Customers are ‘going to buy the outcome–and that’s where it’s evolving,’ Bill Karpovich, CEO of Synaptic Era, tells CRN.
Synaptic Era is building a solution-as-a-service and agent-as-a-service business model to the Salesforce ecosystem that brings together intellectual property with managed services–backed by a $5 million seed round of funding recently raised.
Bill Karpovich, CEO of the Baltimore-based Salesforce solution provider, told CRN in an interview that Synaptic offers prebuilt vertical-specific Concierge chatbots customers add to their websites with guardrails already added and Synaptic manages the bots, keeps the data the bots leverages current and bills based on the outcomes as opposed to hourly. The company also maintains more traditional consulting services around Salesforce products.
“Who wants to buy an hour? An hour is an input,” Karpovich said. Customers “want to buy an output. (They are) going to buy the outcome–and that’s where it’s evolving.”
[RELATED: Salesforce To Buy Informatica: 5 Things To Know]
Synaptic Raises $5 Million
Jonathan Perl, general partner at Boulder, Colo.-based venture capital firm Boulder Ventures, one of Synaptic’s investors, told CRN in an interview that he’s known Karpovich since 2005 and led Series A rounds of funding for prior companies Karpovich led.
Alongside Boulder Ventures, other seed investors are Inner Loop Capital, Bloomberg Beta and CoFactor Ventures.
Karpovich co-founded one of those companies, Zenoss, in 2005 and led it as CEO until 2015. Virtana acquired the AI operations (AIOps) vendor this May. Karpovich took on the CEO role of the other company, Youreka, in 2020. He led the Salesforce-native mobile automation provider through its sale to Dispatch Technologies in 2022.
Part of the attraction to Synaptic was getting in on the market for AI agents and a new wave of enterprise automation, Perl said. He was also won over by Karpovich’s ability to marry past experiences in selling software and selling migration and other services around software.
“This was really a bet on the team (and) the market,” Perl said.
Karpovich sees his company as a possible evolution for how solution providers work with customers. System integrators can reach up to 70 percent of the total cost of ownership (TCO) of a Salesforce project, he estimated. But through the prebuilt AI chatbots and through leveraging AI internally, the CEO hopes to bring costs down for customers and create a longer-term relationship with them outside of one-off project work.
“What this is going to do is create this opportunity where we can deliver a solution that is actually what the customer wants,” he said. “We won’t bill them for the hours. We’ll just bill them for that solution and for that outcome ongoing.”
Here’s more of what Karpovich had to say about Synaptic’s chatbots and business model.
What are customers demanding from Synaptic lately?
Industry AI agents are keeping us busy.
What’s exciting for me, being in the middle of enterprise software for 30 years, is–the value opportunity here for businesses is incredible.
That’s the business opportunity–how to really help organizations figure out where to start with AI agents. Because there’s so much you could do with them. What should you do? How do you get started in a way that you can start to prove value and get on the court, begin the cycles of learning. Because it is a whole new way of thinking and operating.
And how do we make it safe and secure so that you’re not taking undue risks in that process?
That’s what our mission is–is to make this technology digestible for businesses as well as government agencies.
It’s all about putting it into a package that is very consumable and gives them a great starting point to get on this journey to create value in a responsible way.
What industries are Synaptic focused on?
No. 1 is the health care market, where we have a lot of history delivering cloud applications and, now, bring the next generation with these AI agents.
We have government and public sector, which is both federal and state and local. And then third are companies–whether it’s financial technology or (software-as-a-service, SaaS).
What we have is a library of industry agents. While we have the know-how and can help you consult, what we want to do is come to you with a prebuilt solution so that out of the box, it does 80 percent of what you need. That accelerates the time-to-value.
(It) helps folks get return on investment more quickly with something prepackaged
We have what we call our Concierge family … a chatbot you would put on your website to help a customer, whether it’s a patient or a member in a health care organization or whether it’s a citizen in the public sector or whether it’s a client in the technology world.
They’re prebuilt for those scenarios with the compliance needs of each of those industries.
When you typically would put an AI bot into an environment, the big question can be, ‘Where is it getting the data right?’
It’s the quality of the data and having it in a place where you can use it and access it so that we don’t have the garbage in, garbage out problem with the AI capability.
Every company has done a lot of work to get at least one pool of information into a curated place–and that’s your website. Your website is basically a little knowledge base.
Traditional websites today, you either navigate through clicks or you do a traditional search, which is a keyword-based search. And keyword-based search is really inferior to natural language in a whole number of ways because you can speak to it. You can have a conversation. It’s the reason why, of course, we’re all using ChatGPT. We know we never have to spell anything correctly. We just say the utterance that’s on our mind, and it figures it out.
Why isn’t every experience like ChatGPT? The answer is, it will be.
How does this translate into opportunities for Synaptic?
There are a lot of things we can do inside your internal processes in a number of other areas. But let’s first start with this idea of making your public information much more accessible. Answer questions for people. Because people are still just frustrated with, ‘Why can’t I have a conversation with your website?’
That’s what our Concierge products do. As opposed to requiring you to build a whole new knowledge base (which can take months).
(Instead), let’s go find a way to train the bot on your existing information. We’ve built a facility that can train the bot on your existing information (and) as you change it.
Phase one is, give me a great experience navigating the website. If I’m a patient, it may not answer all the questions about–what do I do about a certain condition? In fact, our patient Concierge has prebuilt into it the guardrails to make sure it doesn’t go into clinical questions that we don’t have enough data to answer. We’re not going to give you a recommendation on what to do with your heart problem.
What we can do is tell you exactly where you should park if you’re going to come to a division of this hospital for your heart condition. It’s dealing with all the operational stuff that there’s already plenty of public information and then knowing when to stop in that conversation and bring in the right professional–the right physician, in this case.
How is your business model evolving in the AI era?
We are a partner in the Salesforce ecosystem. We’re playing that channel game. What we have done is become evolved to be AI first.
In this AI world, people really need help not only with implementing and finding these starting points as described, but they also need help managing this stuff.
No one knows how to manage a chatbot. No one knows how to monitor it. No one knows how to analyze it. No one knows how to make sure it’s sticking to its guardrails and evolving for compliance.
We’ve evolved from more of a classic professional services model to what we call agent-as-a-service. Everything is delivered as a service so that we can basically be accountable for delivering success. And as opposed to delivering based on labor hours, we’re now billing based on outcomes.
Salesforce themselves had to go from billing on seats to billing on usage. Now if you’re in the solution game, as opposed to just billing based on hours and in those traditional measures, people are much more oriented toward–what’s the business outcome that I’m getting? And that’s where our model has evolved. It’s a longer relationship, not just that one-time (engagement) for the hours. We own the solution. We’ll make it current. And we’ll bill you for the outcome.
You’re almost an ISV, right?
We look a little bit like a SaaS software company. We look a little bit like a consulting company. We look a little bit like managed services.
That’s why we came up with the term solution-as-a-service.
When it’s an agent, agent-as-a-service. We would just say, ‘We’re a solution provider. But we’re doing it in this way that we don’t have language for. So we’ll invent it. We call it solution-as-a-service.’
My personal journey has been in the whole history of cloud. First it was a software-as-a-service company (USinternetworking, or USi, where Karpovich served as vice president of product and strategy–AT&T later acquired the company in 2006), back when it was called ASP (application service provider).
We were always trying to automate things, bring them together and create more value for customers.
Even in the SaaS world … in the Salesforce ecosystem, about 70 percent of the TCO (total cost of ownership) still goes to the SIs (system integrators).
The AI part of this is not only the solution, but we can actually deliver more efficiently because we’re using AI internally. What this is going to do is create this opportunity where we can deliver a solution that is actually what the customer wants. We won’t bill them for the hours. We’ll just bill them for that solution and for that outcome ongoing.
That’s quietly the transition in the packaging that’s happening.
We see that business model changing. And that’s why we’ve pushed this concept of solution-as-a-service because we think it’s a much better way to buy it. Who wants to buy an hour? An hour is an input. I want to buy an output. I’m going to buy the outcome–and that’s where it’s evolving.
Is listing on the Salesforce marketplace important to Synaptic?
(Salesforce) AppExchange is a huge resource. And they’re evolving. Now there’s an agent exchange. That’s where you can take your very specific AI agents and make them a part of that ecosystem play.
Ecosystem marketing is critical. And that’s why we like working inside these ecosystems. And same thing if you’re in the Microsoft ecosystem, AWS (Amazon Web Services). Everyone has their marketplaces.
Those marketplaces are evolving. Just like we saw with the evolution of the cloud marketplaces over the last 10 to 15 years. Now the agent marketplaces and the AI marketplaces that specialize in corners are becoming areas of rapid growth.
How large is Synaptic?
Our team is pretty compact … (about) 30 or 40 people working on the business at any point in time.
But our goal is to scale the business by many factors and to not have to expand the team that significantly.
Getting more efficient every day, how can we scale the business without having to add just as many people? That’s the difference between the old business model and now. As we continue to scale, we will have to add incrementally, but the ratios are much different than they were in the traditional business.
Anything you want to see from Salesforce that would help your business?
They’re doing a great job evangelizing. That’s really a big help for a partner is that you can now kind of come in behind the marketing machine that is Salesforce. And they’re obviously doing everything they can to create awareness. And building sales staff (Salesforce recently disclosed plans to hire at least 1,000 salespeople to grow the company) for us is terrific. That’ll just mean there’s more qualified prospects.
That part of it is really useful and really creates opportunities for partners because they’re in many ways long on the sales capacity.
One of the challenges that Salesforce has had is pricing. (The vendor’s new Flex Credits value-based pricing) is really helpful. It makes it much more flexible and more affordable.
(Innovation) is moving so fast. And so imagine how customers feel. But again, for us as a partner ecosystem, we see all these things as opportunities because we can stay close to things. We can distill what’s important to the customer.
In our case, we’re not only not only taking the knowledge, but putting it into an add-on and a software product that can really make it a lot easier for the end user to consume. We’re happy with what they’re doing. We’re really impressed by the platform.
It’s difficult (for startups) to get through compliance. And then how do I trust (them) with all the concerns and potential risks. How do I trust a little old startup? That’s the nice thing about being in the Salesforce ecosystem and building on top of Agentforce.
This hard work has been done through the trust layer, and all the architecture has been built out to make sure that there isn’t any data leakage to deal with all kinds of other guardrails and compliance concerns. And you don’t have to trust us as a relatively small company.
How’s the competitive landscape for Synaptic and Salesforce when AI could bring down the barriers to entry into any technology market?
What we see is people making decisions on their AI platform. It’s been encouraging to see how Salesforce has gotten positioned there–positioning themselves as truly an enterprise platform. And having (communications application division) Slack and other pieces that make it more broadly applicable is really powerful.
The CRM (customer relationship management software) is going to be the strength. Salesforce also has been smart to be open in any one of these things. You can only control so much. You need to be able to work with the various LLMs (large language models), the various clouds.
Any strategy that’s going to succeed is going to have to–even if it is a platform strategy–still be relatively open just because of all the value that happens outside of the specific platform.
The pricing (of AI) was the major concern because you have things like DeepSeek and open source to create value. And there’s more value being on a platform like Salesforce, but the economics have to make sense.
Part of the opportunities over time in this market are–how do you manage a heterogeneous AI landscape? Just like in the cloud world, we had hybrid cloud. We have multiple cloud providers. Need to figure out how to manage across multiple clouds and geographies. In a similar way, there’s going to be this heterogeneous AI management challenge that is going to happen down the road. We all should be prepared for that.
