Don’t Buy Another Tool Until One Model Works

A new platform will not save a messy model. It will host the mess with a better logo.

If you cannot name one semantic product the company trusts through a real meeting, you are not ready for another SKU. You are ready to finish a loop: source, model, steward, refresh, decision.

This is not the CEO platform call, and it is not a capacity order

Fabric vs Power BI is about whether you run one analytics platform or a pile of files, covering ownership, landing zone, and enough runway for four quarters. Premium capacity is not a strategy is about headroom versus product. We will not retread either one here.

This post is about the sprawl of tools: another catalog, another notebook environment, another “AI layer,” another departmental warehouse, another visualization toy, each bought because the last stack did not produce a number the room would use. The gap was never missing software. It was one unfinished model.

If the semantic model is the product does not exist yet, every new tool is just a brochure printer.

What “one model works” actually means

“Works” means a short contract, not a refresh on your laptop.

You pick one domain, such as the finance close pack, plant throughput, or inventory cash. One, not twelve.

You name one steward who can freeze the grain and the sentence for the measures that meeting uses.

You set one refresh window the meeting can see, with a visible fail state.

And you ship one app or connected Excel view that leadership actually opens. Usage is evidence. A demo is not.

If you cannot point to that loop in production, a second platform is just a second place to fail. Analytics programs fail in delivery, not for want of a feature.

What buying ahead of the loop costs

1. You migrate arguments. Five revenues in the old tool become five revenues in the new one, and the meeting still starts with whose file is right. Every team building their own model survives any catalog.

2. Delivery pauses for procurement. While legal and IT score vendors, month-end is still a scavenger hunt and inventory is still late. The reporting drag at close does not wait for a bake-off, and the opportunity cost hits the P&L first.

3. The pile gets a forwarding address. Every pbix, extract, and shadow warehouse is promised a new home, and nothing is retired. You paid a transformation tax to relocate dashboard sprawl.

4. Skills fragment before a standard exists. Analysts learn the new canvas while the official number still lives in last year’s file. You train two stacks and trust neither, and you can end up with unused licenses in both.

5. Leadership learns that “analytics” means shopping. The board sees software while the plant still runs on a clipboard. After two purchases, the next proposal is dead on arrival, or worse, rubber-stamped out of sunk-cost reasoning. Strategy turns into a catalog.

6. The one loop that would have proved value stays unfunded. The modeler who could have frozen margin, on-hand, or the huddle grain is busy on a migration. Proof of value has its own post. The rule here comes first: do not buy a second factory until the first line ships.

Skip the ROI on the vendor slide. Check whether last year’s purchase changed the Monday meeting. If it did not, the next purchase is a habit.

What not to do

Do not start with a feature matrix. Start with the meeting that already exists and the number it still pastes.

Do not wait for a perfect lakehouse before you certify one model on the sources you have. Waiting for perfect landing is how the pack stays a paste.

Do not add a conversational layer on an untrusted model. Chat on fog is still fog, as covered in conversational analytics needs a trusted model.

Do not treat a departmental tool as a harmless sandbox when it publishes “official” numbers. Sandboxes are labeled. Official means a steward.

Do not treat “the team wants to learn X” as a capital request. They can learn on the certified model you already own.

If you cannot name the decision the next tool is for, you need a roadmap, not a cart.

How to prove one loop before you shop

1. Pick the ugliest recurring meeting. The close flash, Monday ops, or the inventory cash review, whichever already burns the most calendar. Write down the three questions it must answer. If you cannot, you are not ready to model, let alone buy.

2. Build or repair one semantic product for those questions. Freeze the grain, write the measures in plain English, and name the source. Single source of truth is a decision, and dual publishers mean no publisher. Park every other canvas.

3. Put a steward and a clock on it. Decide who may change the sentence, when the data must be fresh, and what the room sees when it is not. Ownership is part of the product, as in why your Power BI project has no owner. A tool with no owner is how you got here.

4. Run the meeting on that product for a full cycle. That means a real close, a week of huddles, or a cash meeting, not a demo. Connected Excel is allowed. Parallel rebuilds are not. If people still paste, you have a behavior and definition gap, not a tool gap.

5. Retire the twins the loop replaced. A working model next to three unofficial copies is not proof. Proof is the copies gone, or labeled as working paper with an expiry.

6. Only then ask whether the platform is the constraint. If the loop works and the constraint is landing, scale, or a real shared data product across domains, you have a platform conversation, so have it. If the loop does not work, a new SKU will not create one. Capacity and architecture come after evidence.

A bloated first model is cheaper to fix with dashboard optimization than a second estate is to build. Ongoing ownership is what Managed Data & AI Advisory is for.

What good looks like

One domain, one model, and one meeting that no longer pastes.

The next vendor has to explain why that loop is not enough, and most cannot. When you do expand, you expand a product, not a shopping list.

If the strategy slide is a tool landscape, you do not have a strategy. You have a postponement.

Frequently asked questions

What if our current tool is actually the blocker? Then you can name the job it cannot do after one model was honestly tried. “People are bored” is not a blocker. “We cannot share a certified model across finance and ops” might be. Prove the model first anyway, because a migration without a product just moves the paste.

Can we sandbox a new tool while we finish the model? A labeled sandbox that cannot feed the pack is fine. A silent second “official” source is how you buy sprawl. Fence it or do not start it.

Isn’t one model a bottleneck? One certified model per domain is a bottleneck on arguments, which is the point. Exploration still happens on that model. Forking datasets is the old habit.

How does this relate to unused licenses? Seats and new tools both fail when there is nothing trusted to open. Build the product first, then match tools to the jobs that use it.

Get started

Prove one trusted loop. Then shop, if you still need to.

Not sure whether you are buying a platform or postponing a model? Book a session. We’ll map one domain, one steward, and the meeting that should already run on it. Or start with a free Model Health check.