Interesting. My team just went through this to plan offsite. Took about 1/2 a day of one person's time. So I asked your tool to help me:
"I want to have a two day offsite for a team of 12 in Cambridge in April."
It then started pulling up results in Cambridge UK. I meant Massachusetts. I didn't say that in the prompt, but I figured since there are two equally famous Cambridges, it would ask me for clarification.
I redid it specifying Massachusetts and it worked pretty well (although all the options it found were about double the price of what we actually booked).
An interesting idea!
BTW I didn't continue, but I assume you manage the whole booking process? How do deal with questions from the venue and other human in the loop issues?
Cool that it adapted from UK to Massachusetts. Do you remember what was the price per person per night? Our client usually book around $200-500 room rate per night.
Yes we manage the whole booking process, I think the true value is here as aggregating venues with AI nowadays is quite easy .
For questions between venues and client, we have a specific chat but the Agent can also try to reply if it knows the answer or decide to contact the venue via email (we haven't released that tool yet)
I asked it for locations in the Randstad (area in The Netherlands around Amsterdam) and I got:
1. Hoofddorp, Noord-Holland, Netherlands (actually ok location)
2. Marysville, Ohio, United States
3. Lisboa, Portugal
4. Nashville, Tennessee, United States
5. Kenmore, Washington, United States
6. Golden, Colorado, United States
I would expect there to be some reviewer agent that ensures that all found locations are at least within the same country?
We are actually pushing an update on that, we want to make the distinctions between Exact Matches (respecting destination criteria) and Non Exaft Matches , that. are suggestions.
The below options were suggestions (because we have limited supply in Netherlands ) but it not clear at all for the user:
2. Marysville, Ohio, United States 3. Lisboa, Portugal 4. Nashville, Tennessee, United States 5. Kenmore, Washington, United States 6. Golden, Colorado, United States
I did not see this mentioned: a very hard part of organizing events for remote teams is dealing with visas, for example, choosing to host an event in Europe will often prevent someone from India attending. Do you handle that, finding a location that is suitable for the largest number of the team?
Another challenge is travel, e.g: scheduling an event in Europe for a distributed team of U.S. people during bad weather leads to people stranded at airports, missing the event.
I think this is a great idea, but I am surprised to learn that organizers are spending most of their time communicating with hundreds of venues. Once you have a location and budget, finding a venue is straightforward.
Hello, the visa question is quite recurring especially for larger team that have distributed teams.
The algorithm is quite smart about that because it has legacy data from ~1500 events. IT has been shown a lot of example of" When a team is located there, they have VISA issues --> they should actually go there because it worked in the past for XYZ client"
For example the algorithm will know that Dominican Republic is usually VISA frienly and can send larger groups there etc.
For the second part: Finding a venue with the requirements is hard, our clients really have a pain point, but what it is even harder tis to agree on terms, prices, food, meeting space, transportation coordination etc, it is a logistical nigthhmare
> For venue recommendations [...] we do not rely purely on the language model. We embed both user requirements and venues into vector representations and retrieve candidates using similarity search. Hard constraints such as capacity and dates are applied first, and results are ranked before being presented.
Huh this surprised me as a forgone opportunity.
I heard second-hand about the process for organizing our last offsite. Searching for venues was not the time-consuming part.
The time-consuming part was actually engaging with the venues to confirm specific details not available online. Our teammate who did this engaged with _hundreds_ of venues. It was a lot of work on their part ... and probably not the most fun part of their job.
You are right, venue recommendation is only the first step or the process.
What is time consuming is the communication with the venue to agree on "terms" , this is exactly why if you click on "Request Quote" you will have a real quote process with the venue that will share all the details and cost estimate with the client , we also offer to talk directly with the venue manager to talk about the final details and close the deal, that is where the value is at --> end to end booking process (not just aggregating results)
Maybe I misunderstood the use case this is for, but I asked it to search for a "venue for team outing for 8 people in the City of London" and it just came up with random hotels in London. I clarified that I'm looking for venues for a team activity and that it needs to be limited to the City, but it just returned hotels again.
I initially didn’t read this post, fixated on “company event”, and thought it could be used for a single-day, one-off “thing to do tonight given (location) and (preferred activities/venue type) after (work end time)” It presented hotels and then some potential activities, but didn’t look up the time. When I asked about what is open after the time, the agent seemed to realize my request was not in your typical use case flow and gave me a refusal.
It’d be cool to offer one-off event suggestions, but I understand that’s probably not as easily monetizable.
Right now the AI flow is optimized for multi-day events where people stay at least one night, like offsites, retreats, and conferences. When you shifted it to a same-day “what should we do tonight after work” use case, you basically stepped outside its current planning model, so the refusal you saw is on us.
We do support day events and activities on the supply side, but they are not yet fully integrated into the AI agent flow. Over the next few weeks, we are plugging that inventory into the system so it can handle more one-off and shorter formats.
Monetization is part of the equation, but it is also a product focus decision. We started with the higher-friction, higher-stakes planning problem. Expanding into lighter-weight, single-day coordination is definitely interesting and your comment is a good nudge in that direction.
Since there has been more than one case of this misunderstanding in the thread, I've changed the title to say "retreats" instead of "events". That (to me at least) that implies overnight stays.
Haven't organized large meetups, but for regular enterprise companies this could be a difficult to buy decision, because you have ChatGPT + bunch of connectors which can get company policies.
This could be good idea for event companies who regularly schedule things, but even for them, probably difficult to justify the value when you have access to ChatGPT and other connectors
Are you talking about data security policy for enterprise companies? For that we can just protect the data (SOC2 etc) and comply with enterprise companies.
For the second part, ChatGPT only could potentially aggregated options but It will not get quotes, booking etc, it will stay shallow. The value here is that it gives you good venues vetted by our network but you can also book them and continue organizing your event with process and credibility of a real company.
A retreat is $$$ expensive, you need a real company in the background to insure the booking and finance safety, ChatGPT won't do that.
Important distinction: we are not in the same segment as Booking.com. Most hotel platforms support small group bookings, usually up to around 10 rooms. We operate in MICE, where you are negotiating room blocks, meeting space, F&B minimums, contracts, and attrition clauses. That is a very different workflow from self-serve booking.
LLMs can make search look nicer, but getting an actual group quote still requires going through property sales teams and contracts. That is operational and relationship-driven, not just a UI problem.
Over 1,200+ events, we have also built proprietary data around pricing patterns, responsiveness, and contract structures. That is not publicly accessible today.
Also our proprietary data is unique to us for now.
How would you make it more defensive? I take any idea
Well, I checked your chat cause I'm planning a company retreat to Athens for two teams - one from Poland, the other from Sweden. And the chatbot told me (after saying what a wonderful destination Athens is) that there are direct fligths from both cities to Athens which is simply not true...
Globally, meetings, incentives, conferences, events, and group travel together represent a 500B+ market all in. Almost every mid-sized or large US company runs some form of in-person event each year, whether that is a retreat, sales kickoff, or team meetup. Since COVID, distributed teams have made these gatherings more important, not less.
Corporate offsites are just our entry point because the pain is clear and budgets are structured. Almost every mid-sized or large company runs in-person events every year, and since COVID those gatherings have become more important for distributed teams.
Long term, we are not limiting this to corporate. The underlying problem is group coordination with real budgets, contracts, and logistics. That applies to associations, communities, weddings, large friend trips, and more. Our ambition is to expand into every type of group travel and event where planning is complex and high stakes.
Latency is something we’re actively working on. Because the agent sometimes calls multiple tools (venue retrieval, cost estimation, ranking, etc.), it can feel slower than a traditional search UI. We’re optimizing tool chaining and caching right now, but it’s definitely an area where we need to improve. If it ever feels sluggish, that’s on us.
Filtering UI: Also agree. We leaned heavily into conversation because planning is iterative and constraint-driven, but that doesn’t mean everything should require typing. A hybrid approach (chat + explicit filters/sliders/toggles) probably makes more sense for power users. We already have structured results on the right adding faster, direct manipulation controls there is a logical next step.
Appreciate you calling it out. If you were using this for real, what filters would you expect to be immediately clickable instead of typed?
Has there been a rebrand as of late? What was the product pitch before that? I guess "AI for planning company retreats" (and possibly SaaS for company retreats before that)
This capacity to pivot into these buzzwords shows that at least sometimes they are more phenomenons with marketing (or at least UX) definitions rather than technological ones.
As an AI engineer I made mysefl a promise when started a company:
"You will only use Ai, if it really useful for your customer" IT wasn't from 2021 to 2025 but then LLM reached a point that now the experience for my customer is way better using a completely different product that is a conversation instead of a classic Airbnb search. What do you do when you discover that? You change you product
Also we do not use AI "buzzwords" for the marketing at all, if you go to our website https://www.teamout.com/ we do not even mention it, we focus on the experience you will get at the event.
On TAM, corporate retreats and offsites in the US alone represent roughly a 500M+ venue booking market by our estimates, and that is just one slice, not counting flights, activities, or international events.
Since COVID, distributed teams have made in-person gatherings more important, not less. Almost every company does some form of corporate event, whether it is an annual retreat, sales kickoff, leadership offsite, or team meetup.
Almost all US company do corporate event and retreats, every year.
The bigger bet for us is not just that events are a sizable market. It is that this is exactly the kind of messy, coordination-heavy workflow that AI can now handle. Two years ago this would not have worked. With current multi-step reasoning and tool use, it finally does.
I'm on the technical end but to me this looks like just another ChatGPT wrapper with a Booking.com & flight planner API key. Nothing more. Expedia was on the list of ChatGPT plugin developers in 2023. What's stopping you? What keeps you in business the moment any travel agency decides gets into Gemini, ChatGPT or the like? I'm sure you make a lot of money per commission, but I don't see what is unique about you as company. What stops anyone else from vibecoding what you'e built in an afternoon?
If we were just a ChatGPT wrapper with a Booking API, we’d be already dead.
Our value isn’t the interface it’s the supply. We have direct relationships with hotels that log in daily to quote, adjust pricing, negotiate, and close deals. That’s not something you get with an API key.
You can vibe-code an Airbnb clone in an afternoon. Without supply, contracts, and operational execution, it’s useless. Marketplace take time to build network effect
LLMs can display data. They can’t negotiate, contract, invoice, manage edge cases, or execute group bookings end-to-end.
We’ll distribute via all major LLMs. But the defensibility is in our network and operations and date, not the UI that everyone could replicate
Not the person you're replying to but this response uses different quotation marks and capitalisation compared to your shallow ChatGPT output from earlier. I took the time to write out what I thought and I'm not going to reply to LLM output. If you wanted genuine dialogue put in effort to reply yourself.
Anyone making such a claim should be providing reasons or evidence of their claim about someone else's comments being AI. Not everything is AI just because you don't like someone who uses AI in other areas.
Interesting. My team just went through this to plan offsite. Took about 1/2 a day of one person's time. So I asked your tool to help me:
"I want to have a two day offsite for a team of 12 in Cambridge in April."
It then started pulling up results in Cambridge UK. I meant Massachusetts. I didn't say that in the prompt, but I figured since there are two equally famous Cambridges, it would ask me for clarification.
I redid it specifying Massachusetts and it worked pretty well (although all the options it found were about double the price of what we actually booked).
An interesting idea!
BTW I didn't continue, but I assume you manage the whole booking process? How do deal with questions from the venue and other human in the loop issues?
Thank you for trying it and giving me a feedback.
Cool that it adapted from UK to Massachusetts. Do you remember what was the price per person per night? Our client usually book around $200-500 room rate per night.
Yes we manage the whole booking process, I think the true value is here as aggregating venues with AI nowadays is quite easy .
For questions between venues and client, we have a specific chat but the Agent can also try to reply if it knows the answer or decide to contact the venue via email (we haven't released that tool yet)
Thanks
I am asking because I want to see if my Cambridge (MA) supply overall is too expensive. Have you tried asking for cheaper options? Thank you
I asked it for locations in the Randstad (area in The Netherlands around Amsterdam) and I got:
1. Hoofddorp, Noord-Holland, Netherlands (actually ok location) 2. Marysville, Ohio, United States 3. Lisboa, Portugal 4. Nashville, Tennessee, United States 5. Kenmore, Washington, United States 6. Golden, Colorado, United States
I would expect there to be some reviewer agent that ensures that all found locations are at least within the same country?
Thank you for reporting that. It is a real issue.
We are actually pushing an update on that, we want to make the distinctions between Exact Matches (respecting destination criteria) and Non Exaft Matches , that. are suggestions.
The below options were suggestions (because we have limited supply in Netherlands ) but it not clear at all for the user:
2. Marysville, Ohio, United States 3. Lisboa, Portugal 4. Nashville, Tennessee, United States 5. Kenmore, Washington, United States 6. Golden, Colorado, United States
Thank you for your help
I did not see this mentioned: a very hard part of organizing events for remote teams is dealing with visas, for example, choosing to host an event in Europe will often prevent someone from India attending. Do you handle that, finding a location that is suitable for the largest number of the team?
Another challenge is travel, e.g: scheduling an event in Europe for a distributed team of U.S. people during bad weather leads to people stranded at airports, missing the event.
I think this is a great idea, but I am surprised to learn that organizers are spending most of their time communicating with hundreds of venues. Once you have a location and budget, finding a venue is straightforward.
Hello, the visa question is quite recurring especially for larger team that have distributed teams.
The algorithm is quite smart about that because it has legacy data from ~1500 events. IT has been shown a lot of example of" When a team is located there, they have VISA issues --> they should actually go there because it worked in the past for XYZ client"
For example the algorithm will know that Dominican Republic is usually VISA frienly and can send larger groups there etc.
For the second part: Finding a venue with the requirements is hard, our clients really have a pain point, but what it is even harder tis to agree on terms, prices, food, meeting space, transportation coordination etc, it is a logistical nigthhmare
> For venue recommendations [...] we do not rely purely on the language model. We embed both user requirements and venues into vector representations and retrieve candidates using similarity search. Hard constraints such as capacity and dates are applied first, and results are ranked before being presented.
Huh this surprised me as a forgone opportunity.
I heard second-hand about the process for organizing our last offsite. Searching for venues was not the time-consuming part.
The time-consuming part was actually engaging with the venues to confirm specific details not available online. Our teammate who did this engaged with _hundreds_ of venues. It was a lot of work on their part ... and probably not the most fun part of their job.
That seems like an ideal agent scenario?
You are right, venue recommendation is only the first step or the process.
What is time consuming is the communication with the venue to agree on "terms" , this is exactly why if you click on "Request Quote" you will have a real quote process with the venue that will share all the details and cost estimate with the client , we also offer to talk directly with the venue manager to talk about the final details and close the deal, that is where the value is at --> end to end booking process (not just aggregating results)
Maybe I misunderstood the use case this is for, but I asked it to search for a "venue for team outing for 8 people in the City of London" and it just came up with random hotels in London. I clarified that I'm looking for venues for a team activity and that it needs to be limited to the City, but it just returned hotels again.
You are right, for now, on the AI product, we only support events where people have to stay for at least one nigh: offsite, retreats, conferences etc.
We do support day event and day activities and we plug this supply in the AI in the coming weeks to make the supply stronger and cover more usecases
I initially didn’t read this post, fixated on “company event”, and thought it could be used for a single-day, one-off “thing to do tonight given (location) and (preferred activities/venue type) after (work end time)” It presented hotels and then some potential activities, but didn’t look up the time. When I asked about what is open after the time, the agent seemed to realize my request was not in your typical use case flow and gave me a refusal.
It’d be cool to offer one-off event suggestions, but I understand that’s probably not as easily monetizable.
That is interesting feedback, and you are right.
Right now the AI flow is optimized for multi-day events where people stay at least one night, like offsites, retreats, and conferences. When you shifted it to a same-day “what should we do tonight after work” use case, you basically stepped outside its current planning model, so the refusal you saw is on us.
We do support day events and activities on the supply side, but they are not yet fully integrated into the AI agent flow. Over the next few weeks, we are plugging that inventory into the system so it can handle more one-off and shorter formats.
Monetization is part of the equation, but it is also a product focus decision. We started with the higher-friction, higher-stakes planning problem. Expanding into lighter-weight, single-day coordination is definitely interesting and your comment is a good nudge in that direction.
Since there has been more than one case of this misunderstanding in the thread, I've changed the title to say "retreats" instead of "events". That (to me at least) that implies overnight stays.
Thank you Daniel, it makes sense to me
> Where would you expect this to fail?
Haven't organized large meetups, but for regular enterprise companies this could be a difficult to buy decision, because you have ChatGPT + bunch of connectors which can get company policies.
This could be good idea for event companies who regularly schedule things, but even for them, probably difficult to justify the value when you have access to ChatGPT and other connectors
Are you talking about data security policy for enterprise companies? For that we can just protect the data (SOC2 etc) and comply with enterprise companies.
For the second part, ChatGPT only could potentially aggregated options but It will not get quotes, booking etc, it will stay shallow. The value here is that it gives you good venues vetted by our network but you can also book them and continue organizing your event with process and credibility of a real company.
A retreat is $$$ expensive, you need a real company in the background to insure the booking and finance safety, ChatGPT won't do that.
> Where would you expect this to fail?
Booking.com and similar moving into this space with their own generic AI tool.
Or even Gemini improving their UI so it presents search results more neatly.
Fair question.
Important distinction: we are not in the same segment as Booking.com. Most hotel platforms support small group bookings, usually up to around 10 rooms. We operate in MICE, where you are negotiating room blocks, meeting space, F&B minimums, contracts, and attrition clauses. That is a very different workflow from self-serve booking.
LLMs can make search look nicer, but getting an actual group quote still requires going through property sales teams and contracts. That is operational and relationship-driven, not just a UI problem.
Over 1,200+ events, we have also built proprietary data around pricing patterns, responsiveness, and contract structures. That is not publicly accessible today.
Also our proprietary data is unique to us for now.
How would you make it more defensive? I take any idea
Well, I checked your chat cause I'm planning a company retreat to Athens for two teams - one from Poland, the other from Sweden. And the chatbot told me (after saying what a wonderful destination Athens is) that there are direct fligths from both cities to Athens which is simply not true...
This is great, congrats on the launch. I never would have discovered half these venues / options without this.
Thank you for testing it, and thank you for the feedback
#social https://youtu.be/v54CcKEN0Sk?t=27m00s
How big is this market? Feels pretty narrow if it stays focused on company outings but are there plans for additional categories?
Good question.
Globally, meetings, incentives, conferences, events, and group travel together represent a 500B+ market all in. Almost every mid-sized or large US company runs some form of in-person event each year, whether that is a retreat, sales kickoff, or team meetup. Since COVID, distributed teams have made these gatherings more important, not less.
Corporate offsites are just our entry point because the pain is clear and budgets are structured. Almost every mid-sized or large company runs in-person events every year, and since COVID those gatherings have become more important for distributed teams.
Long term, we are not limiting this to corporate. The underlying problem is group coordination with real budgets, contracts, and logistics. That applies to associations, communities, weddings, large friend trips, and more. Our ambition is to expand into every type of group travel and event where planning is complex and high stakes.
Can it calculate how many people you need to layoff to make the company retreat affordable for next quarter?
Actually I have been of both side of this story.
2019 -- 2020 : we had an office in New York with my previous company for 30 people
2020 -- 2026 : We are fully Remote at TeamOut, we are 30 and we only do retreats
We pend 5X less than a company going to an office everyday
Looks nice. I'd speed up queries and add a filtering UI to the results; don't make me type everything.
Latency is something we’re actively working on. Because the agent sometimes calls multiple tools (venue retrieval, cost estimation, ranking, etc.), it can feel slower than a traditional search UI. We’re optimizing tool chaining and caching right now, but it’s definitely an area where we need to improve. If it ever feels sluggish, that’s on us.
Filtering UI: Also agree. We leaned heavily into conversation because planning is iterative and constraint-driven, but that doesn’t mean everything should require typing. A hybrid approach (chat + explicit filters/sliders/toggles) probably makes more sense for power users. We already have structured results on the right adding faster, direct manipulation controls there is a logical next step.
Appreciate you calling it out. If you were using this for real, what filters would you expect to be immediately clickable instead of typed?
Price, location, time, event type, group size, for starters.
make sense, we are actually starting to enable people to edit these fields on the right panel.
Aren't you scared that we will have 2 concurent ways to control the experience:
- Chat - Buttons
We may have the syndrom "too many cooks in the kitchen" don't you think?
I'd sync the UI to the text.
Ok yes that is for sure , maybe with an udpate in the chat saying "we udpate your dates to XYZ"
>"Agent"
> (2022)
Has there been a rebrand as of late? What was the product pitch before that? I guess "AI for planning company retreats" (and possibly SaaS for company retreats before that)
This capacity to pivot into these buzzwords shows that at least sometimes they are more phenomenons with marketing (or at least UX) definitions rather than technological ones.
As an AI engineer I made mysefl a promise when started a company:
"You will only use Ai, if it really useful for your customer" IT wasn't from 2021 to 2025 but then LLM reached a point that now the experience for my customer is way better using a completely different product that is a conversation instead of a classic Airbnb search. What do you do when you discover that? You change you product
Also we do not use AI "buzzwords" for the marketing at all, if you go to our website https://www.teamout.com/ we do not even mention it, we focus on the experience you will get at the event.
Thank you
[flagged]
Fair enough, I promise Garry is doing fine.
On TAM, corporate retreats and offsites in the US alone represent roughly a 500M+ venue booking market by our estimates, and that is just one slice, not counting flights, activities, or international events. Since COVID, distributed teams have made in-person gatherings more important, not less. Almost every company does some form of corporate event, whether it is an annual retreat, sales kickoff, leadership offsite, or team meetup.
Almost all US company do corporate event and retreats, every year.
The bigger bet for us is not just that events are a sizable market. It is that this is exactly the kind of messy, coordination-heavy workflow that AI can now handle. Two years ago this would not have worked. With current multi-step reasoning and tool use, it finally does.
I'm on the technical end but to me this looks like just another ChatGPT wrapper with a Booking.com & flight planner API key. Nothing more. Expedia was on the list of ChatGPT plugin developers in 2023. What's stopping you? What keeps you in business the moment any travel agency decides gets into Gemini, ChatGPT or the like? I'm sure you make a lot of money per commission, but I don't see what is unique about you as company. What stops anyone else from vibecoding what you'e built in an afternoon?
Fair question.
If we were just a ChatGPT wrapper with a Booking API, we’d be already dead.
Our value isn’t the interface it’s the supply. We have direct relationships with hotels that log in daily to quote, adjust pricing, negotiate, and close deals. That’s not something you get with an API key.
You can vibe-code an Airbnb clone in an afternoon. Without supply, contracts, and operational execution, it’s useless. Marketplace take time to build network effect
LLMs can display data. They can’t negotiate, contract, invoice, manage edge cases, or execute group bookings end-to-end.
We’ll distribute via all major LLMs. But the defensibility is in our network and operations and date, not the UI that everyone could replicate
It's not a human response, it's an AI
I am Vincent, TeamOut's CTO, if you want to see my face you can book a call here: https://calendly.com/vincent-363/15min
no problem for me!
Not the person you're replying to but this response uses different quotation marks and capitalisation compared to your shallow ChatGPT output from earlier. I took the time to write out what I thought and I'm not going to reply to LLM output. If you wanted genuine dialogue put in effort to reply yourself.
Anyone making such a claim should be providing reasons or evidence of their claim about someone else's comments being AI. Not everything is AI just because you don't like someone who uses AI in other areas.
Agreed. Thank you for the support. There was even grammar issues in my response, what sort of Ai is doing grammar mistakes?
The risk isn't "anyone", it's Tripadvisor or Trivago.