Kev Tanna

A creative generalist with an analytical decision-making process.

I build brands and the content systems behind them. I diagnose before I make, check whether it worked, and say plainly what isn't proven yet.

  • 61 episodes of Founder Stories filmed
  • 190 pieces of content from one event
  • 148K impressions from the first 18 episodes
  • 28 sales calls from the first 18 episodes
01

Strategy and positioning

Navrangi Navratri by Red Lotus Events: a packed hall dancing garba beneath the festival's branded stage screen.
Navrangi Navratri

Red Lotus Events

Brand architecture for a cultural events business.

5 formats
Problem:
Red Lotus ran many successful events, but they weren't connected, and the brand was carried by the founders' personal profiles rather than a proposition of its own.
What I did:
Developed a brand and content strategy to bring the separate touch points together, covering brand architecture, vision and mission, and IP frameworks for five formats: Navrangi Navratri, Sip & Paint, Meditate & Create, Migration Map and Rangeelu Gujarat. Delivered as strategy documents.
Result:
Delivered. Early, with no measured outcome to report. Status: Delivered, Early
  • Page from the Red Lotus Events vision and mission strategy: the vision statement and what it commits the organisation to.
  • Page from the Red Lotus Events brand guidelines: the IP identity framework that every event format follows.
  • Page from the Red Lotus Events brand guidelines: the Rangeelu Gujarat format's logo, colour palette and typefaces.
  • Page from the Red Lotus Events brand guidelines: the parent logo in colour and on a dark ground.
Pages from the strategy and brand guidelines

Patient Growth

A North London GP practice, and the product that came out of it.

26,120
Problem:
A North London GP practice came in asking for new photography. The real issue was visibility.
What I did:
I reframed the brief and built a digital marketing effort within NHS guidelines. Later I turned the approach into Patient Growth: a free monthly NHS list-changes page, a monthly report and a done-for-you programme. Every figure links to its NHS England release.
Result:
The practice's list grew by 8,488 patients to 26,120 since April 2024. That growth also reflects a relocation, so I don't claim all of it. The free monthly page started this month and has no results yet. Status: Measured (practice), Early (page)
The Patient Growth dashboard for a client practice: 26,120 registered patients, up 8,488 since April 2024, charted against neighbouring practices using NHS England data.
Client dashboard, live data

The Open Journal: search and AI visibility diagnosis

Brand, site and search strategy for a personal documentary studio.

Problem:
A personal documentary studio in a category few people search for.
What I did:
I built the brand identity, the site and the search strategy, then analysed SEO and AI-search visibility with DataForSEO. Search volume was low, and AI answers were citing older American companies whose films looked like 1990s television. So I changed the plan: localised for the UK, and began bringing clients into community on Reddit and Quora.
Result:
No measured result yet. Status: In progress
  • Video: Tejal's Baa, married at 16

    Tejal's Baa, married at 16

  • Video: Mahesh Gandhi, left home at seven

    Mahesh Gandhi, left home at seven

  • Video: Bishnu Das, building a life in England

    Bishnu Das, building a life in England

Films from the studio
02

Content engines

Founder Stories and Podcast Foundry

Co-founded with Priyesh Pankhania. Podcast Foundry is the studio; Founder Stories is its front door.

  • 148K impressions from the first 18 episodes
  • 8% engagement rate, no paid advertising
  • 28 sales calls
  • 25% lift in impressions and engagement after the rebrand
Problem:
A new studio with no audience and no pipeline needed a front door.
What I did:
I built Founder Stories from nothing: format, guest strategy, production and distribution. One recorded conversation becomes eight to ten pieces of content. Halfway through, the data showed the thumbnails and posts weren't standing out, so I rebranded the whole project.
Result:
61 episodes filmed, 50 of them at IdeaFest and GBEA, and 190 pieces of content from IdeaFest. The first 18 episodes reached 148K impressions at an 8 percent engagement rate, with no paid advertising, and led to 28 sales calls. The rebrand lifted impressions and engagement by 25 percent across our social output. Short-form clips drove it, and human-centred cuts beat brand-led ones. Organic shareability stayed low, and I'd build for it from the start now. Podcast Foundry has six recurring clients and was profitable from its second quarter. Status: Measured
Founder Stories, full series

Conservation Chronicles for Lancashire Wildlife Trust

Nine-episode podcast series.

Problem:
The public loves the land the trust manages but rarely sees the work behind it.
What I did:
Nine episodes built on human stories, thought leadership from the team and expert voices. I also built an asset bank their PR network could use in local and national press and on their own channels.
Result:
The trust's marketing and PR team liked it and used the asset bank. It led to a conversation about a national podcast, which is early and not a commission. Status: Early
Two Lancashire Wildlife Trust staff recording an episode of Conservation Chronicles at a table among glasshouse plants.
Recording Conservation Chronicles
  • Female Founder Stories

    I built the brand, site and content system for Siya Kantaria's founder interview series. 14 episodes published and two scheduled, with guests including Dame Alison Rose of NatWest Group, Sarah Dean of Noble Foods and Happy Eggs, and Emma Woods, formerly of Wagamama.

    Status: Early
  • Free tools as lead magnets

    A four-minute brand gap audit, a nine-question content strategy scorecard, and a free monthly NHS list-changes page. Each helps first and sells second.

    Status: Early
  • My own LinkedIn experiment

    I built a Claude Code automation that turns meeting recordings into posts, then tested it against my own writing. My own writing won. The posts that aired a real disagreement with my co-founder drew roughly ten times my usual engagement, because they voiced a question other founders were already asking.

    Status: Measured (on engagement only)
03

Production craft

On set for the MSC Cruises virtual dance class series: Flavia Cacace with the camera crew, jib and monitor rig.
Virtual dance class series

MSC Cruises virtual dance class series

Executive producer, four-part series with Flavia Cacace.

six-figure
Problem:
MSC's Head of Entertainment needed a four-part virtual dance class series with ballroom star Flavia Cacace, produced end to end.
What I did:
I ran the entire production: a three-person core crew plus a creative director and photographer, all reporting through me, with talent contracts negotiated through my company.
Result:
A six-figure production, negotiated and delivered in full. Status: Delivered
  • Cardboard Creative design work

    Brochures, event stands and packaging, mostly for FMCG clients.

    Status: Delivered
04

AI and systems

  • The Claude production pipeline

    It turns one recorded conversation into the weekly content set of clips, captions, shownotes, blog and social, with human editorial kept at the judgement points. Roughly 80 percent less repurposing time.

    Status: Measured
  • The Council

    A multi-agent system I built to test business ideas before I commit time or money to them. Every idea goes in as a structured brief. Research agents gather the evidence and check the facts, a judge weighs it, and a secretary keeps the record, so the verdict stays impartial. I've run it 14 times on brand developments across the businesses I've already built.

    Status: Active, Measured
  • SEO, GEO and AEO research skills

    Claude skills that run deep search research using DataForSEO, Apify and Firecrawl: keyword demand, competitor rankings, and which sources AI assistants cite. Every source is found live and fact-checked across all three tools. I used them for The Open Journal diagnosis and to set the page titles on this site.

    Status: Active, Measured

AI motion consistency R&D

Animated narrative short.

Problem:
The hard problem in AI motion work isn't a good frame. It's a consistent sequence.
What I did:
An animated narrative short that holds one locked visual style and three recognisable characters across separate image and video generations. Style-key locking, character reference elements, and concrete visual language over named styles.
Result:
The finished film, Our Home, is live. Status: Delivered
Video: Our Home, the finished film

Our Home, the finished film

  • Storyboard frame: all three characters in the kitchen, daylight.

    Kitchen, daylight

  • Storyboard frame: the same three characters on the sofa, lamplight.

    Sofa, lamplight

  • Storyboard frame: the mother alone in a third room, morning light.

    Third room, morning

  • Character reference turnaround used to lock the mother's design across generations.

    Character reference

The film, and storyboard frames
05

How I work

  • 01 Diagnose before making.
  • 02 Check whether it worked.
  • 03 Say plainly what isn't proven yet.

Thought leadership built for SEO, AEO and GEO has delivered the best return across the brands I've built. Third-party citations decide who AI quotes, so community and earned mentions matter as much as the site.

06

Developing makers

Three people who started with me on the wedding film business with limited experience now run their own businesses. A digital marketing specialist I've worked with for five years now runs her own agency on the workflows and SOPs we built together.

07

Earlier work

Aperture X Media

2018 to 2025

A wedding film studio for the Asian wedding market, run on a freelancer network. 120+ films delivered.

  • Video: Sonal and Sanjay

    Sonal and Sanjay

  • Video: Kirran and Prab

    Kirran and Prab

  • Video: Trisha and Niraj

    Trisha and Niraj

  • Video: Chhaya and Arjun

    Chhaya and Arjun

TEDGroup

2013 to 2018

Five and a half years running in-house production for a live entertainment group. Took the studio from a £50k cost centre to a £140k commercial operation, won the passenger photography concession on TUI's Marella Explorer, and recruited ten international staff.

Ragatip

A music brand for South Asian music, built from inception. I created the identity and produced its content and first recording project.

  • Video: Krishna Dance, Kathak choreography

    Krishna Dance, Kathak choreography

  • Video: Sitar vs Guitar

    Sitar vs Guitar

  • Video: Keshava Madhava x Made to Worship

    Keshava Madhava x Made to Worship

If any of this is useful to your team, I'd like to hear about it.

Full CV available on request.