How to Run a GEO Strategy Workshop with Clients

The short answer: the workshop is the single highest-leverage hour in a GEO engagement — not because of what you present, but because of what the client builds with you. A slide deck tells a client what you think should happen; a scored board, built live on a touchscreen with their hands on it, produces a plan they wrote. That's the gap between a plan you re-sell every quarter and one the client defends to their own boss.

This is the second playbook in our Becoming the GEO Agency series — the hub covers the whole operating model; this one goes deep on the single move that makes the rest of it sellable. If you've already won the work, the workshop turns a signed contract into a plan the client will actually run. If you haven't run one yet, read this before your next pitch — the strongest pitches now end with "let's build the first version of this together," not "we'll send you a proposal."

Build a first-pass board before your next client session — it's the whole workshop in one screen. Create a free account →

Why the workshop is the center of the practice

Every GEO engagement has a moment where the client either buys in or doesn't, and it isn't the proposal — it's the workshop. A proposal is something you defend; a workshop is something the client helps build, and people don't argue with their own conclusions.

This matters more in GEO than in most disciplines, because the tactics genuinely trade off against each other. There's no single right answer to "should we chase third-party citations or fix our schema markup first" — it depends on the client's team, timeline, and risk appetite. A slide deck can only assert a priority order; a workshop lets the client feel their way to one, with you steering, and a plan someone feels their way to is one they keep believing in after you leave the room.

The mechanism is co-authorship. Present a finished plan and the client's only options are to accept it, push back, or quietly file it away — that third option is what happens to most strategy decks. Build the plan together and the client has already made the calls: which tactics matter, how much risk they'll tolerate, what their team can carry. They can't shelve a decision they made themselves. That's the entire case for the workshop over the deck, and why it sits at the center of the practice rather than at the end of it.

Before the room: build the rough first-pass board

Never walk into a workshop with a blank board. An empty screen puts the burden of the first move on the client, and most rooms go quiet when you ask "so, where should we start?" Instead, spend an hour beforehand building a rough first-pass board from the client's public footprint — their site, existing content, a handful of prompts run against the assistants their buyers actually use. You're not trying to be right; you're giving the room something to react to, because people correct a draft far more readily than they generate one from nothing.

Use that pre-work to set the agenda around the four lanes rather than a flat list of forty tactics. Walking in with "here's roughly where you stand on PR, technical, content and brand" gives the client a mental map before you've dragged a single bubble, and it means the session can move fast instead of spending its first twenty minutes on orientation. If the client hasn't seen the lane breakdown before, send a short primer beforehand so the room isn't learning the framework and applying it at the same time.

The rough board also does quiet diagnostic work: pushback on where you've placed a tactic tells you how the client sees their own capacity and risk tolerance, and where the real conversation is headed. Treat it as a hypothesis, not a pitch — you'll be moving most of the bubbles within the first half hour anyway.

On-site or online? Run the workshop in the room

The workshop itself should happen on-site whenever it's remotely feasible — in the client's office or yours, with the relevant stakeholders physically in the room. The co-authorship effect that makes the whole method work depends on people getting up, touching the screen, and arguing across a table, and that energy is genuinely hard to reproduce over a video call. A Zoom or Teams session runs on a stricter clock, one person holds the screen share, and the quieter voices — often the executors who know the real capacity — tend to stay muted. On-site, the room self-corrects: someone leans in, someone else reaches for the stylus, and the disagreements that need to surface actually do.

There's a second reason on-site wins, and it has nothing to do with the board. A workshop is one of the few times you get the client's whole team in one place for a few hours, and the value isn't only in the agenda — it's in the coffee beforehand, the lunch in the middle, the ten minutes in the hallway after. Those informal exchanges are where you actually get to know the people you'll be working with, pick up the context that never makes it into a brief, and build the relationship that makes a retainer stick. An online meeting, by design, skips all of it: it starts on the minute, follows the schedule, and ends when the calendar says so. Treat the unstructured time around an on-site workshop as part of the work, not overhead.

Follow-ups are a different story. Once the plan exists, the recurring reviews — walking the board against progress, adjusting scores, re-sequencing — work perfectly well over a call, and insisting on on-site for every quarterly check just adds travel for little gain. The right rhythm for most engagements is a live, on-site workshop to build the plan, then online reviews to keep it moving — dropping back to on-site only for a bigger reset, like a full re-planning session, a new senior stakeholder, or a strategy shift that again benefits from everyone in one room. And if an on-site kickoff genuinely can't happen, a video workshop still beats a slide deck by a wide margin; just protect the co-authorship as hard as you can — send the board in advance, get everyone on camera, and hand out control of the screen deliberately instead of presenting.

The room setup

Room setup matters more here than for almost any other agency meeting, because the method depends on the client physically touching the plan. A laptop screen shared over a call won't do it — it turns the session back into a presentation, just a smaller one. You need a big touchscreen, ideally wall-mounted or on a stand at the front of the room, loaded and ready before anyone sits down.

On attendance, resist inviting everyone who might have an opinion. The right room is small: whoever owns the budget, whoever executes the tactics day to day, and — if it exists — whoever owns content or engineering resourcing, since those two lanes generate the most capacity debate. Three to five people is the sweet spot; beyond that, the session slows to a crawl and the quietest people never touch the screen.

Roles should be explicit before you start. You are the facilitator, not the presenter — keep the board moving, surface disagreement early, translate what the room wants into a tactic on the screen. The client's job is to drag, argue, and decide. Say it out loud at the top: "I'm going to ask you to touch this screen a lot today — the plan only counts if you build it." Naming the format up front heads off the instinct to sit back and wait for a deck.

Match the four lanes to the client's real team

The four lanes — PR, technical, content and brand — are how GEO work is organized, but they almost never line up with a client's org chart. A large brand might have a division for each; the clients you'll actually run workshops for tend to have four lanes and maybe two people. Some lanes are owned by an external partner, some by one overstretched generalist, and at least one — usually brand — by nobody in particular. Before the capacity slider means anything, the room has to work through each lane and answer three questions: who actually does this work, are they inside the company or outside it, and is the real constraint people or budget?

Those sliders are not cosmetic — they decide which tactics stay on the board. Every tactic in the library carries a difficulty level, and a lane's slider sets how far up that ladder the client can realistically climb. Set PR to 2/5 and the lane keeps its entry-level plays while the ambitious, higher-ceiling moves grey out as over-capacity; push the same slider to 4/5 and those bigger plays come back. The slider doesn't just shorten the to-do list — it sets the ceiling on what each lane can contribute to the result.

What that means lane by lane:

• PR. A low setting limits you to the quick, low-effort plays — review velocity, answering journalist requests. The compounding big bets that drive the most AI citations, like data-driven PR campaigns and earning your way into the sources models already trust, need a higher team level before they'll stay on the board. A small PR function still contributes, just slowly and at a lower ceiling.
• Tech. Most of the tech lane's value lives in low-level quick wins — freeing facts trapped in images, fixing contradictions, footer and nav corrections — so even a modest technical setting keeps the highest-ROI work in play. What a low slider drops is the heavy infrastructure (closing indexability gaps, server-side rendering), which needs real engineering and stays greyed out until the tech capability, in-house or bought-in, is there.
• Content. A low setting keeps the answer-first retrofits and definitional pages any marketer can ship; it greys out the citation-moat plays — original research, proprietary data, sustained information-gain programs — that need a real content or research function. Raise the slider and the lane's ceiling lifts from "tidy the existing pages" to "become the source everyone cites."
• Brand. The slowest lane: a low setting leaves the baseline audit and a few proof points, while the durable moat plays — named author entities, a real video program, recognised internal voices — demand sustained capacity and only stay on the board when the brand slider reflects a team that can carry them over quarters.

This is why the sliders are a strategy conversation, not a settings screen. Where the client invests in capability — hiring, reassigning someone, or budgeting an external partner to lift a lane — directly decides which tactics, and therefore which results, are on the table. A lane left low isn't punished, it's capped, and the board shows exactly what's being given up. The presets (Lean SEO team, Content-led team, Locked-down enterprise) are quick starting points for common shapes, but the real value is tuning each lane to this client. And it only works if the settings are honest: a slider set to the team you wish you had puts tactics on the plan that never ship — the fastest route to a strategy that impresses in the room and stalls the week after.

Ask them lane by lane, out loud. "Who here owns technical changes to the site?" is a more revealing question than it sounds — and the answer is often "our external dev agency," "a freelancer we use," or an awkward silence. That silence is information: a lane with no clear owner is a lane where good tactics go to die, and naming it in the room is half the fix.

When a lane lives outside the company

The technical lane is the one this happens to most. Plenty of clients don't have in-house engineers — their dev work runs through an external agency, a contract developer, or a product team that treats marketing requests as the bottom of the backlog. That changes the lane completely. Capacity is no longer "hours our developers have free"; it's "how quickly can we get our partner to ship a change, and is there budget in the relationship to commission it." A schema fix that's a two-day job for an in-house dev can be a three-week, billed-by-the-hour job through an external shop — which genuinely changes its speed score for this client and pushes it later in the sequence. When you gate the technical lane for a client like this, you're really gating their access to, and budget for, someone else's dev team. The same logic applies to an outsourced content studio or a PR agency on retainer: the tactic isn't executed by the person in the room, it's briefed to a partner, and collaboration means translating each tactic into something that partner can actually run.

Identifying the resource is usually a budget question, not a headcount one

It's tempting to treat a missing lane owner as a dead end — "we can't do brand-entity work, nobody here does brand." But most missing capacity is a budget decision in disguise. The client usually can get the work done; the question is whether they'll assign someone to it or pay a partner to do it. Framing it that way in the workshop is far more useful than writing the lane off. A lane with no internal owner isn't automatically out of the plan — it's a lane that needs either an assigned owner or a budget line, and deciding which is a call the room can make on the spot.

Sometimes the most valuable output is a structural recommendation, not a tactic

This is where a GEO workshop earns its keep beyond the tactic list. When the mapping exposes a gap — no one owns brand consistency, the technical lane is stuck behind a slow external partner, content is one person already at capacity — the right advice isn't another tactic, it's a change to how the client is set up. "Assign a single owner for brand-entity consistency, or budget for us to run it." "Pre-book a block of your dev partner's time each quarter, or the technical quick wins simply won't ship." "Content is maxed out — either free up capacity or the publishing tactics stay parked." Naming those adjustments — who needs to own what, and where budget has to go for GEO to work at all — is advice a client can act on the same week, before a single tactic starts, and it's what makes an agency feel like a strategic partner rather than a vendor with a checklist.

Running the session

1. Start from the four lanes

Open on the lane view rather than the full tactic list — PR, Technical, Content, and Brand — and ask the room, lane by lane, what's already working and what clearly isn't. It's a warm-up, not the real work, but it gets everyone talking before anyone's touched the screen and surfaces which lane the client is most anxious about, which is usually where the real budget conversation lives.

2. Drag tactics onto the speed×impact board

Move into the scored board itself — every relevant tactic plotted by speed and impact, shaded by risk. Hand the stylus, or just the moment, to the client. Have them drag the tactics they already know they want onto the board first, a low-stakes way to get hands on the screen before the harder calls arrive. Resist narrating every score as you go — let a few placements sit and see whether the room agrees before you explain the reasoning.

3. Draw the quiet people in — especially the executors

The most important voice in the room is often the quietest: the person who will actually build the tactics knows the real capacity and the real friction, but they're usually junior to the budget-holder and won't volunteer it. Ask them directly, by name — "you'd be the one shipping this; what's the honest timeline?" A board built only on the loudest opinions is a board that breaks the moment execution starts. It also quietly changes the room's dynamic: once the executor has spoken and been taken seriously, the capacity conversation that comes next stops being a negotiation and starts being an honest accounting.

4. Set the capacity slider to the client's real team level — not their aspirational one

This is the step agencies most often rush, and the one that decides whether the plan survives contact with Monday morning. Ask directly: how many hours a week does content actually have free, not counting what's already committed? What's engineering's real queue depth — or, if dev is external, how fast will your partner actually move, and is there budget to push them? Set the slider to that number, out loud, in the room, and watch what happens next.

5. Watch over-capacity tactics grey out — and let it happen

As the slider drops toward reality, tactics the room already dragged onto the board start to grey out as over-capacity. Don't rescue them. This is the most credible moment in the session, because it isn't you saying no — it's the math saying no, based on numbers the client just supplied. A findings list can't do this. A slide deck can't do this. This is the thing the board is for.

6. Read the quadrants

Once capacity has trimmed the board to what's real, walk the room through what's left, quadrant by quadrant. Quick wins — high speed, high impact — go first, almost by default. Big bets — high impact, slower speed — are where the real strategic conversation happens, since they require committing resources now for a later payoff. Fill-ins — fast, lower impact — keep momentum visible between the bigger moves. Money pits — slow, low impact — should mostly get cut on sight; if the client still wants one, ask why out loud, because there's usually a political reason underneath worth surfacing.

7. Sequence into now, next, later

Close the working portion by dragging the surviving tactics into three buckets: now (this month), next (this quarter), later (parked, revisit at the next review). This is the step that gives a scored board an actual shape, and it should happen with the client's hands on the screen, not yours.

8. Capture the "why," not just the "what"

As tactics land in their buckets, jot the reason beside the ones that were contested — why the brand play got parked, why the schema fix jumped the queue. The exported board then carries its own argument, so when the client re-opens it in three weeks, or forwards it to their boss, the logic is still attached and nobody has to reconstruct it from memory. This is also what makes the board survive a change of personnel on the client side: the rationale lives in the plan, not in the head of whoever happened to be in the room.

9. Read the three headline cards

Above the board sit three summary cards that read the whole plan at a glance — treat them as the room's sanity check before you close. Risk · brand exposure aggregates how much the plan is gambling the brand's reputation, set by the riskiest tactic actually placed (a plan pulled to "High" by self-published listicles, say) — if that's past the client's comfort line, cut or de-risk the single tactic dragging it up. Speed · time to core impact reads the plan's tempo — when the first win lands, when the real payoff arrives, and how it decays if work slips — and, most usefully, it flags an overcommitted lane ("Content, Tech capacity exceeded — dates will slip"), which is the board telling you the timeline is fiction until you either cut tactics or raise the team's capacity. Impact · needle-move shows the aggregate upside and — the part most rooms miss — how the plan is spread across the Seen → Believed → Chosen journey; a note like "nothing ladders to Chosen" means you've built awareness and trust but nothing that actually closes the buyer, a gap worth fixing before anyone signs off. Read together, the three cards turn a wall of bubbles into a one-line verdict — too risky, fast enough, moves the needle across the whole funnel? — which is exactly the summary a busy stakeholder wants.

Using the scores as a neutral arbiter

Every workshop hits a moment where two people in the room disagree — the CMO wants the flashy brand play, the person who has to build it knows it'll blow the quarter. This is normally where facilitation gets uncomfortable, because you're refereeing a disagreement between people who are paying you. The scores solve this before it becomes personal.

Instead of arbitrating yourself, point back at the board: "the speed and risk scores here say this is a six-week build with real technical risk — does that match what you're both picturing?" You're not making the call, you're making the disagreement legible, which almost always turns a standoff into a negotiation, because now there's a shared reference point instead of two people asserting confidence at each other.

Remember, too, that the scores aren't gospel — they're defaults from the planner library, and the room can adjust them with better information. A client who knows a tactic took twice as long as the default assumes should say so, and the score should move. Editing a score live does more for the tool's credibility than presenting it as fixed — it signals the board reflects their reality, not a generic template.

Closing the session

A workshop that ends without three specific things attached to it will quietly dissolve within a week, no matter how good the conversation was. Before anyone leaves: the agreed board, exported so it exists somewhere other than a screen about to go dark; an owner named for each now-bucket tactic, by name, not department; and dates — actual calendar dates for when the now-bucket work starts.

Then do the thing most agencies skip: book the first quarterly review before the meeting ends. Not "we'll follow up in a few weeks" — an actual date on an actual calendar, agreed while the plan is still fresh. This single habit is the difference between a workshop that produces a one-time deliverable and one that produces a renewing retainer, because the review is what turns "we built a plan together" into "we're the agency that keeps this plan moving." Send the exported board within 24 hours, but the review date should already be on the calendar before anyone stands up.

Common pitfalls

Presenting at the client instead of co-authoring with them. The most common failure is falling back into deck habits — narrating every tactic instead of handing over the stylus. If you're the only one who's touched the screen twenty minutes in, you're running a presentation with extra steps. Fix it by asking a direct question and going quiet until someone reaches for the screen.

Loading too many tactics onto the board at once. Forty scored tactics across four lanes is the full library, not the workshop agenda. Bring the client's most relevant fifteen to twenty — a board that's too dense turns the drag-and-drop into noise, and the room stops engaging once there's too much to hold in mind.

Letting the client set a dishonest capacity number. Every room wants to say their team has more bandwidth than it does, because admitting otherwise feels like a weakness in front of a vendor. Push gently but directly — "is that hours available, or hours available after everything already on their plate?" A plan built on fictional capacity fails in month one, and that failure lands on you.

No follow-up. A brilliant workshop with no scheduled review is a brilliant one-time event. If the board isn't exported, the owner isn't named, and the review isn't on the calendar before people leave, the energy evaporates by the time everyone's back at their desks.

All tactic scores referenced are the planner library's July 2026 defaults, refreshed quarterly. The four lane deep-dives linked above are live today; this playbook and its sibling spokes publish across the "GEO for Agencies" series.

FAQ