Built by Logan / Case study / Business intelligence

Business Bottlenec Bottleneck

Most businesses optimize symptoms. This advisor software finds the real operating constraint, tests what happens when it is relieved, and turns the analysis into a versioned, evidence-linked assessment a client can act on.

DBusiness intelligence
CSoftware you own
BHard-to-get data, made usable

$400,000+: what a dedicated development team would charge to build it
$35,000: what Logan would charge to build it for you
About$110: in software and AI usage to build it internally (your team's time excluded)
4weeks: from first commit to a working multi-tenant advisor platform, built alongside other projects

Team figure is the founder's estimate of what a dedicated team would charge, not a quote. Internal figure is reconstructed from usage records. 228 commits across 5 active build days, plus a three-day pressure-model sprint in July.

What problem does it solve

Everyone can see the symptoms. Nobody agrees on the cause.

An owner feels the pressure in sales, delivery, cash, and people at the same time, and every advisor has a favorite fix. Spreadsheet models hide their logic, benchmarks assume every business is average, and the recommendation changes with whoever built the model last.

What impact does it have

48 metrics. 8 domains. 12 levers. One answer.

A deterministic engine takes a baseline P&L and 48 operating metrics, calibrates them to the client's industry, and ranks the constrained domains. Advisors pull 12 strategic levers and see the cascaded before-and-after P&L, guardrails, and dollar attribution. The methodology is backed by 974 passing tests, so the same inputs always produce the same answer.

What value does it have

A methodology the practice owns.

The engine, the industry calibration, the document intake, and the report studio are one platform, multi-tenant from day one, ready for a firm of advisors rather than a single consultant. A newer pressure model follows requested growth through the operating chain to find the first thing that breaks. Every assessment is versioned and every number traces to its input.

What it does today

A working advisor platform, not a spreadsheet. Everything below runs today.

Advisor workspace

Dashboard, client records, contacts, notes, and a visual engagement pipeline
Organization administration with role-based access and audit history
Custom metrics and configurable engagement phases
Multi-tenant from the ground up, with row-level security

Assessment engine

Baseline P&L plus 48 operating metrics across 8 business domains
Bottleneck ranking with severity and headroom per domain
12 strategic levers, including three dual-mode levers, with guardrails
Six-stage deterministic pipeline: headroom, lever effects, cascading, reconciliation, guardrails, attribution
Focused snapshots and immutable finalized versions for comparison

Data and calibration

Document-to-metrics intake: PDF, CSV, Excel, and images extracted with confidence and source location
Split-view review where an advisor accepts, overrides, rejects, or redirects every value
Industry calibration from a plain-language description, with cached researched profiles and provenance
Curated Default, SaaS, Services, and Distribution profiles

Reporting

Branded PDF with executive summary, P&L, lever attribution, data quality, and commentary
Report preview, history, delivery record, and side-by-side version comparison
Pressure assessment: 82-field baseline, scenario engines, questionnaires, and an eight-page report

On the roadmap

Promote the pressure model after pilot comparison
Close the remaining report coverage gaps
Benchmark context from the Industry Benchmarking Engine on every assessment

Why owning it matters

The method is the product.

An advisory practice that rents assessment software rents someone else's opinion of what matters. This engine encodes the practice's own methodology, and it can change when the thinking changes.

Deterministic, so defensible.

When a client asks why the model says operations before sales, the answer is a calculation with inputs, not a consultant's intuition dressed as a chart.

Built for a firm, not a founder.

Organizations, roles, audit, and versioning were in the first build. Adding advisors adds capacity; it does not add licenses.

Why the operator built it

Why the person who does the assessments built the software.

The methodology existed before the code.

Years of walking businesses through their constraints produced a way of thinking that no vendor's template matched. Describing that method to AI, lever by lever and guardrail by guardrail, turned it into a testable engine in weeks. A dev team would have needed the method explained first and would have flattened it in the retelling.

Every edge case was a real client.

Dual-mode levers, industry calibration, document intake that refuses to trust the model: each of those exists because a real assessment needed it. When the person who hit the edge case is the person building, the fix lands the same day instead of in the next quarter's backlog.

Four weeks, and the firm-grade parts came free.

Multi-tenancy, audit trails, and versioning are usually the expensive second phase an agency sells after the demo. Here they were in the first four weeks, because asking AI for them costs a sentence, not a change order.

That is what the Vista AI Cohort teaches: operators building their own software, with a community for the parts that need a second set of eyes.