By Chaz Ripoll, VP of Customer Success at PerfectQuote, where he leads software onboarding and adoption for benefits agencies
Summary:
• Stalled agency rollouts usually come down to change management: no named owner, training that happens once, and a launch timed inside renewal season.
• The agencies that get adoption to stick decide what the new tool replaces, train on their own groups, attach one number to success, and retire the old spreadsheet on an announced date.
• Buyers should treat adoption as shared work with the vendor and ask before signing who owns onboarding, what week five looks like, and how new hires get trained.
The contract is signed. The kickoff call goes well. Everybody’s on camera, and somebody on the agency side says the thing I’ve heard a hundred times: “This is going to save us so much time in Q4.”
Then week six arrives. Only one person is on the new platform, usually the one who was excited about it in the demo. Everybody else is back in the spreadsheet they’ve used for nine years, and nobody has said a word about it out loud.
I’ve spent about twenty years watching that happen from both sides of the table, first selling enterprise software and now leading customer success for a benefits quoting platform.
So I come at this from the outside. I wasn’t an account manager at an agency, and I’m not going to pretend I was. What I do have is an enterprise adoption playbook that mostly transfers, and a pretty clear view of where it doesn’t.
Here’s the part that surprises people: the tool is almost never what went wrong. The rollout is. And most agencies plan carefully for go-live, then stop planning entirely for everything that comes after it.
Why do so many agency software rollouts stall?
The pattern I see most often is that buying the software gets treated as the decision, and using it gets treated as a detail. Leadership runs a real evaluation, sits through demos, compares pricing, negotiates, signs. Then the hard part starts and nobody has planned for it.
Three things go wrong, usually together.
Nobody actually owns the rollout
It belongs to “the team,” which in practice means it belongs to whoever has a free hour. There’s no one person whose job it is to notice that half the staff hasn’t logged in.
Training happens once
There’s a session, people nod, and then three weeks pass before anyone touches the platform again. By then most of what they learned is gone. That’s how memory works: unused learning fades fast, and most of it goes in the first few days. Meanwhile, the old workaround is right there where it’s always been. New hires show up in March and get taught the old process by whoever trains them, because that’s what’s written down.
The launch gets timed against the calendar instead of the business
An agency signs in September, wants value immediately, and starts rolling out in October with renewals stacking up. Renewal season is already the most stretched time of year before anybody adds training on top of it. Everyone’s working long days. Asking a team to learn something new in that window is asking them to choose between the new tool and getting their work done, and they’re going to choose getting their work done every time.
None of those are software problems. They’re change management problems, and they’d show up with any vendor you picked.
Where does alignment break down?
Agencies give individual users a lot of autonomy over their own workflow. That’s genuinely valuable, and it’s part of why good account managers stay. But it means a leadership decision to buy software doesn’t automatically become an expectation to use it. An owner or principal makes the call. The account managers are the ones who have to change how they work. In between sits whoever runs the team day to day, and if that person’s weekly check-in doesn’t change, nothing below it changes either.
That middle layer is where a decision becomes an expectation. They set the agenda for the status meeting, decide what a renewal handoff has to include, and notice when somebody has quietly gone back to the old spreadsheet. If nobody has told them the rollout is theirs to carry, they’ll keep running the process they already know, and their team will follow them rather than the announcement.
At a lot of agencies that layer is thin or informal. The person supervising accounts is usually carrying their own book too, so enforcing a new process competes directly with their own renewals during the busiest stretch of the year. That’s worth naming, because it means more end-user training won’t close the gap. What helps is giving middle management a reason to own the rollout and the room in the week to do it.
Is “Will my staff actually use this?” the right question?
I hear this one constantly, usually from an agency principal, usually near the end of an evaluation. It’s a fair question. Most agencies have bought something before that nobody used, and they’re not eager to do it twice.
But it puts the question on the staff, and that’s the wrong place for it. A better one to sit with is: what are we going to stop doing once this is live?
That question is harder, which is why it gets skipped. If the old spreadsheet is still on the shared drive, people will use the old spreadsheet. Not out of stubbornness, but because it’s Thursday, the client needs the comparison by Friday, and they already know how to do it that way. Every hour you leave the old process available is an hour the new one has to win on the spot, under deadline, against muscle memory.
The account manager lens is worth sitting with here too. In my experience, account managers (AMs) don’t resist tools. They resist extra steps. An AM who’s quoting a dozen groups a week will adopt anything that gets them home earlier and reject anything that adds a click, and they’ll make that judgment in the first week, during the busiest stretch of their year.
So the real version of the principal’s question is something like: have we decided what this replaces, told people clearly, and given them a window to learn it where learning is actually possible? If the answer is yes, staff use it. If the answer is no, the issue was never whether the software was any good.
What does real change management look like inside an agency?
None of what follows requires a particular platform. These are things an agency owns, and they work the same way, whichever software you end up buying.
Before go-live
Map the process you have now. My years selling business process software taught me that process change fails when nobody wrote down the process. You can’t replace a workflow you’ve never described. Spend an afternoon on it. You’ll usually find three people doing the same task three different ways, which is useful information before you’ve paid for anything.
Pick an internal champion who has authority. Not the most enthusiastic person; the person who can tell an AM that the new way is the way, and have it stick. Prosci’s benchmarking research has ranked active, visible sponsorship as the top contributor to change success in every edition of its Best Practices in Change Management study since 1998.
Define value in numbers, up front. When I was a business value engineer, we pulled everything back to three levers: savings, sales, and satisfaction. Are we saving money by reducing manual effort? Are we growing by creating more opportunities and converting more of them? Are we improving the experience for employees, clients, or both? Pick one lever and one number under it. Quotes per week run in the new system is a perfectly good place to start. A rollout with no number attached to it can’t succeed or fail, which means it will quietly fail.
Pick the window around renewals, not inside them. Urgency helps, since agencies engage when Q4 is close. But the version that works is doing the prep in slower months, so the process is already established by the time renewals land.
During rollout
Train on your own groups and your own carriers. Generic demo data teaches people how the software works. Your real cases teach them how their job works now, which is the thing you actually need them to learn.
Start with one workflow, end to end. One real case, all the way through, before you expand. People need a complete trip, not a tour.
Let the people doing the work show you where training should focus. I always start with the outcome: what should somebody be able to do differently after this session? Then I bring in the people who do the work every day, because they know where it hurts. That involvement does something else too. It creates internal change champions, people their colleagues trust, who can explain why the process is changing in terms that land.
After go-live
Check in at 30, 60, and 90 days. Put them on the calendar before go-live, while everyone’s still paying attention.
Keep training alive for new hires. Whoever onboards your March hire will teach them something. Decide now whether it’s the new process or the old one.
Retire the old process deliberately. Pick a date, say it out loud, archive the spreadsheet. This is the step most agencies skip and the one that decides the whole thing.
One expectation worth setting. In my experience, it takes most teams eight to twelve weeks to get fully confident on a new quoting platform, and the range depends almost entirely on how much real work they run through it early. Teams that apply training to live cases in week one get there fast. Teams that touch it occasionally take much longer, and some never arrive.
What if the rollout has already stalled?
Usually, alignment never happened at the start. One leader believed in the software, and everyone else was left to figure out the goal, the owner, and how anyone would know it worked. What restarts it is that leader turning belief into clear expectations and a consistent process, then staying with the team through the objections rather than announcing the change and stepping back.
What should agencies expect from a software partner after the sale?
Everything above is the agency’s side. There’s a vendor side too, and these questions are easiest to ask before you sign and nearly impossible to ask after.
Adoption is shared work. A vendor who hands you a login and a recorded training has decided it’s your problem. You can find that out in the evaluation if you ask, and it belongs on the same list as the questions you’re already asking about features and price.
Four questions I’d put to anyone (including us):
- Who owns our onboarding, by name? Not which department. Which person, and what happens when they’re out of office.
- What does week five look like? Most vendors can describe week one well. Week five is where you learn whether there’s a plan or a handoff into silence.
- How does somebody hired after onboarding get trained? Agencies turn over account staff. If training only exists as a live session that already happened, your rollout has a shelf life.
- Has anyone on your support team actually worked in benefits? There’s a real difference between support that knows the software and support that knows what a level-funded renewal looks like in July. You’ll feel it the first time something gets complicated.
I’d also listen for how a vendor talks about warning signs, because the useful ones are early and unglamorous. Attendance and logins are the weak signals. The strong one is whether somebody has run a real case and started asking the kind of questions that only come from using the thing.
A partner who is watching that will tell you adoption is slipping in week three. A partner who isn’t will tell you at renewal.
What I’d tell any agency buying new software right now
I’ve also sat on the kickoff call that worked. Same contract, same optimism, but this time there was a name on the rollout and a number to watch. They mapped how they quoted before changing it, started in May instead of October, trained on their own groups, and archived the old spreadsheet on a date somebody announced. At week six, people were using it because somebody treated the rollout as work.
If you’re setting a 2027 technology budget right now, and a lot of agencies are, I’d spend some of that planning time on the rollout rather than all of it on the evaluation. Decide who owns it. Decide what you’ll stop doing. Decide what number tells you it worked. Then pick your window, and pick it somewhere other than the middle of renewals.
My colleagues have written about two related pieces of this. Curtis has written on what honest AI looks like in insurance, and Justin on what clean, consistent data makes possible. Both depend on the same thing: a team that actually opens the platform.
The best tool is the one your team is still using in month six.
Chaz Ripoll is VP of Customer Success at PerfectQuote. He has spent about twenty years in enterprise SaaS, including business process software at major enterprise technology companies and nine years building customer success at a global customer engagement software company.
Explore more practical guidance on choosing and rolling out benefits technology.