codiac: a go-to-market foundation built from nothing for a platform product.

i took on the company's entire marketing from scratch. the product sat at the product-market fit stage, and the positioning had to help the company find that fit.

my plan had three goals. grow qualified lead volume. position the company as a thought leader. build a repeatable acquisition framework from paid search and event-led nurture.

i worked with the company through Ascend GTM as its outsourced marketing lead.

i started with the foundation. a company at this stage cannot run a campaign until it can say who the product is for and why it wins.

the system ran to eighteen documents. a complete messaging toolkit, an ICP discovery framework, a brand vow, a product narrative framework, a proof point curator, a brand belief system, a tone of voice guide, and an archetype document.

around those sat the operating pieces. a KPI and experiment framework, a messaging-to-content mapper, a growth-gap mapper, a pillar forge, a brand guardian with anti-patterns, and an internal training and adoption playbook.

the toolkit itself covers a brand foundation, taglines, three elevator pitches, three value propositions, competitive positioning, a proof asset library, and an implementation checklist.

it also holds four persona kits. a product manager, an engineering leader, a startup founder, and an operations leader. each kit applies jobs to be done, voice-of-customer language, buying committee dynamics, and journey maps.

one document questions the others. a strategic reframing layer asks whether the category, the vertical, and the ICP are right to begin with. it includes a category misalignment detector, a contrarian persona search, and a blue ocean grid.

the method is a chain of prompts that i built and reuse across clients.

the plan called for four to five Google Search campaigns. core product, competitor comparison, pain point, branded, and feature. it added conversion tracking, dashboards, landing pages, a workshop-led nurture funnel, retargeting, and a month-six audit.

i ran the account with Optmyzr for weekly reports and alerts, and identified Semrush as the source for competitor ad data. hygiene was strict. desktop only, negative targeting, and a spam filter.

the first data moved me off the core product keywords. i shifted the angle to cloud cost, using a feature that schedules shutdown of idle environments.

i tested one explanation before accepting it. was the company being outbid. i read the auction insights. the top-of-page and outranking rates said no. so the constraint sat in keywords and offer.

that is the discipline i care about. a test that rules out the easy answer is worth more than a spend increase.

in november 2025 i audited the site, with an AI audit agent, for classic search, answer engines, and generative engines.

the headline finding was a site-wide directive that blocked search engines and AI crawlers from the content. everything else was secondary until that was removed.

the audit also found no structured data, content not shaped for machine parsing, and weak author authority signals.

my recommendations were concrete. remove the directive. add Article, FAQPage, and Person schema. add key-takeaway boxes and FAQ sections. add author bios. build comparison and alternative-to pages. show up where engineers ask questions, on Reddit, Stack Overflow, and review sites.

an automated blog tool was connected with a daily idea feed, and i trialed one more tool that integrates directly with the site.

generic AI content does not work for a technical buyer. we agreed on a founder-voice approach. a co-founder records voice notes and they become articles.

i committed to send a daily feed of content ideas for review, so the team could tune the prompts and the tone.

the client's team then built an approval pipeline in a workflow automation tool. it generates LinkedIn, X, and blog drafts from an idea and asks for approve or reject. i asked the design questions: which model writes, and whether anything auto-posts.

i also drafted a thought-leadership white paper and a supporting research brief. both are client work product, so i describe them and do not reproduce them.

in november 2025 i produced an AI-assisted growth strategy paper that cites 26 public sources.

the same month i ran a working session with the team that reset the plan toward founder-led and partner-led distribution.

i asked the team for two competitors per product so the competitor data could steer the next campaigns.

later i introduced the founder to a media production firm for video work.

i can claim a full go-to-market foundation built from nothing, across seven kinds of work. messaging, ICP and personas, paid search, a search and answer-engine audit, content strategy and tooling, competitive and category strategy, and growth advisory.

i can claim decisions with the reasoning attached. i diagnosed before spending more, tested the easy explanation, and moved effort toward what the data supported.

i cannot claim growth. i hold no pipeline, revenue, signup, or customer figure for this engagement.

i also cannot say which recommendations shipped. i have no record that the crawl directive was removed or that the schema went live.

the documents are client work product. this page describes them and does not reproduce them.