How do insurance brokerages choose technology and avoid expensive software mistakes?

How insurance brokerages and agencies choose technology and avoid expensive AMS, portal, and AI software mistakes, with an independent advisor on the owner's side.

TL;DR: Insurance brokerages waste money building custom portals, picking the wrong agency management system, and chasing AI quoting hype. I am Giacomo Balli, an independent technology advisor. I sit on the owner's side, not the vendor's, and help non-technical principals make $25k to $250k software decisions without learning the hard way.

Most agency technology mistakes are not bad software. They are the wrong software bought for the wrong reason, or a custom build that should have been an off-the-shelf purchase. Applied Epic, Vertafore AMS360, EZLynx, HawkSoft, and NowCerts all work. The expensive errors happen in the decision before the contract, where a non-technical owner is talking to people who get paid when you say yes.

What technology decisions actually matter for an agency?

The decisions that move money are your agency management system, your comparative rater, your client portal, and how AI fits in. Get the AMS wrong and every renewal, commission, and ACORD form runs through friction for years. Get a custom build wrong and you own a maintenance bill forever. These few choices decide your client retention and your margin.

Everything else, the small tools and add-ons, is reversible. The AMS is not. That is where I spend most of my attention with P&C and benefits brokerages, because that single platform touches carriers, policies, producers, and commissions every day.

Should we build a custom client portal or buy one?

Buy, in almost every case. Applied, Vertafore, and NowCerts already ship portals that issue certificates of insurance, store policy documents, and let insureds self-serve. A custom client portal runs $40k to $150k to build and then you maintain it forever, through every browser change and security patch. The math rarely favors building.

I have watched owners in many industries fall for the same trap: a portal feels strategic, so they build it, then it rots because no one budgeted for upkeep. Build only if the portal is your actual product. For a brokerage, it is a feature your AMS already covers.

What is the biggest risk in an AMS migration?

Dirty data and broken carrier download. Moving from AMS360 to Applied Epic, or off a legacy platform, fails when policy fields, ACORD codes, and commission history map incorrectly. The first time renewals break or commission reconciliation goes sideways, your producers stop trusting the system. That is the moment a migration quietly dies.

Clean the data before migration, not during. Garbage carries over.
Run parallel for a full renewal cycle so nothing falls through.
Test carrier download and ACORD/EDI feeds against real carriers early.
Plan producer training, because adoption is where most migrations actually fail.

Is AI quoting and underwriting worth the hype?

For narrow tasks, yes. For the big promises, not yet. AI that pre-fills ACORD forms, reads loss runs, summarizes policies, or drafts renewal emails saves your service team real hours this quarter. AI that claims to replace a comparative rater or underwrite accounts on its own is mostly a demo, not a tool you should bet the book on.

My rule for any vertical: buy the boring automation that removes keystrokes, ignore the magic that promises judgment. Test any AI tool on a slice of your own book before you roll it across every producer. A confident demo on someone else's data tells you nothing about yours.

Why is a non-technical agency owner so exposed here?

Because the people explaining the technology are the people selling it. A non-technical principal cannot easily tell a real integration from a slide, or a fair contract from one that locks in price hikes. The asymmetry is the problem. You have the money to fund the right move, but not the in-house expertise to know which move is right.

That gap is normal. You run an agency, not a software company. The fix is not to become technical. It is to rent someone technical who owes loyalty to you and reads the deal the way you would if you spoke the language.

What does an independent advisor change for you?

I do not sell software, build it, or take referral fees from Applied, Vertafore, EZLynx, or Salesforce. I earn the same whether you buy, build, or walk away. That independence lets me give the unpopular answer: the shiny tool is wrong, the contract clause is a trap, or the build you want should never happen.

I translate between your team and the vendors and developers, so you stop nodding along to words you cannot verify. You get a plain recommendation in writing, in language you can act on.

How does cross-industry experience de-risk the spend?

Because the same mistakes repeat across verticals. The build-versus-buy trap, the integration that looked simple and was not, the vendor lock-in clause, the AI pilot that never shipped. I have seen these play out in dozens of industries, so I recognize the expensive one in your $25k to $250k project before you sign, not after.

Insurance has its own lingo, COIs, renewals, E&O, commission reconciliation, but the technology decisions rhyme with what I have watched in wealth management, accounting, and other regulated, relationship-driven businesses. Pattern recognition is the edge a single industry vendor cannot give you.

Key takeaways

Your AMS, rater, portal, and AI plan are the few decisions that move money. Most other tools are reversible; the AMS is not.
Buy the client portal. A custom one costs $40k to $150k and you maintain it forever.
AMS migrations fail on dirty data, broken carrier download, and poor producer adoption, not on software quality.
Buy AI that removes keystrokes; ignore AI that promises to replace your rater or underwriter.
An independent advisor earns the same regardless of your choice, so the advice serves you, not a vendor quota.

Related guides

FAQ

Should my agency build a custom client portal or buy one?

Almost always buy. Applied, Vertafore, and NowCerts ship client portals that handle COIs, policy documents, and self-service certificates already. A custom portal costs $40k to $150k and you maintain it forever. Build only if a portal is your actual product, not a feature your AMS bolts on.

What is the biggest risk in an AMS migration?

Dirty data and broken carrier download. Migrating from AMS360 to Applied Epic, or off a legacy system, fails when policy fields, ACORD codes, and commission history map wrong. Producers lose trust the first week renewals break. Plan data cleanup, parallel running, and download testing before you sign anything.

Is AI quoting and underwriting worth it for my agency?

Sometimes, for narrow tasks. AI that pre-fills ACORD forms, reads loss runs, or drafts renewal emails saves real hours. AI that promises to replace a comparative rater or underwrite accounts is mostly hype today. Buy the boring automation, ignore the demo magic, and test on your own book first.

How does an independent advisor differ from a vendor or consultant?

I do not sell software or take vendor referral fees. Applied, Vertafore, and EZLynx resellers earn when you buy. I earn the same whether you buy, build, or do nothing. My job is to sit on your side of the table, read the contract, and tell you what not to spend money on.

How do you protect client data when adding new software?

Treat every integration as a data exit point. Insureds' SSNs, driver records, and loss histories flow through your AMS, rater, and portal. I check encryption, access controls, vendor SOC 2 reports, and how carrier download and ACORD/EDI feeds are secured, so a convenience tool does not become your E and O claim.

What does an engagement with you actually look like?

It starts with a free 20-minute call. If it makes sense, I review your AMS, integrations, and the decision in front of you, then give a plain written recommendation: buy this, skip that, negotiate this clause. Most engagements run a few weeks and cost a fraction of the mistake they prevent.

About the author

I am Giacomo Balli, an independent mobile and product technology advisor. I help non-technical owners make expensive software and AI decisions with confidence, drawing on patterns I have seen repeat across many industries. I do not sell software or take vendor fees, so my advice answers to you.

If you are weighing an AMS migration, a portal build, or an AI pilot and want a straight read before you commit, let's talk. Book a Find the right move call, free for 20 minutes, or email [email protected].