There is one equation that governs the next three forces of your business, and most organisations have never written it down.
Customer experience equals expectation minus reality.
Marketing sets the first term. Sales confirms it. Value Delivery determines the second. The gap between them is the entire customer experience, and it is decided long before anyone completes a satisfaction survey.
This matters now because of a specific asymmetry. Artificial intelligence (AI) has made it dramatically cheaper to raise expectation. It has done far less to raise reality, and what it has done there is slower, harder and less visible. An organisation that applies AI enthusiastically to marketing while leaving delivery untouched is not improving its business. It is systematically widening the gap that produces customer disappointment, and doing so at a lower unit cost than ever before.
This is Part 2 of six. Part 1 examined Demand, the evidenced understanding of what your market actually wants. Marketing is where that understanding either gets communicated faithfully or gets improved upon.
What Marketing actually is
Marketing is a relay function. It takes the insight gathered in the Demand phase and communicates it compellingly, through the channels and media that connect best with the intended audience, while setting expectations the business can meet.
That last clause is the one organisations quietly drop. Marketing is routinely measured on attention, reach and lead volume, and almost never on the accuracy of the expectations it creates. So it optimises for the metric it is judged on, and the cost of the expectations it sets is absorbed somewhere downstream by a delivery team who were not in the room.
The definition also carries a dependency worth stating plainly. Marketing relays what Demand discovered. It does not invent it. If your Demand force is weak, and Part 1 argued that in most organisations it is, then your marketing is not communicating market insight. It is communicating internal opinion, confidently, at volume. AI will help you do that faster.
What AI genuinely changes here
Three things change, and only one of them is strategically interesting.
The first is production cost. Copy, variants, imagery, localisation, channel adaptation and creative iteration have all collapsed in cost and cycle time. This is real and it is immediately available.
The second follows from the first, and it is the one that matters. When production cost approaches zero, the cost of testing a hypothesis approaches zero with it. The constraint on marketing has historically been that you could only afford to find out whether a small number of positions, messages and framings resonated. That constraint has largely gone. The valuable output is not more content. It is faster, cheaper learning about which articulation of your value actually moves the market, fed back into Demand as evidence.
Most organisations have captured the first change and ignored the second. They produce four times the content and learn nothing new from it.
The third change is to the channel itself, and it is the one Directors are least prepared for. A growing share of buyers now form their shortlist by asking an AI assistant rather than by working through search results. The job of your published material is shifting from ranking in a list to being retrieved, understood and cited as a source. Content that is vague, promotional and unsubstantiated performs poorly at that task. Content that makes specific, verifiable, well-structured claims performs well. This is a genuine change in the economics of the Marketing force, it is happening now, and it rewards precision over persuasion for the first time in the discipline’s history.
What AI does not change
It does not decide what is true about you.
Positioning is a judgement, not a generation task. Whether your organisation is the fastest, the most specialised, the lowest cost, the safest pair of hands or the only provider who will take on the difficult work is a fact about your business, established in Value Delivery and discovered in Demand. A model can articulate that position well. It cannot determine it, and it will confidently produce a plausible one if you do not supply it.
There is also a convergence problem that deserves board attention. When every organisation in a category briefs similar models with similar inputs, the outputs converge. Tone, structure, vocabulary and argument all drift towards a common centre. The result is a market in which everyone sounds broadly the same and nobody sounds like themselves. Distinctiveness, which was already the scarcest asset in marketing, becomes scarcer.
Apply the defensibility test to this directly. A marketing capability built on production volume is replicable by any competitor within a quarter, for roughly the same cost, using the same tools. It fails Richard Rumelt’s standard completely. There is no expense, delay or complexity involved in copying it.
The bench strength question
Ask it honestly. Is your organisation successful because of market-leading marketing?
The test is not whether you have a brand you are proud of. It is whether you can evidence that your marketing outperforms. Do you know your cost per qualified opportunity by channel? Can you attribute revenue to specific positions rather than to activity? Do you know which of your claims prospective customers actually repeat back to you in a first meeting?
Organisations with genuine marketing bench strength answer these quickly. Most cannot answer them at all, which is a useful diagnosis rather than an embarrassment. Establishing which force your advantage genuinely sits in, before committing budget to any of them, is the entire purpose of an AI readiness assessment.
If marketing is not why you win, then applying AI here produces efficiency, not advantage. Efficiency is worth having. It is not a strategy.
From efficiency to strategy
A weak strategic statement in this force sounds like this: “We will use AI to scale our content production and improve engagement.” No obstacle is diagnosed, nothing is traded off, and the measure of success is an activity count.
A real one names the obstacle. For example: our buyers now shortlist before they ever contact us, increasingly through AI assistants, and we are absent from that process because nothing we publish makes a specific enough claim to be retrieved or cited. That is a significant obstacle, it is measurable, and closing it changes commercial outcomes rather than marketing metrics.
The guiding policies are where the discipline shows. We will publish nothing we cannot substantiate with delivery evidence. We will not increase volume, we will increase specificity. We will not make a claim in marketing that Value Delivery has not confirmed it can meet consistently. Each of those tells the team what not to do, which is what makes trade-off decisions fast later.
Someone senior owns it. The measure is not impressions. It is the share of qualified inbound enquiries that reference a specific published position, tracked quarterly.
Now test defensibility. This strategy holds because its inputs are proprietary. The specific claims come from your delivery data. The language comes from your customers’ own words, captured in the Demand force. A competitor can copy your phrasing in an afternoon. They cannot copy the operational evidence that makes the phrasing credible, or the delivery capability that lets you keep the promise. That is expensive, slow and complex to replicate, which is precisely the standard.
Where this leaves you
One exercise before Part 3. Take the three strongest claims currently on your website. For each, ask the person accountable for delivery whether the organisation meets that claim consistently, for every customer, not for the best ones. Where the answer is no, you have found an expectation gap that your business is already paying for in customer experience, renewals and reputation, whether or not it appears in a report.
Part 3 examines Sales, the conversion of interest to commitment. It is the force where the expectation term gets locked in contractually, where AI is being sold hardest, and where the highest-value application is almost certainly not the one your CRM vendor is demonstrating.
Unisphere works at the convergence of technology, strategy and AI, with the business knowledge to connect them. Our AI and digital transformation practice begins with a readiness assessment and a roadmap, not a tool selection. If your marketing is setting expectations your delivery cannot meet, that is a strategy problem before it is a marketing one, and we should talk.