Press 1 for Hype: The Rise of AI Receptionists & Sales Agents in Vertical SaaS
"Your Call Is Important to Us": How AI is Taking the Message—and the Market
Hi, I’m Clement. The Big Picture is a newsletter for founders and investors who want to make sense of what is happening in the SaaS industry.
Article of the week
Research and thoughts
Press 1 for Hype: The Rise of AI Receptionists & Sales Agents in Vertical SaaS
More than a year has passed since the global emergence of native generative AI SaaS, and the dust is starting to settle. We’re now beginning to see which approaches seem to work from a business perspective. On the application side (as opposed to infrastructure), one of the models that appears to create viable AI native SaaS is the vertical AI-powered receptionist or sales agent.
If you’re unfamiliar with these types of products, you can just look at the ones that YC backed:
Sameday: AI sales agents for home service businesses like plumbers and roofers.
Zorba: Real estate sales AI copilot.
Distro: All-in-one AI-powered sales platform for counter staff and inside sales at industrial wholesale distributors.
Dodo: The AI receptionist for vets.
Arini: AI receptionist for dentists that answers phone calls, and schedules appointments.
Sandra: AI employees for car dealerships and auto repair shops, starting with AI receptionists.
Drillbit AI receptionists for residential contractors.
These are basically verticalized AI agents that function as receptionists or sales for specific industries.
Here are four personal observations about this type of AI product.
Observation #1: The importance of choosing the right vertical
When choosing a vertical for an AI-powered sales or receptionist solution, imo, the choice of the vertical is key and two of the most important factors are: vertical size and vertical dynamics.
Market size is particularly important if the goal is to build a massive vertical SaaS (VSaaS) company capable of generating hundreds of millions in revenue and less important if you want to bootstrap. Many VCs argue that generative AI and automation unlock vast opportunities by targeting labor spend, but I have to admit that I am not yet fully convinced. Why? Because a significant portion of labor spend—especially physical labor—cannot easily be replaced by software alone, at least not yet. Without a granular breakdown of which components of labor are addressable, it’s hard to assess the true potential of many of these verticals to provide VC level returns. So, yes, there are more opportunities, but I still believe that many of them won’t deliver the returns most VCs seek and are better suited for bootstrapped or lean venture businesses (see observation #4).
Another critical factor is market dynamics, specifically customer health and the presence of market tailwinds or headwinds in the vertical. Ideally, you want to focus on businesses that are stable or thriving rather than those constantly battling to stay afloat. There’s a world of difference between targeting customers with healthy margins and stable revenue and those operating on razor-thin margins while fighting to maintain revenue or even survive.
For example, during a recent customer discovery session, I explored the travel agency space. While the industry is super interesting, clearly I wouldn't build an AI receptionist or sales agent for brick-and-mortar travel agencies. Many of these businesses are struggling and face significant headwinds as the direction of the market seems to go toward big online travel platforms and travel operators.
On my market assessment tool, you ideally want to rank 3 or higher for the customer health dimension, and at least 3 on the market size dimension if you go the VC path. If you don’t want to take VC money early on, market size doesn’t really matter imo (customer health much more).
Observation #2: GTM execution beats product execution at early stage
In my opinion, before $1M ARR, go-to-market (GTM) execution tends to be more important than product execution for AI-powered sales and receptionist solutions.
Why? Mainly because of two reasons:
First, the barrier to entry for creating a sellable AI-powered receptionist or sales agent is relatively low. It’s basically the “thin OpenAI wrapper” syndrome. One or two good developers can build a sellable AI powered product in a matter of months.
Second, AI receptionists and sales agents don’t fall victim to the “all-in-one software” curse that plagues many VSaaS solutions. Unlike other traditional VSaaS, AI-powered sales or receptionist products don't need to offer comprehensive functionality across the entire business to gain traction. They don’t need to be the ERP of their vertical in order to be successful, which shortens considerably product development. They can focus on being the best receptionist or sales agent.
These two factors explain why product development can move quickly in this space. From what I saw, what sets the best teams apart is their ability to aggressively execute on sales and marketing. Strong GTM execution enables companies to scale revenue quickly, often without needing to raise a ton of money.
To be sure, I’m not saying that product doesn’t matter. With a terrible product you won’t go far. And a stellar product combined with exceptional GTM execution is, of course, ideal. But I believe that GTM execution is the more critical factor in reaching the first $1 million in ARR.
Observation #3. The VC track and private equity exits
While I believe the opportunity to build successful VSaaS AI receptionists and sales agents is growing, I’m not convinced that most can scale to hundreds of millions in revenue (see my previous point about market size). Also, unlike horizontal SaaS, which often has a wider pool of potential acquirers and a clearer path to IPO, vertical solutions usually face more limited exit options.
I believe that many of these VC-backed businesses that will reach tens of millions in revenue will ultimately be acquired by private equity firms or rollup funds. And if you want an outcome that is lucrative for the founders, then it’s probably better to maintain a “clean table and avoid chasing “hype” funding rounds with inflated valuations that can create unnecessary pressure to deliver hypergrowth. Companies that can’t meet these expectations often find themselves stuck, with limited options for future financing or exit opportunities.
Many VC backed AI receptionists and sales agents should probably prioritize sustainable growth to ensure that their company remains attractive to this type of buyers.
Observation #4: The lean venture or bootstrapped startup track
I personally believe that the real explosion of opportunities AI brings to entrepreneurs lies in bootstrapped or lean venture startups.
Given the characteristics I’ve outlined earlier—such as the ability to build great products quickly with a small development team, the potential for rapid revenue growth through aggressive GTM execution (because these are emerging and not yet saturated markets), and the limited options for large-scale exits—I think we’ll see more successes with this approach. These entrepreneurs will bootstrap or raise modest funding rounds (between $1M and $5M in total) and go on to create AI receptionists or sales agents generating millions in ARR, or even exceeding $10M.
It’s possible that I’m wrong, but my sense is that this model is particularly well-suited for many verticals. In contexts where heavily funded AI receptionist or sales agent startups are likely to hit a growth ceiling that fails to meet investor expectations, lean venture startups seem much better adapted. Overcapitalized startups in these niches often end up trapped by the need for hypergrowth. By contrast, bootstrapped or lean venture companies have the flexibility to grow sustainably while remaining super attractive for PE types of buyers.
Business Breakdown of the Week
Interesting startups and products
Since I covered vertical AI receptionists in my article, let’s dive into one for this week’s business breakdown.
To do this, I asked Jana (The AI VC agent I’m working on) to evaluate Dodo, an AI receptionist designed for veterinarians, using my assessment toolkit.
Below, you’ll find her assessment along with my personal take.
Market Assessment
Question 1: Market Equipment Rate
What percentage of the market is already equipped with software that solves the same problems as Dodo does?
What Jana picked:
Rating 3/5: Market equipment rate of 40% - 60%. “Dodo's target market (veterinary clinics) is at a transition stage. While many clinics use software for administrative tasks, the integration of AI-driven voice receptionists is relatively new and represents a shift from traditional systems to advanced automation.”
What I picked:
I would have rated this question just a bit higher at 4/5. From the perspective of AI-native receptionist software, the adoption rate is much lower (very few vets have adopted AI receptionists). However, many veterinarians use some form of online appointment booking system, which redirects an increasing number of calls. As a result, I consider these systems as part of the receptionist software stack.
Question 2: Competition & Differentiation
How competitive is the market, and how difficult is it to differentiate the product?
What Jana picked:
Rating 3/5: Moderate Competition. “There is competition from broader AI receptionist and veterinary software providers, but Dodo's niche focus on veterinary-specific protocols and deep integration with clinic workflows offers differentiation opportunities.”
What I picked:
I would have answered this question differently and would have rated it 4/5: “Limited Competition. Some competition, but there's still clear room for product differentiation.” I think we’re still at the beginning where everything needs to be invented product wise.
Question 3: Customer Health
How would you describe the overall financial health and stability of Dodo’s target customers?
What Jana picked:
Rating 2/5: Mostly Strained but Viable Customers“Veterinary clinics, while essential and increasingly in demand due to the pet boom, often operate with limited budgets and face challenges such as high staff turnover and thin margins, making affordability and ROI critical for Dodo's success.”
What I picked:
I would have rated higher: “3/5 Mostly Stable Customers. Your target customers are financially stable, with steady revenues but don’t necessarily experience robust growth or benefit from particularly favorable economic/market conditions”. Jana has some good points about staff shortage and turnover, but looking at the stats, overall veterinarians tend to have lower default rates and longer business lifespans than most other types of businesses.
Question 6: Market Size
How would you evaluate the size of your market in terms of its potential to deliver VC-level returns?
What Jana picked:
Rating 3/5: Potentially VC Return Compatible“The veterinary market shows significant growth potential due to the increase in pet ownership and the ongoing staffing challenges, but for Dodo to achieve VC-level returns, it may need to expand its product suite or branch into adjacent markets such as general medical practices or broader SMB automation.”
What I picked:
I picked the same rating for the same reasons :-)
Interestingly it seems that Jana has a tougher (maybe more realistic) assessment of the market compared to me (I’m a nice person).
Let’s move to the pain point assessment.
Pain Point Assessment
Question 1: Frequency
How often does this pain point arise in the work of the target users?
What Jana picked:
Rating 5/5: Daily. “The pain point of managing client communications, particularly phone calls for scheduling and inquiries, arises daily or even multiple times a day. Veterinary clinics often experience high call volumes, which can overwhelm reception staff, leading to missed calls and decreased operational efficiency.”
What I picked:
The same 🙂. It smells like a “high usage” type of product (which is good).
Question 2: Complexity
How complex is it for the user to solve this pain point or complete this task?
What Jana picked:
Rating 4/5: High Complexity. “Handling calls effectively involves several steps, such as understanding the client's needs, providing accurate information, and scheduling appointments while ensuring no overbooking occurs. This process becomes complex during busy periods or with staff shortages, necessitating a robust solution like AI assistance.”
What I picked:
The same 🙂. And I think this is also why it’s not (at the moment) trivial to build an AI receptionist that is more efficient than a human.
Question 3: Task Importance
How important is it for the user to solve this pain point or complete this task?
What Jana picked:
Rating 4/5: High Importance. “Efficiently managing client calls is crucial for maintaining clinic operations and ensuring client satisfaction. Failure to handle this task effectively can result in lost revenue, client dissatisfaction, and a tarnished reputation for the clinic.”
What I picked:
The same 🙂
Question 4: Pain Point Awareness & ownership
How aware are users of this pain point? Is it clear who solves it?
What Jana picked:
Rating 4/5: High Awareness and Clear Ownership. “Veterinary clinics are aware of the challenges in managing client communications and the impact on their operations. Reception staff or administrative teams are typically tasked with addressing this pain point, though solutions like AI receptionists are gaining traction as a viable option.”
What I picked:
The same 🙂. I like this kind of pain point, where users are aware of the issue and there is clear ownership of who is solving it. Some might argue that AI threatens the very job of receptionists, but due to the nature of the role (see the frustration-related question) and the labor shortage, it likely won't be a problem.
Question 5: Required Expertise
What level of expertise is needed to solve this pain point or complete this task?
What Jana picked:
Rating 2/5: Low Expertise Required. “Addressing client calls typically requires basic communication and organizational skills, which most reception staff can handle with minimal guidance. However, during busy times or when dealing with demanding clients, some training in client management or scheduling software can improve performance.”
Question 6: Emotion and Frustration Level
How much frustration or emotional stress do users experience related to this pain point?
What Jana picked:
Rating 5/5: High Frustration. “Clinic staff often experience significant stress when managing high call volumes, especially when short-staffed or during peak hours. Missed calls, long wait times, and dissatisfied clients add emotional strain, making this a pain point they strongly desire to solve.”
What I picked:
The same 🙂. TBH, what is more frustrating than answering people on the phone all day long?
Question 7: Willingness to Pay
How willing are users to pay for a dedicated solution to this pain point?
What Jana picked:
Rating 4/5: High Willingness to Pay. “Veterinary clinics recognize the value of streamlining client communications and minimizing operational stress. They are generally willing to invest in AI-driven solutions like Dodo, especially if it improves efficiency and reduces the workload on staff”
What I picked:
The same 🙂. But to be fair, I think there is real questions about safety and that AI doesn’t go crazy when they answer customers. So while willingness to pay for this pain point is high, you need to reassure your customers a lot before they buy it.
In terms of pain point assessment we basically agree on everything Jana and me.
That’s it for this week!