The Prospector Gap: Why SaaStr's Custom Agent Stack Is a Build Opportunity for the Rest of Us
Every B2B company I talk to is drowning in fragmented tools and missed follow-ups. Sales reps are switching between a dozen tabs just to answer one email, and leads are slipping through the cracks because nobody owns the follow-up. The data backs this up: we track 142 problems in Sales with an average severity of 3.9 out of 5, and 37 problems in Communication at 3.8 out of 5.
Jason Lemkin over at SaaStr just published a detailed walkthrough of their internal agent-led growth stack, including a custom tool called "The Prospector" that funnels every inbound, outbound reply, renewal, and support ticket into one queue with an AI agent working 24/7. It's a compelling vision, and they claim a 10-minute inbound booking as proof. But here's the thing: SaaStr has a team building custom agents on Replit and wiring up 30+ tools. Most B2B teams don't have that luxury.
That gap—between what SaaStr built and what the average company can actually buy—is where the real product opportunity lives. Our database of nearly 12,000 app ideas is full of problems that map directly onto the capabilities Lemkin describes, but for teams that need an off-the-shelf solution, not a custom build.
The Custom Build vs. The Market Gap
Lemkin's article is a playbook for companies that can afford to build their own agent stack. They started with a Notion helper 16 months ago and expanded to 30+ tools. They built their own meeting booker to replace Calendly, their own prospectus generator, and their own queue system. That's impressive, but it's also a signal of how immature the current tooling is. If there were a dominant off-the-shelf solution for agent-led GTM, SaaStr might have used it instead of building in-house.
We track a category called "AI Agents" in our database, and it's growing fast. The underlying pains are clear: missed follow-ups (severity 4.5/5), sales rep context switching (4.2/5), inbox overload (4.1/5). These are the exact problems The Prospector aims to solve, and they're widespread. The article even admits that AI-generated outbound emails score about 5 out of 10, while most human outbound scores 2 or 3. That's a massive quality gap that product builders can close.
Where Builders Should Focus
Lemkin's playbook has three distinct phases: inbound, outbound, and renewals/support. Each has different challenges and different existing tools. For an indie hacker or small agency, the most accessible entry point is probably inbound. Here's why:
Inbound is more predictable. The article describes a four-step flow: first-party data check, third-party enrichment, personalized prospectus, and email with meeting booker. Each step is currently a separate tool or manual process for most companies. An integrated inbound agent that does all four would be immediately valuable.
Outbound is noisy. SaaStr runs four outbound tools and only unifies the replies. That's because pulling in every outbound email was "too much data to be useful." This suggests an opportunity for a tool that intelligently filters outbound replies and only surfaces the ones that need attention, similar to how The Prospector works.
Renewals and support are underserved. The article mentions that renewals lived "somewhere else" before The Prospector. This is a common pattern: sales and support tools are often separate, but the same agent that can book a meeting can also handle a renewal reminder or a support ticket. A unified queue with different goals per record type is a product concept that hasn't been fully explored by existing vendors.
The First-Party Data Debate
Lemkin makes a strong claim: "First-party data beats every signal tool." Our data suggests a more nuanced picture. The author's own demo uses third-party enrichment (Exa, Sumble, CoreSignal) after checking first-party data. That order of operations matters. For new leads, first-party data is often sparse or nonexistent, and third-party signals are the only way to get context. The real pain, from our data, is not having enough first-party data, which is why third-party enrichment remains essential.
A better framing: first-party data is your foundation, third-party data is your bridge. The most valuable product might be one that automatically merges both into a unified lead profile, so the agent and the human rep see the full picture without manual research. That's exactly what SaaStr built with their Prospector, but it's not yet a plug-and-play solution.
What to Build
If you're an indie hacker or small agency looking to enter this space, here's a roadmap based on the gaps we see:
- Inbound agent for SMBs: A tool that connects to a form or shared inbox, checks your CRM and email history, enriches with third-party data, generates a personalized response, and books a meeting. No custom coding required.
- Follow-up agent for outbound: A tool that monitors all your outbound campaigns, pulls in replies from multiple platforms, and creates a single queue with smart timers and reminders. This is the missing piece for many sales teams.
- Renewal and support agent: An agent that watches for upcoming renewals, checks usage data, and sends personalized renewal reminders, while also handling simple support tickets. This can dramatically reduce churn for subscription businesses.
None of these require building a full platform like 10K. Start with one use case, nail it, then expand. The data shows the pain is real and urgent. The fact that SaaStr built their own solution is proof that the market is ripe for a product.
The Bottom Line
Lemkin's article is a valuable look at what's possible with agent-led growth today. But for most companies, it's a blueprint, not a product. The tools they assembled—Replit, Exa, Sumble, CoreSignal, custom meeting booker—are still developer tools, not business tools. The company that can package this into an off-the-shelf solution will have a massive head start.
We're tracking the space closely. If you want to dive deeper into the specific problems and app ideas we've cataloged, check out our full database. There's a lot of opportunity here, and the window is open.
This article is commentary on the original article by Jason Lemkin at SaaStr. We encourage you to read the original.
Explore more problems and app ideas across B2B SaaS, Marketing Technology.
Browse App Ideas