Looking for a business intelligence consultant?
Here's the honest answer.

If you need dashboards, reports, or a data warehouse, we're the wrong firm — and we'd rather tell you that here than twenty minutes into a call. Integral doesn't do traditional business intelligence. We build AI systems that act on your data instead of charting it, and the rest of this page is about how to tell which of those two things you actually need.

What you were probably searching for

In IT, "business intelligence" is a specific discipline. It means pulling data out of your accounting system, your CRM, and your operations software, cleaning it, storing it somewhere sensible, and putting charts on top so the people running the company can see what happened and why. Power BI, Tableau, Looker, a data warehouse underneath. It is mature, useful work, and there are good firms in Los Angeles and Orange County who do nothing else.

If that's your project, hire one of them. Three questions are worth asking before you sign:

• Who owns the data model when the engagement ends? The charts are the cheap part. The logic that defines "active customer" or "gross margin" is the asset. Make sure it's documented and yours.
• What happens when a source system changes? Dashboards break quietly. Ask who notices, and how fast.
• Who is going to look at this every week? A dashboard nobody opens is the most common outcome in the category. Name the person and the meeting before you build.

So why is "Business Intelligence" in our name?

Because we mean the words literally. The intelligence we work with is the kind that reads, reasons, answers questions, and gets things done: artificial intelligence. The business is where we put it — not a research lab, not a consumer app, but a company with real operations, real records, and real consequences when something goes wrong. We integrate one into the other. That's the whole name.

We know the phrase was already taken by an industry built around dashboards. We don't do that work, and nothing on this site should suggest we do. The longer version of the story is on our name page.

Business intelligence, as a category, helps you understand your data. We think the more interesting software is the kind that does something about it.

The gap a dashboard leaves

A dashboard can tell you that forty invoices are past sixty days. Someone still has to chase them. It can tell you the intake backlog doubled in March. Someone still has to type those forms into the system. It can show that support tickets about one policy tripled. Someone still has to find the policy, read it, and write back.

That last mile — the part after the insight — is where most small and mid-sized businesses actually spend their payroll. It's also the part that reporting tools were never designed to touch. They end at the chart.

What software that acts looks like

Less dramatic than it sounds. In practice it's things like:

• Documents that file themselves. An invoice, an application, or a signed form arrives. The system reads it, pulls the fields, checks them against your rules, and puts the record where it belongs. A person reviews the exceptions.
• Answers from your own records. Staff ask a question in plain English and get an answer drawn from your policies, contracts, and procedures — with the source passage attached, so they can check it.
• Drafts waiting for approval. The routine reply, the status update, the follow-up request for a missing document — written and queued for a human to send or fix.
• Work that routes itself. New requests get classified and sent to the right person with the context already gathered.

None of this removes people from the decision. It removes them from the retyping, the searching, and the first draft.

Why where it runs matters more here

A dashboard works on totals. An AI system works on the documents themselves — the contract, the client file, the application with a Social Security number on page two. That changes the privacy question completely, and it's why we build these systems to run in-house: on a server in your office or on private infrastructure you control, not on a public chatbot.

We hold ourselves to the same standard. The assistant on this website runs on a single 24GB graphics card in our own office — a 27-billion-parameter open-weight model, compressed to fit, using about 23 of those 24 gigabytes and starting its answer in well under a second. The first time we loaded it, it spent its entire response budget "thinking" and returned nothing at all until we changed one startup setting. That's the kind of thing you'd rather your integrator had already been through on their own hardware.

A quick test for which one you need

Listen to how the problem gets described in your own meetings. If the sentence starts with "show me" — show me revenue by region, show me which jobs ran over — that's business intelligence, and a BI consultant is the right hire. If it starts with "who is going to" — who is going to enter all of these, who is going to read all of that, who is going to answer these emails — that's the work we do.

Plenty of companies need both. They're different projects, bought from different firms, and it's fine to do them in either order.

What it costs to find out

We publish our ranges. A single basic automation starts at $500. A private knowledge base over your own documents starts at $2,500. If you aren't sure where to begin, our AI Readiness Evaluation is a fixed $2,500, two-week engagement that ends in a written, prioritized roadmap you can act on with us or without us. The full list is on our pricing page.

We're based in Southern California and work on-site across Los Angeles, Orange County, the Inland Empire, and San Diego. We build for organizations that handle records they're obligated to protect — law firms, clinics, accounting practices, property managers, nonprofits. You can see how the work differs by field on our industries page.