Tech debt with a number attached

Every platform owner has made this pitch. Some version of: the org is a mess, we need time to clean it up, things are going to break if we don't.

It rarely works, and the reason has nothing to do with whether it's true.

You're asking a CFO to fund something described as "a mess" against projects described as "$1.4M in pipeline influence" and "12 days off the quote cycle." One of those has a number. The tech debt request has an adjective and a feeling of dread.

So it gets deferred, again, and the dread compounds.

The list you'd get today

A mapped org produces something different: an inventory with counts, ages, and dependencies. Not a characterization of the problem, an enumeration of it.

Here's roughly what a mid-size org surfaces.

41 Workflow Rules still running. Salesforce retired the tool. Your org didn't get the memo, because nothing broke. Each one is a piece of automation that fires in an execution order nobody controls, and none of them are visible to anyone building something new on the same object.

8 Process Builders doing work one before-save Flow could do faster. Process Builder is slower, and it consumes CPU time from the same budget your agent actions draw on.

12 Apex triggers on Opportunity. 3 with logic that overlaps. 1 with 6% test coverage that passes deployment because the org-wide average carries it.

Managed packages nobody has written to in 14 months. Still on the renewal. Still in the security review. Still a dependency somebody has to check before a release.

340 custom fields with data in fewer than 1% of records. Some are dead. Some are quarterly-use fields that Finance would notice immediately. The map tells you which is which, which is exactly the question that stops most field cleanup projects in week 2.

Circular dependencies between automations on Case and Account. The kind that has been fine for years because human throughput never hit the limit.

Now you have a document. It fits on two pages, every line has a count, and each one can be assigned an estimate. That's a fundable thing.

Putting cost against it

Four budget lines carry your tech debt, and three of them are already being paid.

SI and contractor spend. This is usually the biggest number and the least examined. Look at what your partner actually bills for on a change request. A meaningful share is discovery: reading the org, tracing what a change touches, working out what will break. It's charged as analysis, and it's the same analysis every time, because nothing captures the result in a form the next engagement can use.

License and package spend. Managed packages nobody uses, seats assigned to people who left, and overlapping tools bought by two departments that never compared notes. This is the smallest number and the easiest to recover, which makes it a useful opening move.

Incident cost. Production breakage traced to a change whose blast radius nobody checked, plus the rollback, plus the week of credibility you spend rebuilding with the team that got broken. The direct cost is small. The second-order cost is that every subsequent change gets slower, because the team has learned that changes are dangerous.

Delivery drag. The one that never gets measured. Every new build starts with archaeology, so every estimate carries a discovery tax. Ask your team what percentage of a typical ticket is understanding versus doing. The answer is usually somewhere north of half.

That fourth line is where the compounding lives. Tech debt's real cost isn't the debt. It's the interest rate it puts on everything you build afterward.

Making remediation cheap enough to do

Knowing the list doesn't clear it. Somebody still has to migrate 41 Workflow Rules to Flow without breaking the automation order, consolidate 12 triggers behind one handler, and refactor Apex that has no tests to refactor against.

Historically, that's a quarter of work and a real chance of a production incident, which is why it keeps losing to the pipeline project.

Tribal does the remediation with the map as context. Workflow Rules and Process Builder consolidated into Flow, with execution order preserved. Triggers merged behind a single handler. Apex refactored and tests generated. Impact simulated against the rest of the org before anything moves, so you know what a change touches before you make it rather than after.

Customers see roughly 10× faster tech debt clearance. Almost all of that comes from the archaeology no longer being the expensive part, and the maintenance cost afterward runs about 80% lower because what replaces the mess is something a person can read.

"Tribal helped us remove tech debt by replacing brittle workarounds with a clean, maintainable setup, and it gave our team a real speed boost, about a 5× upgrade, so we can show the business we're listening."

‍David Kestenberg, Director of IT, Pro-Driven Brands

The cost that doesn't appear on any line

There's one more, and it's the one I'd lead with in the conversation with your CIO.

Look at what your best people spend their week doing. The admin who understands why the escalation rule has an exception for Quebec, who knows which fields Finance actually depends on, who can tell you in one sentence why the renewals process looks strange. That person spends most of their week doing archaeology on a 5-year-old Flow.

That's a strange assignment for someone who understands your business better than almost anyone in the building. Hand that week back and you have a different kind of team, working on the problems you hired them for.

Which is also, incidentally, the answer to the AI question your board is asking. The org that can't clear its own tech debt is not going to deploy agents successfully, because agents inherit every bit of it. Governor limits that humans never hit. Permission gaps nobody exercised. Automation chains nobody mapped.

Clearing the debt is the AI project. It just doesn't sound like one, which is why it's still unfunded.

Start by getting the list. Point Tribal at a sandbox, ask what's in there, and put the two-page inventory in front of the person who controls the budget. The argument makes itself once it has numbers.

‍