Why Does the Simple Internal Panel Sit in the IT Queue for 18 Months?

There is no simple panel

The sentence is short: a login, a list, one approval. Once writing starts, roles, exceptions, and “who may see this record” appear. A shared admin account is not a process. An unowned Notion line does not finish work.

Eighteen months is not a made-up case name. It is a familiar rhythm. There is no discovery, another ticket jumps the sprint, and scope swells with “while we are at it”. The panel does not stall. The queue eats it.

This article does not test whether the spreadsheet has already broken. That split is in when spreadsheets should become an internal panel. The question here is why the internal team cannot finish — and what it takes to finish it outside the queue.

The roadmap will always prefer the customer product

Internal IT has to watch the product the company sells. A customer-facing defect, a legal deadline, a revenue hit — those legitimately crush the queue. The operations panel stays “when we have time” because it is not urgent. It is only expensive.

The ticket system rewards the same mistake. Each request is its own card. A small card looks like small work. Joined together it is product scope: roles, audit, one integration, exceptions. Small cards do not add up. They overwrite each other.

Without an owner, time stretches. Three units comment. Nobody locks a single list. The round looks finished. The rule is not.

How “while we are at it” becomes an ERP

If build starts before scope locks, every meeting adds a screen. Connect accounting. Ship a mobile app. Let the warehouse see it too. The work stops being a fixed core. The internal team is right to pause: this is no longer a side project. It is a second product.

Not writing is sometimes correct. If a ready tool already covers 80% of the process, the panel should not enter the queue. What often inflates the queue is the wrong candidate.

If you do write it, discovery should still include the option to buy instead. Otherwise the 18 months are spent slowly growing the wrong job.

What actually shortens the queue

More developers do not shorten it. A locked process does: one flow, who may see which records, one feedback owner, preferably one connection. A new request is extra work, not a revision. An ERP and a second integration stay out from day one.

Weekly demos let the team review with real records. Feedback is one list. The runbook and source code stay with you at handover; the product lives in your Git. That is not shutting IT out. It is finishing a core the queue cannot swallow.

Fixed scope and the delivery rhythm are written on the custom software page. This article does not replace that package. It explains why the work sits inside for years.

Next step

If the panel is still in Notion, the next move is not another ticket. Lock the process, the exceptions, and the system to connect on one page. Discovery → fixed scope → demo → runbook is the work the queue does not do.

When you are ready, read the custom software package, then talk through the one process.