The problems I’ve solved, and what changed after.

The problem

I was the only content person on a four-person marketing team, writing for a buying committee of five roles: IT, Deans and CEOs, Admissions, Marketing and CFOs. Each one needed different evidence, and there was no way to serve all five at full depth.

What I decided, and why

I asked to be judged on pipeline contribution, because downloads would have looked good while telling leadership nothing. Every content goal traced back to a business goal in one line. Our marketing manager ran the CRM and produced the pipeline reports, while I chose the metric and decided how we’d report it.

I built each persona from sales conversations and customer language, then wrote a priority split every quarter based on live deals and the gaps costing us opportunities that week (plus any persona we’d ignored long enough to stall a sale later). One quarter ran 40% IT, 25% Admissions, 20% Leadership, 10% Marketing and 5% CFO.

The biggest single asset was a 50+ page academic workflow playbook, which I broke down into other formats (ad copy for the agency’s campaigns included) and which became one of the strongest campaigns across channels.

The problem

Six months in, leadership moved the primary buyer from Admissions Directors to IT leaders: CIOs, CTOs and heads of infrastructure. Deals were stalling because IT joined late and wasn’t bought in, and the library had nothing credible for them on integrations or security, let alone architecture.

What I decided, and why

I triaged the whole library in about a day and a half into keep, deprioritise, rework and build from scratch. I moved Admissions content down to the opportunistic tier and rewrote the product pages that led with it, which freed around half of capacity for IT while the other four personas stayed in maintenance mode.

I warned leadership up front that Admissions blog traffic would dip and swapped in transition KPIs to match. I also told them to expect the first measurable results around week six, so nobody would panic at week three.

The problem

I took over content and found 40+ brand assets, each describing a slightly different company. Nobody had written a positioning document or a messaging hierarchy, let alone a tonal framework, and the brief asked me to clean up a brand that had never been defined.

What I decided, and why

I treated it as a rebuild, because a refresh assumes there’s a coherent brand underneath to polish. I audited every asset on audience, tone, what it promised, look and status, then wrote a short positioning document: who the company is, who it’s for (specific enough to exclude someone), what it stands against and how it sounds, with “we say this, we never say that” examples.

I pitched it to leadership on time and money with the audit as evidence, and brought Sales in early, since they used the deck and one-pagers every day. The company later went bankrupt, for reasons far bigger than its brand.

The problem

The championship raced across more than five countries and had a fundraising initiative with a conservative target, because nobody there had run a comparable campaign before. I was the sole marketing lead, coordinating four freelance social media specialists in the UK, the USA, Italy and Brazil on race days.

What I decided, and why

Fundraising has days instead of months, and has to move someone from feeling something to giving in the same scroll. So I built each runner’s story before race day and centred the campaign on a named runner instead of the organisation. Most of the effort went into live coverage, when urgency and emotion peaked together, with the race clock as the deadline and the running community doing the distribution.

The problem

Most readers were engineers with no patience for fluff (and I’m not one), while much of the client work was confidential or regulated, so I couldn’t write about it directly.

What I decided, and why

My credibility had to come from process, so I got to roughly 70% understanding on my own before every expert meeting, then asked narrow questions and stated what I believed so the engineer could correct it, because their time was the scarcest resource I had.

I traced every technical claim to documentation, a standard, a study or a named expert and logged it, and cut anything I couldn’t verify. Each of the six verticals got its own register, from source-heavy precision in healthcare and finance to measured capability claims in defence and aerospace. That source log is now the verification step in my AEO/LLMO framework.

The problem

Cleanwatts had a full marketing team, and I was its only digital marketer for 18 months, covering blog, case studies, PR, web, social, design and email for business buyers, energy community managers and residential consumers. Social and email didn’t exist when I arrived.

What I decided, and why

Sales came first, because the gap between how a company describes itself and how buyers describe their problem is where the content plan comes from. With them, I split the audience by buying motivation instead of CRM fields, and that split drove everything from the blog to the email programme.

Each email carried one idea and one call to action matched to where the reader was, and specific, benefit-led subject lines won every test. I rewrote the product messaging and user flow across 20+ pages and built the changes myself in HTML and CSS instead of waiting on a developer.

What I decided, and why

I ran a fixed mix at a cadence I could keep for a year, 39+ posts a quarter. Product content stayed at 10% on purpose, because pages that give it 40% to 60% watch their engagement collapse. Every blog post went out as a native post with the link in the first comment, and I wrote three or four suggested angles per piece so colleagues could share it in their own words.

What I cut along the way

Video underperformed in every format I tried, so I dropped it within a quarter. Polls got impressions and no conversation, and generic listicles lost to specific case studies with numbers by four or five times.