A delayed Salesforce project has a new go-live date in writing. A stalled one keeps moving the old date and cannot tell you why.
At Fortimize, we have spent too many hours helping good banks and credit unions recover from Salesforce implementations that looked fine on the status report. The pattern is almost always the same, and it shows up months before anyone says the word stalled out loud. At this point, teams typically need a formal Salesforce rescue services engagement to assess the damage and chart a recovery path. So here are the seven signs we look for, what causes them, and what we would do next if we were in your seat. Knowing these signs is easier when you have a baseline for what a proper Salesforce implementation is supposed to look like.
To be upfront: Fortimize gets paid to rescue stalled Salesforce implementations, so weigh this accordingly. It is also why we know what the signs look like.
The 7 Signs your Salesforce Implementation has Stalled
Your go-live date has passed and no one has set a new one, or the date has moved so many times nobody believes it.
A missed date on its own is not a stall. Core banking integrations and data work take time, and sometimes the date has to move. What matters is whether your partner gives you a new date in writing and tells you what changed.
There are two versions of the stall. In the first, the date passes and nobody will commit to a new one. In the second, the date keeps moving, a few weeks at a time, until your team stops planning around it. Both cost you the same way. You keep paying for licenses nobody uses, and the business case you took to your board gets weaker every month.
Scope keeps growing, and change orders show up after the work is done.
Requirements change on every project. That is normal. On a healthy project, the change is written down, priced, and approved before anyone builds it. On a stalled one, scope grows in side conversations and the change order shows up after the fact.
If you cannot trace every extra dollar back to something your team approved, the contract is working against you. Time-and-materials contracts, where you pay for hours instead of results, make this worse, because your partner earns more every time the scope grows.
It does not have to work that way. When we launched one community bank on Salesforce in 14 weeks, we found partway through that some of the data lived in a different system than anyone expected. We changed the plan with no change orders and no escalations. The executives heard about it on the status call that was already on the calendar.
People stopped using Salesforce after launch.
A Salesforce org that is live but unused is still a stall. It just has a go-live date. You see it when relationship managers go back to their spreadsheets, loan officers work around the system, and member service reps only open Salesforce when an auditor asks.
The cause is almost always the same. The system was built around how someone imagined the work, and the training happened in a classroom instead of inside the work people do every day.
Your team does not trust the data.
If your bankers check Salesforce against a core export before every decision, if your leaders will not put the dashboards in front of the board, or if compliance keeps a side spreadsheet, you have a data trust problem.
The rule Fortimize gives every bank is simple: make sure the data is right before you go live. Trust you lose in the first week is very hard to win back. When Credit Union of Texas finished its rescue with us, CIO Angela Faust said it best: “We finally trust our system again.”
Integrations with your core banking system are failing, and duplicates are piling up.
For a bank or credit union, Salesforce is only as useful as its connections to the core, the loan origination system, and digital banking. When those connections break or fail quietly, loan officers see stale balances and service reps pull up the wrong record.
Duplicates are usually the first thing people notice. The same member shows up three times: once from the core, once from the loan system, and once from someone who could not find the first two and created a new record. Now nobody knows which one is right, reports double count, and the household view falls apart. If your team spends time merging records every week, the integration is creating the problem faster than people can clean it up.
The fix starts with how records match. We match people and households on tax ID, not by name or by hand, because a name match will never hold up across a core, a lending system, and a branch. A banker once put it to us this way: she is on four accounts at her bank, but her husband is primary on two and her son on the other two, so if you only look at the primary owner, she looks like she does not bank there at all.
About half of the banking projects we deliver are data and integration work, and this is where generalist partners get stuck. A $4 billion credit union came to us after its Salesforce and MuleSoft build went off track with another partner. We rebuilt the integration, rebuilt the data model around members instead of accounts, and had a working member view in 90 days.
How we took over a stalled MuleSoft rescue for a credit union
The consultants on your project keep changing.
When the people who designed your org roll off, the reasons behind their decisions leave with them. Then the next consultant bills you to figure it all out again. If two or more lead consultants have rotated through your project in under a year, accountability left with them.
We keep teams small, usually five people or fewer, and your own people work with them every day. That way the knowledge stays in your building, not ours.
The invoices grew, and the milestones did not.
Spending more than you planned is not automatically a problem. Spending more without signed-off milestones is. If you have paid invoices beyond your contract and cannot point to a finished sprint, a delivered feature, or an approved change for each one, nobody is accountable. That is the clearest sign your partner’s incentives and your outcomes have drifted apart.
Four things cause most stalled Salesforce implementations at banks and credit unions
The contract pays for hours, not results.
Under time and materials, open-ended discovery and late change orders are what the contract is built to produce. That is why we moved our own business to delivery as a service: a dedicated team, a predictable monthly cost, and scope that can change without a change order. The destination holds, and the route can bend. It is how we deliver 98% of our projects on time and on budget.
Integration was scoped like configuration.
Connecting Salesforce to a core is specialist work. When a partner scopes it in weeks and it takes months, the whole project waits while they figure out what they did not know.
Change management started at go-live.
Change management means helping people actually change how they work, and it has to start in discovery. Tie the project to one number your team cares about, like case time or average balance, and build it with the people who will use it. It is not done to them. It is done by them.
The partner does not know financial services.
Financial Services Cloud has its own data model, its own regulatory requirements, and its own core and lending integrations. A partner who has not run it in production at a bank tends to set it up like a standard sales system. You end up with an org that is technically live and does not reflect how you serve customers.
What to do if your implementation has stalled
Step 1: Stop adding to the pile.
Pause new feature work, hold new scope requests, and write down where every open item stands. A stalled project that keeps collecting half-finished work gets harder to fix every week.
Step 2: Get an outside diagnosis.
Do not ask the partner who was there for the stall to diagnose it. Our Org Reset is built for exactly this. It finds the root cause of the workarounds, shows you the technical debt in your org, and gives you a stabilization plan. It takes about two weeks and two to four hours of your team’s time, and everything we produce is yours to keep, whatever you decide next. If your Salesforce account executive refers you, it may be at no cost.
If you need the full business case, including what another year of the current setup will cost you, our Salesforce ROI Roadmap is a fixed $23,500 and takes four to six weeks.
If your team plans to handle the cleanup itself, start with our 12-step recovery checklist for fixing a broken Salesforce org.
Step 3: Decide whether to replace your partner
Once you have the diagnosis, the partner question gets a lot easier. Sometimes the relationship can be saved with a reset conversation and a new contract structure. Other times, replacing the partner is the right call, and the diagnosis gives the new partner a head start instead of a restart. The question is not whether you like your partner. It is whether they have shown they can deliver the outcome you need, under a contract that rewards them for it.
Step 4: Rescue what you have. Do not start over.
A rescue starts with the org you already paid for. You keep what works and fix the causes, not just the symptoms. Credit Union of Texas stabilized a failing lending platform in 12 weeks. The $4 billion credit union had its member view working in 90 days, and that rescue turned into a multi-year partnership. Starting over is rarely the right answer, and it is almost always the most expensive one.
If you want a lighter first step, our complimentary Technology Health Check is a short workshop that looks at your people, process, and CRM.
Get a written diagnosis in two weeks
The contract pays for hours, not results.
If three or more of these signs describe your project, do not wait for the next sprint. Every month a stalled project runs costs you license spend, your team’s patience, and the outcome you bought Salesforce for.
And before you do, ask your partner one question: what will be live and in daily use 60 days from now?
Once you have confirmed the project is stalled, the next step is a clean operational handoff: our guide on how to switch Salesforce implementation partners walks through every step in the right order.
Frequently Asked Questions
How do I know if my Salesforce project has stalled or is just delayed?
A delay comes with a new date in writing and a reason. A stall has neither, or the date keeps moving. Score your project against the seven signs above. If three or more apply, you are past a delay.
Should we replace our Salesforce implementation partner?
Not before you have an independent diagnosis. Replacing a partner without one means the new team rediscovers the same problems at your expense. Start with an Org Reset, and if the answer is to switch partners, the written findings carry over to whoever takes the project next.
What is the first step when a Salesforce implementation stalls?
Get an independent assessment from a partner who was not part of the original build. The Fortimize Org Reset takes about two weeks and ends with a written stabilization plan you keep.
Can a stalled Salesforce project be rescued without starting over?
In most cases, yes. A rescue keeps what is sound, rebuilds what is not, and protects the investment you already made. Credit Union of Texas and a $4 billion credit union both recovered without a restart, in 12 weeks and 90 days.
Why does our Salesforce org have so many duplicate records?
Usually because the integrations create a new record every time instead of matching to the one that already exists, and people add their own when they cannot find the right one. Fix the matching rules first, ideally on tax ID, then clean up what is already there. Cleaning without fixing the source just means you clean again next month.
How long does it take to get a stalled Salesforce project back on track?
The diagnosis takes about two weeks. Our published rescues reached a stable, working platform in 12 weeks at Credit Union of Texas and 90 days at a $4 billion credit union. Your timeline depends on how much of the problem lives in integrations and data, which is why the diagnosis comes first.