Book a call

How we help

Four stages, and you may not need all of them. We start where you are and build on what already works. Timings are rough and depend on your data and systems.

Learn the business

Usually 1 to 2 weeks

Where most clients start

Business & Data Assessment

Fixed fee · credited toward your build
Book a call →
  • How you make money and where decisions get made
  • Your systems, data, and reporting today
  • Where your notes, docs, and threads live
  • A prioritized plan with timelines

Build the foundation

Usually 4 to 10 weeks

AI-Ready Data Build

Your key systems flowing into one place automatically, checked every time they refresh.

  • Pipelines from your key systems
  • A warehouse sized to your data
  • Tested data models

Data Acquisition

When the data you need isn’t in your systems, we go find it, buy it or start collecting it.

  • Public and third-party data, like demographics and market data
  • Data vendors compared before you buy
  • Lead and account lists built from your CRM and outside data
  • New CRM and ATS fields when the data has to start with you
  • Facts pulled from emails, calls and PDFs with AI, then checked

Context Layer

The why behind your numbers. Notes and calls become searchable, so answers come with the story.

  • Notes, transcripts, and threads, synced on a schedule
  • Each linked to a metric, customer, and date
  • Who sees which notes, agreed with you up front
  • Answers that can cite the note behind them
What most companies lack

KPI & Semantic Layer

Every key number defined once, so every report and every AI answer matches.

  • A metric dictionary, reviewed with finance
  • Business terms defined in plain English
  • Defined once in code (dbt, Cube or your warehouse), so dashboards and AI match
  • Changes versioned and tested

Data Migration

Moving to a new CRM, ERP or accounting system? We handle the data side, so your history comes with you and the new system starts clean.

  • Old fields mapped to the new system and signed off by the people who use them
  • Duplicates and old records cleaned up before the move
  • Test loads checked until counts and totals match
  • History kept in your warehouse, with a way back if cutover goes wrong

Why the foundation comes first

“Can’t we just upload our spreadsheets to ChatGPT?” It’s a good start. But a spreadsheet is a snapshot, and the AI has to guess what your numbers mean. A data foundation keeps them current and defined once, so it doesn’t have to guess.

Put AI to work

Usually 2 to 4 weeks once the foundation is in place

Dashboards & Insights

The numbers you check every week in one place, with what changed and why.

  • Dashboards on governed metrics, in the BI tool you use
  • Weekly KPI digest and monthly CFO brief, with commentary
  • Alerts when a metric moves, with likely drivers to check
  • Sponsor and board reporting packs
Example marketing dashboard on mock data: revenue, spend, return on ad spend, leads, new customers and cost per customer, then weekly revenue by channel, revenue mix, a conversion funnel, return on spend by channel and a campaign table.
Example dashboard · mock data

Growth at a glance

Last 30 days · mock data

Revenue
$611.3K
▲ 20.2%
Spend
$91.5K
▲ 8.5%
Return on spend
6.7x
▲ 10.8%
Leads
5,816
▲ 11.9%
New customers
312
▲ 22.8%
Cost per customer
$293
▼ 11.7%

Revenue per $1 spent

  • Events24.2x
  • Email14.8x
  • Paid search6.5x
  • Paid social1.9x
Open the full dashboard

AI Analyst

Your team asks questions in plain English and gets answers from the numbers you’ve agreed on.

  • Built on Claude, ChatGPT, or similar, in your accounts
  • Your team owns the account and can switch it off
  • Sees approved metrics only, through a standard connector (MCP)
  • Read-only, each team sees only its own data, and queries are logged where your warehouse allows
  • Test questions re-run when your data changes
Example AI Analyst conversation on mock data. An HVAC owner with three Dallas–Fort Worth locations asks which locations and job types drive margin. The answer says Plano leads and Fort Worth is the problem, shows margin by location and by job type, finds about $520K a year left on the table, and says what to do first.
Example AI Analyst answer · mock data

Machine Learning

Forecasts and scores that show what’s likely to happen next. Best with a couple of years of history for forecasts and a few hundred past outcomes for scores.

  • Revenue, volume, and cash forecasts
  • Lead and deal scoring in your CRM
  • Churn, risk, and anomaly detection
  • Monitoring and retraining planned before hand-over

Data Activation

Your data put to work in campaigns, sales follow-up and outreach. Set up in your own accounts, and every sync has an owner, a log and an off switch.

  • Audiences built from what customers bought, like repeat buyers or ones about to lapse
  • Segments synced to your ad and email tools on a schedule, not by hand
  • Prospect lists ranked against your best customers and loaded into your CRM
  • Scores and account details right on the CRM record
  • Sales sent back to ad platforms so they optimize for buyers, not clicks

Keep it running

Month to month, for as long as it helps

Data Team on Retainer

We keep what we built healthy as your business changes.

  • Pipeline monitoring and fixes
  • Warehouse, model and test upkeep, plus cost reviews
  • New sources, metrics and reports as priorities change
  • Testing and retraining for AI and ML models

Fractional Head of Data

Senior data leadership, part time, without a full-time hire.

  • A data and AI roadmap tied to your business goals
  • Direction for your in-house team, contractors or vendors
  • Advice on data hires, tools and vendors
Illustrative examples

What this looks like in practice

Questions a team might bring us, and what we’d connect and build to answer them. Examples, not client work.

The question

Which campaigns turn into revenue, not just leads?

What we connect

Google Ads, Meta, direct mail drops and response codes, and your CRM.

What we build
  • Attribution from first touch to closed deal
  • Direct mail match-back
  • Cost per closed deal by channel

Focused Projects

Have one question, not a build? We take on fixed-scope, fixed-fee projects, like a lead list from your CRM or tying deals to the campaigns behind them. Done in your tools, and you keep the work.

Not sure where to start?

That’s what the Assessment is for.

Book a call →Why us