I build enterprise revenue from zero, and I am most at home exactly where it is hardest: complex data licensing, six- and seven-figure deals, sales cycles measured in quarters and years, and contracts that outlive the people who sign them.
The work is finding revenue other people cannot see. A data asset built for one industry solves a problem in three others its makers never mapped, and that unmapped ground is where the money actually is. I have sold into dozens of verticals: insurance, mortgage and lending, banking, government, real estate and the secondary market; GIS, drone and satellite, consumer portals and people search; marketing, analytics and modeling; academic, economic, medical and scientific research. That range built a recurring-revenue base from zero to $50M, and I carried quota against it. Tens of millions a year, built, not inherited.
My specialty is licensing massive, complex data assets, and that word is doing real work. Most salespeople, including most sales leaders, do not know the difference between a sale and a license. They call it a subscription and move on. In data the license is the product: what the buyer may do with it, for how long, in which systems, and what happens to it when the term ends is the entire negotiation. I have spent a career on the side of the table where that distinction is worth millions, and I hold it against a buyer’s counsel without handing off.
There is a pattern I only saw looking back: I am usually the underdog’s weapon, and I win from behind. I built Palm’s first website when it was the small challenger against the Apple Newton, and it won by being simpler. I ran web development at Ask Jeeves when it was the natural-language upstart and the search giants owned the category. At DataQuick I sold property data against the category’s giants and learned the move I have used ever since: an underdog can do what the incumbent structurally cannot, hand the buyer the good data and the ugly, arm its rivals’ resellers, let a customer sink real work into the product before a dollar changes hands.
I sold through the dot-com boom and its collapse from inside the Bay Area, close enough to the center of it that I turned up in the Chronicle when it broke. I know a hype cycle firsthand, and I know what separates the people who come through it: a level head when the market turns for reasons no one in the room controls. The AI market is running that script right now, and I have already been the calm person in that room.
I run the whole cycle, not a slice of it. I hire, build, and mentor the teams; I write and enforce the comp plans that quietly decide how those teams actually behave; I carry a number myself and report it to executives and boards. And I forecast the way a CEO wishes their sales leaders would: conservatively, calibrated to real probability, and then I meet it. Set my number beside the hockey-stick everyone else is drawing and mine can look unambitious, right up until the quarter closes and mine is the one that came in. That is the number you build a plan on. Optimism is not a forecast.
Every few years a new methodology arrives promising to turn selling into steps anyone can follow. The steps help; they are not the thing. What closes the deal is judgment you cannot write down, built one hard quarter at a time. Put me in a room full of newer frameworks and sharper decks, and I am the one who knows how to proceed.
I came up on the technical side, software development, data, DaaS and SaaS, now AI, so I can go as deep into the engineering of what I sell as into the economics of why it is worth buying. AI runs on data, and the scarce skill in that market is not the model, it is knowing which data is non-fungible and which only looks that way. That judgment is the asset. And I carry deals through the legal redlines, not only the commercial terms, which is where most sellers hand off and where I am strongest.
The specifics
The detail behind the summary. Open what you want to see.
Data and licensing
- Where the engineer’s job ends and mine begins: I do not architect the provenance pipeline, and I have never pretended to. I understand it well enough to sell the assurance it gives, write it into the contract, and hold the technical buyer who asks how we prove it. That is the salesperson’s half of the data business, and it is the half most sellers cannot cover.
- Warehouse and field-population reports: what is complete and, just as tellingly, what is not; data dictionaries, layouts, currency, sources, and record counts. The documents a serious buyer reads before trusting a feed.
- Row-level data provenance, or origin tagging: every row and field carries an immutable tag for its supplier, batch, ingestion time, and contract code, so origin survives the merge instead of dissolving into an untraceable pool. Standards like W3C PROV make those logs machine-readable and audit-ready.
- Multi-vendor entity mapping: the same real-world entity arrives under three suppliers’ identifiers, and the work is reconciling them into one master record, whether the architecture keeps the sources linked by a relational key or merges them with strict cell-level attribution. Which one a buyer needs is a commercial call as much as a technical one.
- Source waterfalling: when suppliers overlap, a priority ladder decides whose field wins, and because the system logs which source won the placement, per-supplier usage can be metered for licensing audits. The audit trail is a byproduct of the architecture, if the architecture was built to produce it.
- Data lineage and usage tracking: automated lineage tools, the OpenLineage and dbt family, trace which suppliers fed any given extract, so an origin trace on a bulk download is a query, not a forensic project. It is the difference between a licensing audit you can answer and one you cannot.
- Contractual enforcement, where the schema meets the contract: contract-lifecycle management tied to the live schema flags when a supplier’s data is heading somewhere the license does not allow; synthetic salted records prove origin if a set is leaked or lifted; token-level API granularity meters exactly what each licensee consumes. This is the part I own outright.
- Regulatory and compliance: the DOJ bulk-data rule, the National Security Division’s Data Security Program restricting bulk transfers of U.S. sensitive personal data to countries of concern, in force since 2025; and privacy and use restrictions across marketing, insurance, underwriting, and mortgage, HIPAA among them.
- Data delivery: formatting and scheduling, and the choice between drop-and-replace and update-and-insert, which reads like plumbing until the wrong one breaks a customer’s production system.
- Cloud hosting and implementation: platform-agnostic across Azure, AWS and the rest; I have contracted hosting providers directly and worked alongside the clients implementing on them.
- Value-added reseller and OEM channels.
- The three loaded objections every data buyer raises, completeness, currency, and accuracy, handled as the technical questions they actually are, not as brush-offs.
The data itself
- Not all data is created equal, and the gap is where the money is. Some of it is genuinely a commodity. Some only looks like one. Tax, deed, mortgage, and foreclosure records are the second kind: sourced directly from thousands of counties and municipalities, and non-fungible in the strict sense. A competitor with $100M and a running start would still spend years building the equivalent.
- Direct sourcing from the origin rather than a reseller: counties and cities, FEMA, NASA, MLSs, appraisal and inspection, insurance, customer databases, and stranger sources when the use case earns them. The value sits in the collection, which is hard, not the redistribution, which anyone can do.
- Governance and fidelity to source. County records stay county records. You do not overwrite an appraiser’s square footage with the county’s number and lose both; raw and formatted live in separate fields, and two sources are never quietly merged into one.
- Fields named for what they hold. Descriptive naming, not the numeric field codes some vendors still ship, which force every DBA downstream to keep a decoder ring.
- Checksums, record-count validation, and the rest of the production discipline that separates a data product you can trust from one that is merely large.
Deals and scale
- Six- and seven-figure enterprise deals, complex, multi-stakeholder, multi-quarter cycles.
- Three-to-seven-year auto-renewing license terms that average roughly fourteen years of recurring, predictable revenue once the relationship takes.
- Closed across the United States, Canada, the United Kingdom, Australia, and with European companies.
- Price architecture: price sheets and volume and price tiers for transactional and bulk products.
- Deal pricing set live against three forces at once, the client’s own factors, the internal exigencies of the quarter, and the competitive landscape.
- Full commercial P&L ownership by product line, with cost of sale, cost to serve, and profitability tracked by account and by vertical, not just the average.
- This is my home, not my stretch.
Contracts and legal
- Grant of license, restrictions, indemnification, representations and warranties.
- Perpetual-use and data-retention-for-compliance clauses.
- SLAs, amendments, renewals, expansions and upsells.
- The difference between production detail and contract language; drafting, redlining, and enforcement, alongside both clients and internal counsel.
Fit and use
- Consumer-facing real estate and people-search portals.
- Modeling of every kind: automated valuation models, replacement cost, actuarial and risk scoring, economic forecasts, home-sales indexing.
- Marketing models: propensity scores, lead lists.
- Match-append against portfolio data; and the hard one, merging and integrating diffuse, mismatched sources into proprietary assets that did not exist before.
- The rules that govern all of it: security, privacy, and regulation across marketing, insurance, underwriting and mortgage, HIPAA among them.
A colleague once asked me not how I sell but what a sale actually is, and I could not answer. This site is where I work that out; the thinking is in the essays and the Understory.
Author’s Note
Yes, I wrote all of this with AI. Of course I did. I am not a writer, and I do not aspire to be one. I am a salesperson and a sales leader, and I used the tool that fit the job, the same way I would use any other. The ideas are mine, drawn from twenty-five years of doing the work. The writing is only how I got them onto the page.
If you have read any of this and you want to talk, you are not looking for a writer either. My expertise is not in the prose. It is in the doing, the doing of sales: reading a room, earning trust, explaining the hard thing simply, and asking for large sums of money without flinching. So let's talk. Not about the writing. About the ideas, and what they might be worth to the way you sell.
Get In Touch
I currently lead enterprise sales at a data company, and take on select advisory and consulting engagements alongside it. I am always open to a conversation worth having. My full background is on LinkedIn. Direct contact below.
For the longer thinking behind the essays, wander into The Understory.
|
