Hey Luis. I took a careful look at your agency. I think you're sitting on something bigger than it looks.
Eight short observations, based on what's public. No pitch, no pricing. I just want to see which of these you think is real, and which I got wrong.
You're the review leader in Lindenhurst. By a lot.
233 five-star Google reviews. The next closest State Farm agent in town has 16. Even the biggest independent broker here (Nusure) has 196. Most five-year-old agencies never get this far.
"I truly appreciate Luis' transparency and professionalism. From the start, he was incredibly attentive, genuinely kind. He took the time to answer my questions and made the process feel smooth and stress free." Michaella Jimenez, Google review
Everything in this note starts from that fact. You've earned the trust. What's below is about the distribution around it.
You have the best reviews in town, and you're not on page one.
If someone in Lindenhurst searches "car insurance Lindenhurst NY" today, you don't appear in the top results. Ten buildings down from your office, Young Insurance NY (196 S 1st Street) does. They're an independent broker with no website, seven Google reviews, and five years in business. Google still surfaces them before you.
Allstate agents, two independent brokers, and your own-street neighbor are all above you at the exact moment a local shopper has their wallet out. This is a distribution problem, not a product problem. Part of it isn't your fault at all: it's the structure State Farm puts every captive agent into. More on that in Play Two.
- Glowsure, a commercial insurance broker +1,576% organic traffic after a local-pages SEO build-out. Case study →
- Elephant Insurance, top-20 keyword rankings grew 10×, organic traffic +130% in one year.
Half your reviews are from Hispanic customers. Your digital presence is English-only.
You're a Latino-owned, bilingual agency. Your Google profile says so. Your reviews confirm it. Rosario Aponte, Jiménez, Pauta, and on and on. Yet:
A Spanish-speaking family searching "seguro de auto en Lindenhurst" has no way to find you. Nusure (a nearby broker) advertises in Spanish and is actively picking up that traffic. So is Young Insurance NY, literally on your block: no website, seven Google reviews, and most of those reviews in Spanish. They're winning a fight you're uniquely equipped to win, because you, Brenda, and Jordy actually are bilingual; they're just loud about it.
- Infinity Insurance runs a fully Spanish-localized version of its site, header language toggle, Spanish landing pages, Spanish quote flow. infinityauto.com/es →
- Seguro-Insurance. Spanish-first brokerage, English secondary. seguro-insurance.com →
Your reviews are trapped on one platform.
When a shopper compares agents on Yelp or Facebook, they don't see you. That's not because customers don't love you, they do, it's because no one has routed them there.
Tom Falletta at Allstate uses an automated review tool (Birdeye) to spread his across every platform, that's part of why he shows up where you don't. The fix is a small, one-time setup: each happy customer gets gently steered to the platform that needs the signal most.
- GAINSCO Auto Insurance (Dallas), went from 1 Google review at 1 ★ to 2,561 reviews at 4.9 ★ via Birdeye.
- TrueCoverage (Albuquerque) 1 → 1,365 reviews, 1 ★ → 4.8 ★.
- LifeQuotes.com 2 → 3,490 reviews.
601 Instagram posts. 223 followers. The content isn't the problem.
You've put in the work. The issue is what's going out: mostly reposted corporate State Farm copy, English only, no story behind any of it. Your "Quotes for Good" donation program is a gift of a content angle, and it's almost invisible outside the website footer.
A bilingual content engine built from the things you already do, customer stories, charity donations, Chamber events, local-market explainers, would reverse that ratio without adding a single hour to your week.
Everything your agency actually needs, built for you, not rented forever.
Your site runs on the standard State Farm template, fine, but the reason every competing agent's site looks the same. Your social is corporate boilerplate. Two years ago, customizing any of this either meant a second job for you and Brenda, or paying a recurring SaaS bill for generic software that covered maybe 60% of what you actually need.
There's a third option now. Modern reasoning AI (Claude, GPT-4) plus low-code automation (n8n, Make, Airtable) lets us build each of these pieces custom for your agency, fast, cheap, owned by you, tuned to your 233 reviews, your bilingual customers, and your tone. Not renting generic tools forever. Not a marketing agency on retainer.
Five plays follow. Each lives in a specific place on your website, your phone, your pipeline. Each is being done somewhere in the wild right now, either in-house at companies like Lemonade, or with open-source tooling a small team can wield. None of them are hypothetical.
Bilingual AI chat + instant SMS, on your site.
Turn every form submission into a live conversation, in the customer's language, within 10 seconds, 24/7.
Today
A Spanish-speaking family lands on your site at 9 PM Saturday. They fill out the quote form. The message goes to Brenda's inbox. She reads it Monday at 9:15. By then, they've also filled out forms at Young Insurance, Tom Falletta's Allstate, and GEICO. Whoever replied first tends to win.
What changes
The same reasoning models behind ChatGPT (Claude, GPT-4) are now fast and cheap enough to run as a chat widget on your site. A custom agent can:
- Detect English or Spanish from the first message, no language toggle needed
- Answer "what does full coverage include?" in your actual policy language
- Give an indicative premium range from ZIP + vehicle year, without a 12-field form
- Book the family onto your calendar for Monday 9:15 AM
- Send you an SMS with the full conversation, so you walk in already knowing them
This isn't a scripted "press 1 for auto" bot. It's a reasoning model that responds like a thoughtful new hire would.
vs. buying a SaaS chatbot
- Off-the-shelf tools (Hyperleap, Intercom, Drift) are trained on their generic corpus; they sound the same for every agency. A custom agent trained on your 233 reviews, your bilingual phrasing, your objections sounds like you.
- Subscriptions charge a flat $400–$800/month regardless of volume; pay-as-you-use AI is roughly $0.01–$0.05 per conversation. At your current volume, that's single-digit dollars a month.
- When State Farm rolls out a new product, you update one prompt file. With SaaS, you open a support ticket and wait.
What your agency gets
No lead lost to weekend timing ever again. Spanish-speaking families who never would have picked up the phone end up in your pipeline. Monday mornings start with qualified, pre-read leads, not an unsorted form inbox.
Where it's already happening
Lemonade Insurance runs AI agents "Maya" (quoting + binding) and "Jim" (claims). As of 2025, 96% of first-notice claims are handled without a human, and 55% of all claims resolve fully automatically, in seconds, not weeks. Lemonade built it in-house with billions in funding. The same building blocks are now openly available: Twilio's public tutorial walks through wiring Claude or GPT-4 to a live voice/SMS line in about a day of engineering.
Local presence that your State Farm template can't give you.
The reason captive agents lose local search to independent brokers, and the workaround the best ones have been running for years.
The structural problem first
Your primary website (luissfinsurance.com) is a State Farm agent template. Brand-safe, compliance-approved, and deliberately rigid. You can edit "About me," add a few photos, tune a small set of fields. That is the whole surface area. You cannot publish a page like "/auto-insurance-copiague-ny". You cannot run a blog. You cannot target town-level or language-level search with custom content. State Farm's guidelines explicitly forbid "independent websites" under the brand, and the template enforces that from the other direction.
This is not your oversight. It is the trade the captive-agency model makes: national brand trust in exchange for local flexibility. For customers searching "State Farm near me," that trade wins. For customers searching "seguro de auto en Lindenhurst" at 9 PM Saturday, it quietly loses every time.
Why Young Insurance beats you on your own street
Young Insurance NY has no website, 7 Google reviews, and no brand authority. They're an independent broker. Because they're independent, they have complete freedom: if they ever wanted a page titled "Auto Insurance in Lindenhurst", they'd make one tomorrow. They don't even bother, and they still pull local search because Google defaults to proximity + presence. Meanwhile your 233 reviews, your bilingual team, and your State Farm authority are locked inside a template no optimization agency can move.
Every captive State Farm and Allstate agent sits in this same structural disadvantage against any independent broker in their zip. It is the single biggest asymmetry in the business.
The workaround: a parallel property, compliance-safe
The top-producing captive agents who actually rank locally run a satellite content site on a separate, personal domain: not branded as State Farm, positioned as a personal insurance resource or local guide. Think "Lindenhurst Insurance Notes by Luis Hernandez" or "Long Island Coverage Guide." It sits outside the "no independent websites under the brand" rule because it is not under the brand. It ranks for town × product queries. It funnels leads to your phone, to your State Farm-approved quote form, or to a simple contact page. Same business, same you, just on a property you actually own.
What we build on that satellite site
- An AI agent reads your 233 reviews and extracts what locals actually care about: flood exposure in Copiague, commuter auto in Babylon, young families in Amityville, Spanish-speaking households across the service area.
- It drafts one page per town × product in English AND Spanish: "Auto Insurance in Copiague: what your neighbors are actually asking us," plus "Seguro de Auto en Copiague: lo que preguntan tus vecinos de verdad."
- It weaves in your real customer quotes, your Quotes for Good donations, and local references that prove genuine authority, not template filler.
- It publishes and schema-marks through a domain we register for you.
- It refreshes monthly with new reviews and State Farm product changes, no touch from you.
vs. trying to optimize the State Farm template
- "State Farm SEO" services extract $2K+/month to tweak the handful of template fields you already have access to. Ceiling is low by design; they can only move the pieces State Farm lets them move.
- A parallel site has no template ceiling. Content depth, schema, internal linking, language targeting, all available.
- Compliance-wise, a personal content site under your name is a clean pattern State Farm has seen many times. We keep all State Farm product marketing on the approved template; the satellite stays editorial, educational, and non-transactional.
What your agency gets
You rank for every town × product in your service area, in English and Spanish. Young Insurance and Nusure stop owning Hispanic Lindenhurst searches your 233 reviews actually deserve. The satellite site becomes a durable, compounding SEO asset that feeds your main pipeline without ever touching your State Farm-approved storefront.
Where it's already happening
The satellite-content-property pattern is how top-performing captive agents already escape template limits; agencies like Agent Branding Marketing publicly document the playbook for State Farm specifically. For the content engine itself: Glowsure (commercial insurance broker) hit +1,576% organic traffic with the same town × product approach. KrispCall built a page per U.S. area code; those now generate 82% of its U.S. organic traffic. The Search Initiative's local AI SEO case study shows the same mechanics for a small local service business.
Review syndication on autopilot.
233 Google reviews; zero on Yelp. Point the firehose at the platform that needs it.
Today
Every time you bind a new policy, a happy customer walks out. Sometimes Brenda remembers to ask for a Google review, they leave one. Your Yelp page has been empty for years. Your Facebook has three. Tom Falletta at Allstate uses Birdeye to distribute his reviews across every platform; that's part of why he shows up where you don't.
What changes
Modern low-code automation turns this into a four-hour build:
- When you mark a new policy as bound in your system, a workflow fires (n8n, Make, or a 50-line script)
- An AI agent drafts a bilingual thank-you referencing the customer's specific policy "Gracias Maria por confiar en nosotros con tu auto y tu casa", not a template
- It picks the target platform based on where your review signal is weakest right now (Yelp first, then Facebook)
- Sends via SMS with a one-click link to the correct review page
- If no review in five days, sends one gentle reminder, then leaves the customer alone
- Tracks who left a review so the same ask never fires twice
vs. paying Birdeye or Podium
- Birdeye: $299–$599/month per location, forever. Custom: one build + pennies per SMS.
- Birdeye's message is Birdeye's voice. Custom uses the customer's name, their policy type, and your bilingual tone.
- Birdeye picks the target platform on a static rule. Custom rebalances dynamically as your review counts shift.
What your agency gets
Your Yelp page fills up over 6–8 months. Facebook follows. Customers who trust Yelp or Facebook can finally find you at the top of their preferred platform. Your review signal stops being a single point of failure.
Where it's already happening
Birdeye's public insurance results show what the outcome looks like: GAINSCO Auto: 1 review at 1★ → 2,561 at 4.9★. TrueCoverage: 1 → 1,365. LifeQuotes: 2 → 3,490. Those agencies paid Birdeye for it. The n8n workflow library and projects like n8n-money (open-source insurance claim automation) show the same plumbing running for free, owned by the business instead of rented.
Bilingual content engine for Instagram + Facebook.
601 Instagram posts got you 223 followers. Let's flip the ratio, without posting more.
Today
You've posted 601 times on Instagram. The follower count (223) tells you the content isn't landing. The feed: reposted State Farm graphics, generic PSAs, English only. Zero local story. Zero Spanish reach. Zero of your actual customer wins.
Meanwhile, your Quotes for Good program, donating to a local charity for every quote, is buried in the website footer. It's the kind of thing that goes viral locally. Nobody knows about it.
What changes
An AI content agent becomes your local marketing person:
- Knows your story. Trained on your 233 reviews, your Chamber of Commerce appearances, your Quotes for Good donations, your Lindenhurst community calendar.
- Drafts 2–3 bilingual posts per week in formats that actually work on Instagram: short carousels, reels scripts, text-on-image quote cards
- Example draft: "Meet Ricardo, full coverage for less than his last agent. This week, his quote helped us donate to @LindenhurstFoodPantry. Quotes for Good, because being a good neighbor should literally mean something." + Spanish version, automatically.
- Approval flow. Drafts land in a shared inbox; you click approve; it publishes via Instagram and Facebook APIs directly.
- Learns. Watches which posts perform and adapts tone and format over the first few months.
Your time: 20–30 minutes a week reviewing drafts.
vs. hiring a marketing agency or using Buffer + Jasper
- Marketing agency: $2K–$5K/month, generic copy, eventually leaves. Custom AI: one build, your voice, forever.
- Buffer schedules posts; Jasper drafts generic ones. The real bottleneck is drafting in your voice, which only a model trained on your specifics does well.
- You stop paying monthly for a feed that could just as easily belong to a Geico agent in Phoenix.
What your agency gets
Followers grow because the content finally sounds local and bilingual. Spanish-speaking Lindenhurst families see themselves in your feed. Quotes for Good turns into monthly earned attention, the local-press-worthy thing it was always meant to be.
Where it's already happening
Research on AI-assisted social content: businesses using AI drafting report a 30% increase in engagement with 25% less production time. Small local businesses regularly 3–5× their follower growth in the first six months with targeted local content. See Postly's case studies on small-business Instagram automation, and the AI vs human content research on Instagram engagement. The open-source stack is Claude + Instagram Graph API + a scheduler, all publicly documented.
Cross-sell and renewal that actually fires.
Your reviews are full of "home AND auto." Let's stop letting the rest slip through.
Today
You sold Maria an auto policy 10 months ago. She insures her house with Liberty Mutual. Her son turns 17 in six months. Her auto renews in 60 days. She hasn't heard from your office since January.
Your reviews are full of "got home and auto with Luis, great rates!", customers who found the cross-sell on their own. How many didn't? How many Marias are sitting in your book right now?
What changes
An AI agent becomes your account-review team:
- Watches your policy data (State Farm back-end or a simple Airtable mirror), who has auto only, who has home only, who's 60 days from renewal, who just had a life event
- Picks the right moment. A renewing auto customer at 45 days gets a confirmation plus a home-insurance quote offer. A recent claimant gets a thank-you and an umbrella-policy explainer. A new parent gets a life-insurance conversation starter.
- Writes in your voice. Bilingual, personal, using the customer's actual name and policy data: "Hey Maria, Brenda here. Your auto renews Aug 12. I noticed you insure your house with Liberty. Quick check: you'd save around $X/year bundling with us. Want me to pull a quote?"
- Hands complex cases to humans. Cold leads are auto-nurtured; warm ones pop up on Brenda's desk already pre-briefed.
vs. buying AgencyZoom, Levitate, or HubSpot
- AgencyZoom: $70–$200 per seat per month, forever. Custom: one build, runs on API + SMS costs.
- Templated platforms feel templated. Custom feels like Luis, because it was trained on you.
- Generic CRMs don't know Lindenhurst, your Chamber relationships, your charity angle. A custom pipeline does.
What your agency gets
Cross-sell opportunities stop slipping. Renewals stop being last-minute scrambles. Your revenue per customer grows without spending another dollar on lead generation. Your book deepens instead of just widening.
Where it's already happening
AgencyZoom's 6,000+ agencies see average 40% growth in year one, driven almost entirely by cross-sell and renewal cadence. Insurance operations using n8n + AI agents have cut processing times by 70% inside 30 days. That's the outcome. SaaS rents it to you at $1,000+/month/user forever; built with modern AI agents and low-code, the same outcome is one initial build + light upkeep, and it evolves with your business instead of a vendor's roadmap.
Why custom is the safer play, and the one that ages better.
The details your State Farm regional office and your E&O carrier will quietly care about too.
Your data stays on your side of the fence
With Birdeye, AgencyZoom, or Hyperleap, your customer list, policy details, and every conversation sit in their database. You own an export button, if you're lucky. With a custom build, that data lives on infrastructure you control: your Airtable, your Supabase, your cloud account. When a customer asks "delete everything about me," you can actually honor it in minutes. When State Farm or a regulator asks where Maria's quote information is stored, you point to a specific server you can audit, not a vendor's shared tenant.
Enterprise-grade AI privacy, by default
Both Claude (Anthropic) and GPT-4 (OpenAI) offer Enterprise API tiers with explicit guarantees: your data is not used to train their models, retention is zero or near-zero by default, and a proper Data Processing Agreement is available. A generic SaaS chatbot typically sits on lower-tier API access and pools your traffic with thousands of other customers. A custom build puts your agency directly on the enterprise tier with your own API key and your own isolated traffic, the same privacy posture a hospital network or a bank uses.
No vendor lock-in, ever
SaaS stacks marry you. Your customer list, your templates, your workflow logic, all trapped inside someone else's database. Migration is painful on purpose. With custom, every piece is yours: the data in formats you own (CSV, Airtable, Postgres), the logic in code any developer can read, the integrations in config files you can edit. If our working relationship ever ended tomorrow, you keep the entire system, and can hire any competent engineer to maintain it for a fraction of a SaaS renewal bill.
Audit-grade transparency
State Farm's compliance rules, NY DFS oversight, and your E&O carrier all care about data flows. A custom build writes a complete audit log for every AI interaction: who asked what, what exact text was sent to the model, what came back, what decision was made, and when. That's evidence a regulator or an auditor can actually work with. With SaaS, you get their compliance brochure and a promise, and their audit logs sit in their tools, not yours.
Economics that improve over time, not erode
SaaS pricing moves in one direction: up. More seats, more add-on tiers, annual increases. LLM API pricing has moved the opposite direction: the same-quality AI call dropped 10× in cost over 18 months, and the curve keeps falling. A custom build locks in today's costs as a ceiling, not a floor, and rides the drop down. Meanwhile you pay no per-seat fees, no renewal pressure, no "we're raising prices 20% this quarter" emails.
The roadmap is yours
When State Farm ships a new product next week, we update a prompt that same afternoon. When you decide to add Massapequa to the local pages, we add a row. When Spanish copy needs different phrasing for Dominican versus Puerto Rican customers, we tune overnight. SaaS vendors ship on their quarterly roadmap and lobby you for upsells; your custom stack ships when you need it to.
Where the industry is going
The shift away from heavy SaaS toward LLM-native, customer-owned automation is already well underway. Andreessen Horowitz calls it "AI eating application software"; public SaaS valuations dropped ~30% in early 2026 as the market started pricing in the shift. Enterprise buyers are increasingly insisting on self-hosted or own-tenant AI before signing any new contract. Insurance, with its PII sensitivity and compliance weight, is one of the industries where this move happens fastest. Going custom now isn't a contrarian bet; it's skating where the puck is going.
To be plain about my side of this.
I'm not pitching you a shopping list of subscriptions. What I do is build the custom systems your agency needs, using modern AI agents and low-code tooling, and then run them. Concretely:
-
• Pick the right 1–2 plays firstNot all five at once. The one with the shortest path to a measurable win for your mix of auto / home / Spanish-speaking customers.
-
• Build and wire them inCustom AI agents, site widgets, SMS flows, bilingual content pipelines. All of it wired into your existing site, State Farm back-end, and phone. You own the output; no per-seat SaaS bill attached.
-
• Train them on your businessYour 233 reviews, your bilingual scripts, your Quotes for Good story, your objection patterns. Off-the-shelf AI sounds like a call center. Yours should sound like Luis.
-
• Run the weekly cadenceContent, review follow-ups, lead triage, renewals, handled in the background. You approve anything that touches a customer; everything else runs itself.
-
• Report what's actually movingMonthly: leads in, quotes out, reviews gained, where the pipeline grew. When one play is earning, we build the next, not before.
If any of this lands (or if I got something wrong and the picture is different from where you actually sit), I'd love to hear it. I'd rather start with the right read than a polished one.
Drop me a line whenever.
No deck to sit through, no slide count. Just 30 minutes, neighbor to neighbor.
dslabakov@me.com