Your Looker Is Not Business Intelligence

It is Friday, 7:52 AM. Your Head of Data opens Looker before the exec review. 1,840 live dashboards across 42 folders, 420 of them untouched in the last 90 days, 94 tiles broken against the last schema migration, 68 metric definitions that drift across three different Explore files, and the CFO's board brief sitting at 61 percent to the Q4 EBITDA commit with five weeks on the clock. The Head of Data opens three dashboards at random, screenshots the revenue tile into Slack, and walks into the 9:00 AM. The three dashboards are 0.16 percent of the shelf. The board pack ships against the other 99.84 percent.

Every operator who confuses Looker with business intelligence is quoting the wrong tool. Looker is a modeling layer with a decent dashboard UI on top. It renders the chart. It does not run the function.

What Looker renders vs what the function owns

Looker renders 1,840 dashboards, a LookML repo nobody has refactored since the Series B pricing change, and a folder tree with 24 subfolders named after people who left the company. Business intelligence is the function that lands the CEO, CFO, Head of Revenue, and Head of Product on the three decisions a week that move the quarter, ships the Monday executive brief with the five metrics ranked against the board commit, and routes every "this number looks wrong" out of Slack into a resolved exception by Thursday.

The gap is not a Looker feature request. The function reads six systems. Looker for the modeled layer, the warehouse for the raw joins, the CRM for pipeline, the billing system for the collected revenue, the HRIS for the loaded cost base, and the Slack channel where every "why is this number down" lands between the Monday and Thursday exec syncs. Looker is one of six inputs. The function is the synthesis across them.

The 28 hours a week the data team spends below the exec line

A $260K Head of Data reads $360K to $460K loaded against the US entity. Three analysts under the seat each read $140K to $180K loaded. The scope reads reporting, forecasting, and executive decision support. The lived calendar reads 28 to 36 hours a week on ticket work the dashboard was supposed to answer on its own.

Metric drift. 68 metric definitions live across the repo, three different versions of "active account" ship in three different Explores, and the number the CFO cites on Thursday is not the number the Head of Revenue cites on Friday. Nobody has reconciled the dictionary since the last pricing migration six months back.

Dashboard rot. 420 of 1,840 dashboards sit untouched in the last 90 days, 94 tiles broke against the last schema change, and the quarterly board dashboard still points at a column renamed in July. The three screenshots in the Monday Slack were last audited by an analyst who left the company in Q2.

Ad-hoc ask queue. Each analyst catches six "quick question" asks a week. 18 asks across the team. 14 of the 18 are variants of a question somebody already answered two quarters ago. No answer library exists, so the queue regenerates itself every Monday.

Forecast reconciliation. The CFO forecast and the Head of Revenue forecast reconcile within two points on week one of the quarter and drift to seven points by week six. Nobody owns the weekly reconciliation. The board sees the gap in the Thursday pack and the CEO spends 40 minutes of the exec sync bridging the number.

What the function should ship on cadence

Monday 8:00 AM. Executive metric brief. The five headline metrics ranked against the quarterly commit day-of, the three segments shifting by more than a point week over week flagged for the Head of Revenue with the proposed call pre-cited, and the Monday exec-sync agenda drafted with the two decisions the leadership team is being asked for.

Tuesday 8:00 AM. Forecast reconciliation. The CFO plan, the Head of Revenue bottoms-up, and the warehouse actuals aligned on the same cut of data, the three accounts most likely to slip the quarter flagged with the renewal date and the owner, and the two expansion deals ranked by expected close value.

Wednesday 8:00 AM. Dashboard-health brief. The 1,840 dashboards scored against the last 30 days of views, the 420 untouched flagged for archive, the 94 broken tiles routed to the owning analyst with the schema diff pre-cited, and the 68 metric definitions scored for drift against the central dictionary.

Thursday 8:00 AM. Board-grade brief. The five quarterly commits scored against the trailing six-week run rate, the two commits most at risk flagged with the proposed scope cut or acceleration play, the four exceptions the audit committee will read on Friday pre-cited, and the three decisions the board is being asked to ratify.

Friday 8:00 AM. Weekly review brief. The three analysts scored against the trailing 60-day ticket cycle time, the two question categories most likely to recur the following week pre-answered into the library, and the four decisions ranked for the following week by expected revenue movement. The Head of Data reads the brief at 8:09 AM and the Monday exec sync opens with the ask already drafted.

The unit economics on the function

A $260K Head of Data reads $360K loaded. Three analysts at $160K each read $480K across the team. 28 hours a week on reconciliation and ticket work across the four seats at a loaded hour of $140 to $220 costs the function $300K to $460K a year on queue work the modeling layer was supposed to run. The quarterly EBITDA commit sits in the $4M to $7M range. One point of commit missed because the forecast reconciliation drifted is $40K to $70K a quarter and a board conversation nobody wants to have.

A 14-day sprint to stand up the BI synthesis agent runs in the low to mid five figures. Ongoing cost lands at $3K to $7K a month on API spend, read access to Looker and the warehouse, and the ops operator who owns the Monday brief review. Year one lands three to five times cheaper than the next analyst hire. The Head of Data and the three analysts stop running the ticket queue and land on the modeling and the forecasting work the ledger was paying them to run on in the first place.

What the sprint ships in 14 days

Day one, the agent reads against Looker, the warehouse, the CRM, and the billing system. Day seven, the first Monday executive brief lands in the CEO inbox with the five metrics ranked and the three segment moves routed. Day fourteen, the Thursday board-grade brief lands at 8:00 AM with the two at-risk commits summarized and the Friday review brief sitting on cadence.

If your Looker reads 1,840 dashboards, 420 of them stale, and the Thursday board pack still ships on three screenshotted tiles, the version where the brief lands at 8:09 AM is fourteen days away. Looker still renders the chart. The warehouse still holds the raw joins. The agent closes the function the stack was already supposed to close. You can keep running the Monday exec sync on three screenshots, or you can scope a sprint and have the Friday review brief landing before the next board read.