The tech-stack lie every operator inherits.

At a franchise ops shop I walked into a couple years back, the head of ops sat me down in a conference room and pulled up what she proudly called the "master dashboard." It was a Google Sheet. It had four tabs. It was updated by hand every Monday morning by a coordinator named Priya, from data she copied out of three separate tools that didn't talk to each other. One tool was the POS, one was the field service platform, one was the marketing analytics dashboard. Priya spent about six hours a week making the sheet. Twice a year she took vacation and the dashboard silently went stale.

Nobody thought this was strange. This is what a "modern stack" looks like at almost every operator I've worked with. Priya isn't the bug, she's the load-bearing feature.

You'll know your stack is Frankenstein when your team celebrates the day Zapier goes down because meetings get shorter.

The anatomy of the Frankenstein stack

Every operator who's been at it more than two years has some version of the same architecture. It looks different from the outside, but the guts are always the same:

• Seven to twelve SaaS subscriptions. Nobody knows what half of them are for. Two of them were bought by someone who no longer works there.
• A CRM (Salesforce, HubSpot, Zoho, pick one) that's being used at about 20% of its capability. There's a whole automation module nobody has ever opened.
• Four Zapier zaps holding the middle together. Two of them are silently failing 5% of the time and no one has noticed.
• An email marketing tool that has a "contacts" list that's 40% overlap with the CRM's "leads" list, plus 60% of contacts nobody has synced in over a year.
• A Google Sheet, a Notion doc, or an Airtable base that serves as the source of truth because nobody trusts the actual source of truth anymore.
• One person who knows how any of it actually works. Usually named something like Priya, Marco, or Dev.

Look familiar? It should. This is the modal stack of any operator between $2M and $50M in revenue. Not because they made bad choices. Because of how each choice looked when they made it.

Why it happens

Each of those subscriptions got bought by someone at some point who had a real problem. The email tool solved "our newsletters look bad." The CRM solved "we're losing track of deals." The marketing analytics platform solved "we can't tell which ads work." Zapier got installed because two of the tools didn't talk. Each purchase was defensible in the moment.

The problem is that no vendor sells to your specific case. They sell to the general case. Salesforce is built for the general enterprise. HubSpot is built for the general SMB marketing team. Your business is neither of those things. Your business is a specific mix of franchise ops or direct sales or service delivery or e-commerce or all of the above, with a specific data model, run by a specific team who use a specific vocabulary.

So each tool gets configured about halfway. Some fields fit. Some don't. The team names custom fields to bridge the gap. Data gets exported and reformatted and re-imported. Over time the seams show up in the reporting. Numbers stop matching between the CRM and the marketing tool. Nobody can fully explain why. Priya starts a spreadsheet.

Multiply that pattern by ten years of the software category exploding, and you get the Frankenstein stack. It wasn't a decision. It was a slow accumulation of half-fits.

What it actually costs you

The subscription bill is the smallest cost. It's usually the only one people pay attention to. Here's the full picture.

Add all six up and the real cost of the Frankenstein stack is easily 10x the SaaS bill. And the SaaS bill isn't small.

The fix isn't "add another tool"

The instinct, when the stack is a mess, is to buy a tool that promises to unify it. A "data platform." A "customer data platform." An "operating system for your business." I've watched founders spend six figures on these. They rarely fix the underlying problem, because the underlying problem isn't a missing tool. It's that the tools you have are each solving a slightly wrong version of the problem, and no vendor is going to sell you a tool that solves your specific version.

Real consolidation looks like this:

1. Audit. Every subscription, every zap, every hand-updated sheet. What is it doing? Who uses it? What happens if it goes away tomorrow?
2. Consolidate. Kill the redundancies. Most operators can lose two or three subscriptions immediately with no operational impact. That's found money. Renegotiate the rest.
3. Own the data. Pick the CRM (or database) that's going to be the source of truth. Migrate the useful stuff. Kill the rest. This is the boring, painful step. It's also the one that unlocks everything downstream.
4. Replace pretty integrations with real ones. A small number of custom tools, built for your specific data model, that do the actual job. Not a Zap that fires once a day and hopes. A real integration that runs on your terms, in your infrastructure, with a paper trail.
5. Document. The tool doesn't work if only one person knows how it works. Get the SOPs out of Priya's head and into a place the next hire can learn from in a week.

What "custom" doesn't mean

Custom doesn't mean "build everything from scratch." Nobody has the runway for that. Custom means the two or three connective tissue tools between your CRM, your ops system, and your reporting are built for your specific data model, instead of being a Zap somebody set up on a Tuesday afternoon two years ago.

A custom-built ops tool for a franchise operator might just be a small dashboard that pulls from the POS, the CRM, and the franchisee onboarding tool, deduplicates the customer records, and spits out a weekly operator scorecard automatically. It replaces Priya's Google Sheet. It saves six hours a week. It doesn't go stale when she's on vacation. It costs less than one of the SaaS subscriptions the team is already paying for annually.

None of that is exotic. It's just deliberate, applied to the specific case, instead of accepting the vendor's default.

The bar for whether you've fixed it

You'll know you're on the other side when three things are true:

• Priya isn't the load-bearing feature anymore. The system runs whether she's at her desk or on a beach.
• Leadership stops asking "which number is right." The numbers reconcile because there's one source.
• Nobody on the team celebrates when Zapier goes down, because Zapier isn't holding anything critical together.

The whole point of the tech stack is to give you leverage. If you've hit the point where the stack is costing you leverage, the fix isn't more software. It's less software, better integrated, built around your business instead of somebody else's idea of your business.

The data stays yours. The workflow stays yours. When we walk out, the tool doesn't.