though badge re-verification can take longer if checks must re-run. Configuration fixes (budget headroom, restored hours, re-checked services) register as soon as the system re-evaluates your availability — usually days. Ranking-signal fixes are the slow compounding tier: answer-rate improvements need 2–4 weeks of consistent data before the responsiveness signal re-rates you, and review velocity moves positions over one to three months as the count and recency accumulate against competitors’ own curves. Two disciplines make the timeline honest: change one thing per layer and log the date (so attribution survives), and track leading indicators weekly — answer rate, new reviews, live-search position spot-checks — rather than staring at the lead count, which is the last metric to move. If leading indicators improve for six weeks and leads don’t, re-run the diagnostic; something in an earlier layer was missed.
LSA gone quiet and nobody can say why?
We’ll run the full four-layer diagnostic — eligibility, answer-rate math, configuration timeline, competitor audit — identify the actual cause, and sequence the recovery while Search coverage bridges the volume gap.
Get an LSA Diagnostic Start with a PPC Audit