For a retreat centre in Altai, the winter months were almost lost commercially. The property had strong nature, silence and good infrastructure, but retreat organisers were looking at Bali, India or Thailand for winter programmes — not Siberia.

The problem was not the property itself. It was market perception. My task was to change how organisers saw winter Altai and create demand for a season they were not actively considering.

Instead of simply promoting the existing offer, we effectively turned winter into a separate product. We repositioned the season as a distinctive setting for retreats, yoga groups and inner work, created content that addressed the main objections, and built simple tools to help group organisers understand the offer and make enquiries.

As a result, winter became a standalone business direction rather than a dead season. The centre gained a clearer system for attracting groups in the winter months and created demand for a product the market had previously overlooked.

Read the case, in full

A 5-star boutique hotel had a distinctive concept and strong offline guest experience, but its digital performance did not fully reflect the value of the property.

The hotel was attractive in reality, but online the booking journey was not converting enough existing demand into direct revenue. Paid search, organic search, website messaging and the booking path needed to work together more clearly.

The work focused on improving the connection between demand and booking: paid search efficiency, website clarity, reasons to book direct, and the steps that helped potential guests move from interest to reservation.
The goal was not to buy more traffic. It was to make existing demand convert better.

Monetary figures below have been converted into euros from the client’s local currency and rounded for the reader’s convenience.

From January–June 2025 to January–June 2026, direct booking revenue from paid search increased from approximately €18,800 to €70,200, while ad spend slightly decreased from approximately €4,200 to €4,000.

Cost per booking fell from approximately €71 to €17, and the ad cost / revenue ratio improved from 22.3% to 5.7%.

Organic search also became more efficient. Search traffic decreased from 10,051 to 7,577 visits, but buyers increased from 78 to 103, and revenue grew from approximately €33,600 to €41,100.

The main improvement was not more traffic. It was better conversion of existing demand into direct bookings.

An independent mountain hotel had a loyal guest base and regular enquiries, but growth had stalled. The hotel wanted to move closer to 85–90% occupancy, yet adding more advertising was not the right first step.

The diagnostic showed that the main constraint was not traffic or campaign performance. The hotel was already generating demand, but too much of it was being lost after the enquiry.

The key finding was simple: 97% of online enquiries were not answered within 24 hours. For potential guests comparing several properties, this delay meant lost bookings.

Instead of increasing the ad budget, we focused on the enquiry-to-booking process. I helped the team review how requests were received, prioritised and followed up, then improved the response process using their existing tools.

Occupancy increased from 72% to 85% in 4 months without additional advertising spend.
The hotel also began converting demand that had previously been lost because of slow communication.

How occupancy went from 72% to 85% without extra ad spend
The mountain hotel case, in full

A 5-star boutique hotel with a strong concept was found almost only by its own name. Guests who did not know the hotel yet never came across it.

At the same time, the way people search had changed. Instead of typing a query and comparing ten links, they were asking an assistant a question and getting two or three names in return. The hotel was not one of them.

The reason was in the content. AI systems look for text they can quote in answer to a specific question. General descriptions about comfort and a good location are of no use to them. What works is detail: what makes this hotel different, what the concept means in practice, who will feel at home here and who will not.

The work started with the guests themselves. The full body of reviews was analysed to find the questions, expectations and doubts people actually have. The website was then rebuilt around them: structure, copy, a detailed FAQ, and articles answering specific guest questions. No mass produced content and no technical tricks.

Traffic figures below are taken from the hotel's own analytics for the period after the work.

Visits from ChatGPT ran about four times higher than visits from a paid placement in a well known city lifestyle magazine with a highly relevant audience.
The behaviour of that traffic was better than the site average: bounce rate 15.5%, more than 2.5 pages per visit, average visit 3.5 minutes.

Paid bookings came from both ChatGPT and Perplexity, converting at around 1.35% of visits, from a channel with no media cost.

A small countryside hospitality property was investing in marketing, but the guest journey was fragmented. The problem was not one weak channel. The website, booking flow, analytics, advertising and content were not working together as one clear path to booking.

My work was to audit the full journey from first contact to attempted reservation and identify where potential guests were likely to drop off.

The diagnostic showed several important constraints.
Room information was spread across different parts of the website, with inconsistent details and unclear next steps. A guest could show interest in a specific room, click to book, and be taken back to a general search form instead of a clear booking path.

The booking form also created friction. Some elements did not work correctly, and the mobile experience made it harder for guests to complete key actions. This was especially important because a large share of traffic came from mobile devices.

Analytics were not set up to show which channels were producing real bookings. This meant the property could see traffic, but could not clearly understand which marketing activities were creating revenue and which were only creating visits.

The advertising strategy was also too direct. It expected guests to move from a first ad contact straight to booking, even though choosing a countryside stay usually requires more time, comparison and reassurance.

The recommendations focused on building a clearer, connected booking journey: unified room pages, a cleaner booking path, corrected mobile issues, proper conversion tracking, more human and trust-building content, and a more staged advertising approach.

The main conclusion was simple: the property did not only need more traffic. It needed to stop losing interested guests between interest and booking.

The hotel case, in full

An independent countryside hotel had several strong assets: a distinctive location, a farm, horse riding, local food, a quiet atmosphere and enough visual material to support a more premium positioning.
But the website was presenting the property mainly as a place to book rooms. This made the hotel easier to compare by price with other regional properties, instead of helping guests understand why they should choose this specific place.

The audit showed that the main issue was not one weak page. It was the way the full guest journey was structured.

The first screen did not make the location clear enough. Some important selling points, such as the farm, food, horses and quiet atmosphere, were present but not organised around clear guest motivations. Room pages did not help visitors choose confidently.

Service pages described what existed, but did not make the experiences easy to understand or book. The booking path also needed clearer calls to action and stronger reasons to book direct.

My recommendation was to rebuild the website logic around three guest scenarios:

Quiet recovery — for guests looking for silence, slow time, farm food and a simple reset.
Base for exploring the region — for travellers who need a comfortable starting point for nearby routes and landscapes.
Farm and horse experience — for guests attracted by animals, local food, horseback riding and a more authentic stay.

This changed the role of the website. Instead of forcing every visitor through the same generic hotel structure, the site would help each person recognise their own reason to stay, understand the experience, and move towards a direct enquiry or booking.

The roadmap included changes to the homepage, navigation, room cards, service pages, messaging, local map content, booking triggers and messenger-based contact points. The goal was to connect the hotel’s real strengths into one clearer commercial system.

The main conclusion was simple: the hotel did not need to look like a bigger hotel. It needed to sell the specific reasons why the right guests would choose it directly.

The company has been in the market for many years and is well known in its region. Its main business is selecting and selling stays at health resorts, mostly in Siberia.

The reputation was strong, and the advertising still did very little for sales. Enquiries were few, and in some campaigns the cost of a single enquiry was very high. That was the cost of an enquiry, not of a paying client.

I started with an audit and saw where the budget was going. The advertising was bringing in a lot of accidental visitors. Part of it was being shown to people who were already searching for one particular resort by name, so the company was paying to appear in front of travellers whose minds were made up.

And the landing pages did not explain why booking through this company was better than going to the resort directly.

I then built three separate landing pages from scratch, one for each type of health resort programme, and rebuilt the Yandex Direct advertising completely. Each one had its own offer, its own copy and its own way of opening the conversation.

The result was a large and steady flow of relevant enquiries. The cost of an enquiry came down by 85 to 91 per cent depending on the programme, and the pages converted visitors into enquiries at up to 5%.

Once the advertising was producing enquiries reliably, the next stage was SEO. The aim there is to lower the dependence on the advertising budget over time, and to take more enquiries from people who are already looking for a resort, a treatment or somewhere suitable to recover.

A company selling someone else's rooms has to give the traveller a reason to come through it rather than book direct. That is the same question a hotel faces with the booking platforms, seen from the other side.

The clinic was spending €5,000 to €6,000 a month on social media advertising and had gone through more than €20,000 over several months. The reports looked reasonable. Thousands of clicks, a good click-through rate, well-made creative. Sales stayed exactly where they were.

The first comparison of the ad account against the website analytics explained it. Of almost 5,000 clicks, 1,256 people actually reached the site. A gap of 75%, and nobody had noticed it.

I started with an audit and found six systematic errors in the campaigns. Then I rebuilt the advertising from the ground up.

• Changed the goal of the campaigns from a click to an enquiry, and for their online shop, to the purchase itself
• Built an enquiry form with extra steps, so that people who were not serious filtered themselves out, and sent the traffic to it
• Split one broad audience into seven kinds of segment: age, location, interests, behaviour on the site, response to the advertising, existing clients, look-alike
• Switched off Audience Network, which was producing cheap clicks from people who were never going to buy
• Set up proper conversion tracking, so that decisions were made on enquiries and sales rather than on clicks
• Put the creative on a regular update cycle, instead of approvals that took a week each time

These treatments are an expensive purchase with a long decision behind it. One to two months pass between the first enquiry and the appointment. That is why the campaigns have to be optimised on the enquiry rather than the click, and why the people who are still thinking have to be collected somewhere while they think.

Results in the first month

• €40,000 in revenue, the highest figure in the clinic's history, in June, which is its quietest month
• €20 to €25 cost per enquiry, down from €35 to €40
• 8.6% website conversion, up from 3.5%
• 150+ enquiries with a name, a phone number and an age
• 250+ people on the list of those who had enquired and not yet bought, who were then brought back through retargeting and email with no additional budget.

The budget stayed the same, €5,000 to €6,000 a month. For the first time the advertising paid for itself and produced a profit.

A hotel selling a wellness programme, or a travel company selling a retreat, is in the same position. The purchase is expensive, the decision takes weeks, and most of the revenue comes from people who were not ready on the day they first saw the offer.

The company organises premium safari trips starting at $6,000 per person.

The problem was not a shortage of clients. It was that the owner was personally carrying a large part of the marketing. Winning clients took too much of his time and energy, and a good deal of that activity produced nothing visible.

My job was to work out which tools genuinely affected sales and which ones only created load.

After going through the marketing, the competitors and the way clients actually behaved, we stopped the activities that were not earning their place and concentrated on the channels that were already producing trust and buyers: the founder's own name and reputation, content, email, and the existing client base. On top of that we put in a proper way of coming back to past guests and asking them for referrals.

Marketing became simpler and easier to manage, the cost of promotion came down, and the owner got time back for the product and for the clients themselves.

At this price, a trip is not bought from an advertisement. It is bought from someone the traveller has come to trust over several months.
That makes a lot of ordinary marketing activity beside the point, and makes a small number of channels worth much more attention than they usually get.

A network with a long history. Founded in 1989, with clinics in Vienna and in several Russian cities.

Its social media was running empty. The audience barely reacted, there were no fresh ideas for the content, and the promotion had no system behind it.
There was also a difficulty particular to this field. Eastern medicine has to be explained. People choose it once they believe in the approach, not once they have seen the price.

I built the promotion as a system of connected parts.

• Set the positioning: Tibetan medicine as a whole approach, the lead doctor's own experience and opinion instead of abstract posts, something practically useful in every text
• Introduced one rule: every post promotes a specific service rather than existing on its own
• Built a content strategy around six recurring formats: practical posts, answers from the doctors, the founder's own story, multi-day online programmes, polls and video
• Ran the promotion across three social platforms, each one for a different segment of the audience
• Set the advertising on enquiries, and sent the traffic to the landing pages for services, seasonal offers and the online shop

The posts stopped being a shop window and became part of the funnel. Instead of likes there were shares and visits to the website, and the monthly offers had a steady flow of advertising behind them rather than occasional bursts.

Results in one month

• 50,000+ people reached in the target audience
• 136 appointment enquiries
• €4.60 cost per enquiry on Instagram
• €4.80 cost per enquiry on Facebook
• €0.18 per click on the advertising for the seasonal offers

The cost per acquired patient is covered by an NDA.

Why this is on a hospitality page
Tibetan medicine, a longevity programme and a week-long wellness stay are bought in the same way. The guest has to believe in the approach before the price means anything at all. That is why content that explains sells more here than any offer does, and why the usual promotional calendar produces so little.

The company designs and runs bespoke journeys in Argentina, Peru and Bolivia. For a long time its clients came mostly through recommendations and social media.

The problem was that the flow of enquiries could not be predicted. The company had no way of knowing how much it would sell in the coming months, and depended heavily on word of mouth and on demand that simply happened to arrive.

My job was to build an acquisition channel the company could actually manage.
We launched paid campaigns in search and in social media, reworked the landing pages, and put in place a regular way of producing enquiries.

The company now has a steady flow of relevant enquiries, and the number of trips it runs has gone from four to six a year to more than ten. The business depends far less on referrals and on seasonal swings in demand.

Why this is on a hospitality page
Word of mouth is the best source of business there is and the worst one to plan around. A hotel is in the same position with its returning guests and the platforms. The bookings do arrive, but not on a schedule anyone can build a year on.

A clinic in Cyprus was developing its medical tourism offering for patients from Israel. Two services were being promoted that were relatively difficult to market: hyperbaric oxygen therapy (HBOT) programmes and comprehensive recovery and rehabilitation programmes.

The main challenge was that, for most potential patients, these were not familiar services with already established demand.

The task was not only to attract attention, but also to explain what HBOT is, what the programme is used for, how the treatment works, why it could be worth travelling to Cyprus for it, and how the clinic’s offer differed from the options available in Israel.

The advertising had to address several things at once: the medical value of the programme, trust in the doctor, clarity of the treatment, and the additional barrier of travelling to another country.

The second campaign focused on recovery and rehabilitation for patients before and after surgery, people with chronic conditions, athletes, and relatives looking for suitable treatment options for family members.

Here too, the task went beyond simply advertising a procedure. We had to connect unfamiliar treatment methods with a specific life situation and a clear patient need.

I managed the full performance marketing scope of the project:

• developed audience segmentation and the advertising strategy;
• set up and managed paid social campaigns;
• wrote ad copy for different needs and stages of the decision process;
• developed concepts for static creatives, carousels and doctor-led videos;
• adapted the messaging to the medical tourism context and the need to educate patients before asking them to convert;
• created lead-handling scripts and follow-up scenarios to help move enquiries towards consultations more effectively.

Results
Across the two service lines, the campaigns generated more than 150 enquiries from potential patients in Israel.
Average cost per lead was below €20.

At the next stage, I also audited how incoming enquiries were being handled. The review showed that a significant part of the potential value was being lost after the advertising stage: responses were sometimes delayed, communication was overloaded with information, and potential patients were often given too many possible next steps instead of one simple action.
Because the clinic did not maintain reliable data on confirmed appointments and actual sales, I deliberately do not attribute the generated leads to a specific number of patients or revenue.

The tour operator specialises in health resort and wellness stays.

The goal was not simply to improve traditional search visibility. The task was to make the company easier for AI systems to find, understand and recommend when people ask natural-language questions such as: where to book a health resort stay, which company to trust with the trip, or which option would be suitable for elderly parents.

The challenge with these searches is that the user is no longer looking for a list of links. They are asking AI to compare options and recommend one, with an explanation of why.

This means that simply adding the right keywords is not enough. AI needs to understand who the company is suitable for, what specific problems it solves and why it may be a good choice in a particular situation.

What I worked on
Over three months, I worked on the company’s website and AI-search visibility:

• audited the brand’s presence in AI search;
• researched real customer questions and decision-making scenarios;
• grouped queries by intent and stage of the customer journey;
• reworked existing website pages;
• created new pages around specific customer needs and situations;
• developed different levels of content, from commercial landing pages to detailed answers to customer questions;
• made the company’s practical value much more explicit: help with choosing the right health resort, travel planning, transfers, documentation, accommodation and support for elderly parents.

Results
After three months, ChatGPT began not only mentioning the company in test queries, but actively recommending it as a suitable option.

For one commercial-intent query about where to buy a health resort package, the company appeared first, described as “my first choice for this region”, with a link to its website and a phone number.

An even more revealing result came from a query about arranging a health resort stay for a mother aged 60+.

ChatGPT explained why the company could be a good choice: a manager could speak directly with the parents, help organise the route and transfer, check the necessary documents, make sure the guests were properly accommodated and help select a suitable room.

In other words, AI was no longer repeating generic marketing claims. It was using specific advantages that had been deliberately and consistently explained on the website.
GEO is not about trying to “make ChatGPT mention your brand.”

The first step is to make sure AI can clearly understand who your company is for, what problem it solves and why it deserves to be recommended in a particular customer situation.