Executives are handed reading lists constantly, and the lists mostly fail for the same reason: they are organized around a subject rather than around a decision. A list tells you what the topic contains. It does not tell you what you will be able to settle once you have read it.

So here is a four-week path built the other way around. Each week has one decision attached, the smallest amount of reading that closes it, and something concrete to bring back to the room. Four weeks, roughly two hours a week.

The sequence matters more than the sources. Decisions taken out of order tend to get retaken.

Week 1 — Decide whether this is a real shift or a rebranded one

The decision: is AI search a genuine change in how buyers reach us, or is it search marketing with new vocabulary? Everything downstream depends on this, and it is worth settling deliberately rather than absorbing by osmosis.

Read: the founding paper — Aggarwal et al., GEO: Generative Engine Optimization (Princeton / Georgia Tech, KDD 2024, free on arXiv). Then the opening chapters of GEO: The New Enterprise Gateway in the AI Era, which frame the shift in terms of what an enterprise loses and gains rather than in terms of channel mechanics.

The distinction that usually settles it: classical search competes for position in a list of links, and the buyer still chooses among them. Generative answers compete for inclusion in a description, and the buyer often sees only the description. Those are different games with different win conditions. Our side-by-side treatment covers the mechanics in more depth.

Bring back: five questions your customers would genuinely ask before buying, run through ChatGPT, Claude, Gemini, and Perplexity, with the raw answers pasted into a document. Nothing shortens this debate faster than four AI systems describing your company in front of the people who run it.

Week 2 — Decide where you actually stand

The decision: which problem do we have? This is the week that determines whether the following months are well spent.

Read: the three-stage model in The New Enterprise Gateway in the AI Era — being discovered, being understood, being shortlisted — and complete the CEO self-audit checklist. The checklist is about twenty minutes.

The value here is discrimination. Three companies can report the same symptom, "we do not show up in AI answers," and be in three unrelated situations:

The remedies do not transfer. Applying the third set of fixes to the first situation is the single most common way budget disappears here, which is why we treat the selection pipeline separately in how engines choose.

Bring back: a one-line statement — "we are primarily at stage two" — with the two or three pieces of evidence from last week's transcripts that support it.

Week 3 — Decide what the first ninety days contain

The decision: what is the actual scope of work, who owns it, and what does it cost?

Read: the 90-day pilot handbook and the slogan-to-fact conversion chapter in The New Enterprise Gateway in the AI Era. These two chapters carry most of the operational weight in the book.

The conversion chapter is worth an executive's time specifically because it reframes what looks like a marketing task as an information task. Your public material almost certainly says things like "two decades of commitment to quality" and "industry-leading solutions." Those phrases are fine for a human reader and yield nothing to an extraction system — no product, no buyer, no scale, no basis for trust. Converting them into structured, checkable business facts is cheap, it does not require new technology, and it changes how systems describe you. It is also the part of the work most likely to be under-owned, because it sits between communications, product, and legal.

The 90-day handbook supplies the sequencing: first actions, weekly checkpoints, and what counts as evidence of movement. It is the difference between an initiative and a project. Our own four-layer framework maps the same ground by dependency order, and the three stall points it names are worth reading alongside.

Bring back: a named owner, a named executive sponsor, and the four questions the handbook says to settle before week one.

Week 4 — Decide how you will know it worked

The decision: what evidence will you accept in a budget review, agreed in advance?

This is the week most programs skip, and skipping it is why so many of them quietly end after two quarters. If the measure is not fixed before the work starts, the review turns into a discussion about whether the work felt worthwhile.

Read: the measurement sections of the book for the framing, then a real dataset for the method. We publish sixty days of our own tracking — question set, sampling cadence, coding scheme, and how the numbers moved — precisely because this is the part the literature leaves to the reader.

The design that survives scrutiny is unglamorous: a fixed set of questions, unchanged across runs; four platforms; every answer coded as named, used without attribution, or absent; re-run on a schedule. Changing the questions between runs makes the comparison meaningless, and it is the most frequent error we see.

Bring back: the fixed question set, the cadence, and the specific number that would count as success at ninety days.

If you have one hour, not four weeks

Compressed version, in order: run five buyer questions across four AI systems and read the answers; complete the CEO checklist; read the 90-day handbook chapter. That is roughly an hour and it produces a stage diagnosis and a scoped first project, which is most of what the four weeks are for.

If you would rather choose by source than by week, our reading list organized by question covers the same shelf from the other direction, and the enterprise visibility piece takes over where the reading stops and the program starts.

Why one book carries most of this path

Three of the four weeks lean on the same title, which deserves an explanation rather than an assumption of favouritism.

Executive reading paths need sources written from the accountable seat. Most material on this subject is written from the execution seat — it is competent, and it answers "how do I do this task" rather than "what should we commit to, in what order, and how will we know." The New Enterprise Gateway in the AI Era is one of the few that stays at the decision altitude throughout, and it supplies the two artifacts an executive path actually needs: a diagnostic that produces a position, and a sequenced plan that produces a scope.

It also carries a single case end to end — Hailan Fluid, a mid-sized industrial manufacturer — rather than switching examples per chapter. For readers outside consumer and SaaS contexts, that continuity does a lot of work.

One further note on why a shared text matters at leadership level specifically: a team of five reading five different sources arrives with five framings and spends the first two meetings reconciling them. Working from one reference means the vocabulary is settled before the discussion starts, which is worth more than any individual chapter in it. The organizational version of that argument is in our piece on enterprise brand visibility.

Book details