Our AI practice

When the question is AI itself.

Most of our work uses AI. Sometimes the work is about AI: where it pays back, how to govern it, how to bring a team with you. That is NowNext, our AI specialist practice, guiding clients in how to implement AI-native marketing governance, proportional to the client's reality and ambitions.

▸ Project scope · AI + marketing

One scope, AI and marketing together.

Every engagement runs the same loop. We audit the AI and marketing you already have, plan in proportion to it, and cost it into a statement of work. Sprints then run on two pipelines, and the results feed back into the Company Brain.

▸ 60 seconds · Proportional governance

You control AI, not the other way round.

Four ways in

Four offers, one team.

01

AI Pay-Back Sprint

Two weeks to a map of where AI pays back.

02

Boardroom

AI direction and scenarios for the board.

03

Govern & Enable

Policy, risk and training at every level.

04

Bespoke

Build, embed, run or train, fixed-quoted.

▸ Three scenarios

Legacy, improvised or AI-native.

01

Legacy

Manual, slow, expensive. Reliable, at yesterday's pace.

02

Improvised

Tools without a method. Shadow AI, unclear IP, no gates. This is where most organisations sit: 95% report no measurable return from generative AI, and only 5% of integrated pilots produce real value.1

03

AI-native

Foundations, guardrails and essentials in place. Fast, owned, measurable. It compounds.

The middle scenario happens by default, unless the third is chosen deliberately.

1 Source: MIT NANDA, "The GenAI Divide: State of AI in Business 2025", July 2025. Preliminary findings based on 300+ public AI initiatives, interviews with 52 organisations and 153 senior leaders surveyed. Read the report · Back to text ↑

▸ Start small

Not ready for a board conversation? Start with a short chat.

Start a conversation
▸ Three tiers

Three tiers, climbed in order.

  1. 1

    The basics

    Right-sized policy and roles, a curated licensed stack, humans verify.

  2. 2

    The guardrails

    IP and the human trail, a Brain without personal data, consent rules.

  3. 3

    The essentials

    Human gates, always-on monitoring, a misuse map, a kill-switch.

▸ Failure modes

Four ways AI adoption fails, and the floor we build.

  • Adoption fadesA weekly cadence until it becomes habit.
  • Brand safety slipsGuardrails enforced at the gate.
  • Capability decaysTerms watch, drift checks, joiner onboarding.
  • Fear blocks the startProbabilistic-thinking training first.
▸ Govern & Enable path
GovernUnderstandApplyReadyDemonstrate
nownext.ai

NowNext is our AI specialist practice. It helps boards and leadership teams decide where AI pays back, govern it, and train their people, from the same senior team.

Visit nownext.ai →