Welcome to a new edition of Coworkings AI!

Last week we had the opportunity to take part in the Coworking Spain Conference. Being there one more year felt really special and it was exciting to find that the interest in AI was everywhere. Across the many conversations we had with operators, one question kept coming up: How can we use AI to sell more?

This edition takes that question head-on. We'll keep it light on the technical side and focus instead on concrete use cases worth exploring.

To organize them, we're using the member lifecycle. There are four key moments where operators can benefit from deploying AI: Acquisition, Conversion, Retention, and Expansion, and mapping use cases to each stage makes it easy to find inspiration for whichever part of the journey is the priority right now.

In this edition we focus on three: Acquisition, Conversion, and Expansion, pairing each with one concrete AI use case operators can act on to sell more.

Let's dive in. 🙂

Three use cases, one per stage of the member lifecycle:

AI search is becoming a new acquisition channel, and it's still mostly unoptimized.

Consumer behavior is shifting: a growing number of people now discover businesses through AI assistants rather than traditional search, and the coworking brands that start optimizing for these new channels will gain an immense advantage over those not yet paying attention.

Gartner predicted traditional search volume would drop 25% by 2026 as AI chatbots absorb queries that once went through search engines, a shift that's now underway. ChatGPT alone surpassed 900 million weekly users in early 2026, and many of those users get their answers without ever clicking through to a website.

More and more prospects no longer open Google to find a coworking space. They ask ChatGPT, Claude, or Perplexity: 

"What's the best coworking in Barcelona with meeting rooms and a flexible plan?",  and get a direct answer naming two or three spaces.

The question for an operator is simple: when someone asks that, does the space show up?

This is a different game from traditional SEO. A classic search engine returns ten results and lets the user choose; an AI search engine synthesizes a single answer or a short list of recommendations, and cites a handful of sources. Positioning for this, called GEO (Generative Engine Optimization), isn't about keywords. It's about being a source the model treats as reliable: easy to extract, clearly attributed, and trusted enough to cite.

This new acquisition channel (GEO) doesn't make SEO obsolete, the two are complementary. AI answers still tend to draw from pages that already rank well in organic search, so a strong traditional presence is the foundation, not a substitute. 

LLMs aren't killing SEO; they're expanding the acquisition and discovery landscape, and GEO is the opportunity to capture that new demand.

So where does a coworking operator start? 

Before optimizing anything, the first step is to measure. That means analyzing:

When the coworking space doesn't appear but competitors do, the next move is to reverse-engineer why: which sources do the models cite when recommending them, and which features or benefits do they highlight?

Most coworking operators can't yet answer even the basic question: which prompts, and which AI search engines, surface the space, and which don't?

A growing set of software tools now helps operators measure and improve their visibility in AI search: Profound, Peec AI, Scrunch, Ahrefs, Otterly.ai, and Semrush, among others. These platforms audit visibility across AI models, analyze which prompts surface a brand, and which surface its competitors, and identify opportunities to improve positioning in generative search.

Profound is one of the platforms built specifically for monitoring AI search, so we'll use it as our example. That said, the others offer a similar value proposition, so any of them is worth exploring.

Profound is a GEO and AI-visibility platform that shows how a brand appears in AI-generated answers across ChatGPT, Perplexity, Gemini, Claude, etc.

All of the data is organized around the assets an operator configures, their brand, their competitors, and the topics they care about. Prompting works three ways: auto-generate prompts from the brand and topic configuration, upload custom prompts tailored to the brand and its products, or pull high-volume real-user queries from Prompt Volumes.

So an operator would set their brand, add competitors, and define topics like "coworking Barcelona" or "best flexible office London," and Profound tracks who surfaces in the AI answers for those prompts. The platform tracks when and how brands are cited in AI responses, benchmarks against rivals, and evaluates how AI portrays brand positioning, including a visibility ranking by topic that flags where competitors get more visibility and citations than the space.

For a coworking operator, this means checking whether a prompt like "best coworking in Barcelona" surfaces their space, learning which sources the models trust, understanding how their site is being crawled, and closing the gaps with content optimized for AI search.

A prospect lands on your website late at night, or while the community manager is busy running the floor, with a specific question: is there a private office for four, what does a dedicated desk cost, can a meeting room be booked for Tuesday?

A static FAQ makes them dig. A contact form makes them wait. Either way, much of that interest cools before anyone follows up.

An embedded sales agent closes that gap. It answers in the moment, learns enough about the prospect to point them to the right plan, and books the next step (a tour, a sales call, a meeting room) while intent is still high. Instead of a cold contact form, the sales team receives a qualified, prioritized profile.

Nexudus has released a sales agent that any operator can deploy on their website and member portal. It isn't a single web chat widget: it's a set of connected channels (chat, email, and WhatsApp for now) managed and tracked from one place.

It also covers the full member lifecycle, because in coworking the relationship doesn't end at signup; that's where it starts. The same person might ask about a day pass, compare plans, book a tour, become a member, and keep using those channels to manage bookings, and the experience shouldn't reset at each step. So the agent serves both sides of the journey: prospects exploring locations, rooms, passes, memberships, and private offices, and members creating, modifying, or canceling bookings or asking any question.

Three things stand out in how it's built: consistency across channels, operator control over what the AI can see and sell, and conversion tracking that ties conversations to outcomes.

1. Consistency across channels

Prospects and members don't all pick the same channel (Chat, Email & WhatsApp) but every answer draws on the same Nexudus records: the same plans, products, offices, rooms, FAQ articles, and visibility settings. Operators manage the data once and deliver the same experience everywhere.

2. Visibility and control

Not every plan should be public, not every price visible, and some products should never surface for AI-assisted purchase. Operators decide what the AI can access: whether prices are shown, whether an AI-specific price applies, and what contextual notes and tone enrich an answer.

3. Conversion tracking

Knowing a conversation happened isn't enough; what matters is whether it led to a tour booked, a day pass purchased, a meeting room booking created, or a membership started. By linking conversations to those downstream actions, Nexudus shows which interactions produced real business outcomes, exactly the visibility sales, marketing, and community teams need to understand where demand comes from.

That's what turns the agent from a front-end assistant into a sales engine: it captures intent, qualifies demand, and drives conversion at every stage of the member journey.

More on the Nexudus AI Sales Agent

Dynamic pricing has long been a popular tool in hospitality. Coworking inventory, by contrast, is fixed and has traditionally carried a static price. What if operators could generate extra revenue from the inventory they already have?

A meeting room sitting empty at 3pm on a Tuesday can't be sold back later, that hour is gone. The same room can be in such high demand on Thursday morning that the space turns bookings away. Flat rates can't respond to either situation: they leave revenue on the table during peaks and space sitting empty when demand drops.

Demand moves with the hour, the day of the week, holidays, local events, and the mix of members in the building. A single price applied across all those conditions is wrong most of the time, too low when demand is high, too high when it's low. Dynamic pricing closes that gap: capturing more value from the hours people most want, and making off-peak hours attractive enough to fill.

Dynamic pricing is already available in Nexudus. It adjusts booking rates automatically based on predicted demand, so operators no longer need to reprice resources by hand.

Under the hood is a demand-forecasting machine learning model trained on each space's historical bookings. Rather than relying on a single signal, it learns from several feature groups:

Together, these give the model a granular read on when and why a given resource gets booked.

Pricing rules can be configured at the space level, applying to every resource at once, or per rate, targeting resources on a specific rate. When both are set, the rate-specific rule wins: the broad rule sets the baseline and the rate-specific one fine-tunes it.

Dynamic pricing applies in three scenarios, which can be combined:

1. Projected demand is powered by machine learning. Prediction models trained on each resource's historical data classify every time slot over the next 4 weeks as high, average, or low demand, and rates adjust accordingly. A resource priced at €50/hour with a 10% high-demand increase bills at €55/hour during peak periods, and the same logic works in reverse, with low-demand discounts steering customers toward the hours an operator most wants to fill.

2. Last-minute bookings adjust rates when customers book a resource close to the start time. The change can be gradual, with the price climbing progressively up to a set percentage over a chosen window, for example, up to 20% more if the customer books within 3 hours of the start time. Or it can be fixed, applying a flat percentage adjustment inside that window, for example, a 20% increase on any booking made within 3 hours of the start time.

3. Remaining availability adjusts booking rates based on how booked a resource already is for the current day, calculated from the unbooked time left. Take a resource available from 9 to 5, or 10 hours on any given day. Once customers have booked more than 5 of those 10 hours, the price of the remaining hours adjusts accordingly, capturing more from the slots that are left as availability becomes scarcer.

More on Nexudus dynamic pricing 

Three stages, three use cases, three tools already available to deploy. The goal is the same every edition: practical AI plays across operations and the member lifecycle. In this edition, we covered where AI can help coworking operators sell more, pairing one use case with each stage — Acquisition, Conversion, and Expansion.

Retention is still to come. Within this stage, in a future edition, we'll dive deep into Churn and Engagement, an AI feature already available in Nexudus. It helps operators identify which members are most likely to churn, which ones are most active, and act on both to improve the experience.

That's it for today. If you try any of these, hit reply and tell us how it went. See you in two weeks! 🙂