Services
Seven lines of work, all of them mine.
Most engagements use two or three of these, not all seven, and the first thirty minutes is usually spent working out which. There is no team behind this to hand the rest to. That is the trade: you get someone senior every week, and I take a small number of clients at a time.
Positioning and messaging
01 / 07
Most technical products are described by what they do. Buyers decide on what they no longer want to do manually. The gap between those two sentences is where the pipeline goes. I work backwards from what your customers have already told you: the words they use in tickets and sales calls, the thing they were doing instead, and the moment it became worth paying to stop. Then I write the version of the pitch that survives being repeated by someone who does not work here.
What you get
- Insights from your customers' data, your best-fit churned accounts, and whoever takes the sales calls today.
- A written positioning document: who it is for, what it replaces, what you are claiming, and what has to be true for the claim to hold.
- A messaging hierarchy anyone can pick up: the one-liner, the three supporting proofs, and the objection each one answers.
- The pages and decks that carry it, rewritten, so the positioning ships instead of sitting in a document.
Product launch and go-to-market
02 / 07
A launch is not just an announcement. It is a set of assets, sequences and sales answers that all have to exist before the date, and most of the work sits in the four weeks nobody plans for. I have seen the product be ready and the go-to-market not be, and the date moved anyway. Planning backwards from the date is the only method I trust for this.
What you get
- A plan sequenced backwards from the date, with what each piece depends on and who owns it.
- The asset set: landing page, release note, email sequence, and the in-product message where there is one.
- Sales answers written down before they are needed: pricing objections, competitor comparisons, and the honest 'not for you' cases.
- A read at 30 days on what actually moved, and what to cut before the next one.
Technical SEO and AI search visibility
03 / 07
Technical SEO used to mean crawlability and title tags. The same groundwork now decides whether an AI answer cites you or cites a competitor, and both come down to whether a machine can read your site and trust what it finds. I read what the crawler receives rather than what the repo contains. A single-page app that serves an empty root to a bot has a marketing problem the marketing team cannot see.
What you get
- A crawl and render audit: what a bot actually receives, where JavaScript hides content, and which of your pages are competing with each other.
- Structured data written and validated, so the entities on your site resolve instead of sitting there as text.
- A content architecture built around the questions your buyers ask, one page per question, no near-duplicates.
- Monthly reporting on where you appear, including whether AI answers are citing you, with the honest caveat that nobody controls that surface.
Product and landing pages
04 / 07
Conversion is not a design preference. It is a sequence of things a visitor has to believe before they act, and a bad signup rate usually means one of them is missing rather than that the button is the wrong colour. I can read your front end and your analytics, so the diagnosis comes from what the page does rather than from what a heatmap suggests. These are scoped as projects with milestones so that each issue is tackled and resolved.
What you get
- A conversion diagnosis of the existing page: where people leave, what they came for, and what the page is asking them to take on faith.
- Wireframe and copy before any visual design, because the argument is what converts and it is cheaper to fix in a document.
- Built and shipped in your stack, with the tracking wired in and tested rather than assumed.
- A read at regular intervals against what we agreed on before the build started.
Lifecycle and product-led email
05 / 07
Email is the channel most technical teams already have the data for and are least likely to have built. Onboarding that reacts to what someone did in the product, rather than to what day it is, is the highest-return piece of it. I build the flows and the logic underneath them, CASL-compliant from the first send rather than retrofitted once the list is big enough to worry about.
What you get
- Flows mapped to product events and not just a calendar: activation, stall, expansion, and winback.
- Built in your stack, whether that is your own system, Customer.io, HubSpot or Klaviyo, with the events instrumented and verified.
- Copy for every message, and the logic that decides who does not receive it.
- A CASL review of consent, sender identification and unsubscribe across each flow. General guidance, not legal advice.
Paid acquisition
06 / 07
LinkedIn and Google, and an honest conversation about whether either is right for you yet. Paid is mostly about not wasting money, and for a startup with a long sales cycle it can flatter itself for a full quarter before anyone notices the pipeline is not real. I run it against what happens after the click, and I am comfortable telling you to stop.
What you get
- Channel assessment first: whether your buyer is reachable on either platform at a cost your contract value supports.
- Campaign build with conversion tracking wired to the event that matters, not to the form fill.
- Creative and copy testing on a fixed rotation, with underperformers cut rather than left running quietly.
- A monthly review in the language of cost per qualified opportunity, where 'spend less' is a permitted recommendation.
Measurement and attribution
07 / 07
Most startups can tell you how many signups they got and not where they came from. That is rarely a tooling problem. It is that the tracking was added by three people at three different times to answer three different questions, and nobody has read all of it since. I go through what you have, work out what it is actually measuring, and rebuild it around the handful of questions you need answered before a board meeting.
What you get
- An audit of what is tracked today, what it claims to measure, and where those two disagree.
- A measurement plan: the events worth capturing, each defined once, with the definitions written down.
- Implementation in your analytics and tag manager, plus the specification your engineers need for anything that ships inside the product.
- A first-touch and last-touch view you can defend, with the blind spots named rather than smoothed over.
- Not sure which applies to you?
Most engagements start with two or three of these. Thirty minutes, no pitch, and you will leave knowing which ones are worth your money, including when the answer is none of them.