Week 6 & 7: Tackling Low Conversions; Finding My #1 Customer Opportunity & Competitive Differentiator
Pivoting my channel, not my ICP; A look at my Opportunity Solution Tree & selecting my #1 opportunity; Competitive Research with AI
Background
Over the past 7 weeks I’ve been trying to build a B2B startup from scratch, with AI by my side as a powerful (yet sometimes surprisingly unreliable) sparring partner.
My goals:
Let you look over my shoulder as I apply my product management practices in real life. No abstract theory, but concrete examples.
Share everything I learn about AI along the way
In an ideal world: 5 Letters Of Intent (with pricing and timelines) signed with potential clients after 13 weeks - although that goal seems less feasible today. I’m definitely feeling the pain of only being able to dedicate 5 hrs/week.
Missed the start?
My goals and principles: Building a Bootstrapped B2B Product with Gen AI (0 to 1)
Week 1: Hunting for My Underserved Niche ICP with AI & real-world signal (ICP smoke test)
Week 3&4: Zeroing In On My ICP With Interviews & Prototype Tests
Week 5: The User Lab. A Guide to Integrating Interviews, Landing Pages, and Prototypes
Quick Navigation Through This Post
My priorities during weeks 6 & 7
Priority #1: Improving outreach-to-discovery call conversion rates
Priority #2 Opportunity Solution Tree; Choosing my #1 opportunity (a video walkthrough of my OST today, how I selected my #1 opportunity)
Priority #1: Improving outreach-to-discovery call conversion rates
My biggest problem - surprise surprise - has been getting potential champions on a call.I’ve been using cold outreach via LinkedIn to Field Service Managers working with fewer than 20 service technicians (Linkedin Sales Navigator + LinkedRadar). Over the past six weeks, I sent over 400 connection requests and InMails to Field Service Managers with fewer than 20 service technicians. The result? A meager 3 calls. That’s a 0.75% conversion rate.
This low conversion rate forces a critical re-evaluation. Do I:
Change my Ideal Customer Profile (ICP)?
Change my outreach channel?
Change my messaging?
Let’s break down the decisions.
Change ICP? → No
Despite the low conversion, I'm holding firm on my ICP for at least another three months. Here’s why:
Underserved Niche: My initial conversations revealed a significant lack of dedicated software adoption in this segment. They're often wrestling with paper forms, email, WhatsApp, or clunky Google Docs on mobile.
Low Bar for Software: The three managers I did connect with were genuinely excited by my simple prototype. Two out of three were even responsive to follow-up, which is a good signal.
Clear Pain Patterns: I'm seeing consistent pain points emerge, which is gold for refining the product. (More on this in a future post about Opportunity Solution Trees).
First Signals Willingness To Pay: Two out of three interviewees have indicated that they would be willing to pay several $100/month for a solution like the prototype.
Initial Rationale Holds: My original reasons for targeting this ICP still hold.
Of course, none of this counts as validation. People rarely do as they say. I’ll need to do far more interviews and follow up with real-world-tests to learn more.
Change channel? → Supplement LinkedIn with new channels
I don’t want to abandon LinkedIn entirely, but aggressively explore additional channels alongside it. My initial reliance on LinkedIn was due to two factors: validation from Field Service Management influencers and my own familiarity with the platform. Frankly, neither are strong enough arguments to stick exclusively with it given the current results.
Possible channels to explore:
Partnerships with large manufacturers: Tapping into existing networks can bypass the cold outreach hurdle.
Marketing "Fake Door" Tests: This involves prioritizing a key opportunity, ideating 3 strong solutions, and then running paid campaigns (Facebook or LinkedIn TBD) for each solution idea, driving sign-ups to a waitlist. This will provide real demand signals.
Cold Calling (Phone): I might need to get over myself for this one
Industry Trade Shows: Direct, in-person interaction can build rapport far faster than online methods.
Change the message? → Yes. Quality Over Quantity
This is a definite "yes." My current "spray and pray" approach isn't cutting it. I need to pivot to more personalized outreach.
John Skitt, co-founder at Pathfinder Product Partners, shared a strategy that's currently working for his team:
LinkedIn Connection Request → Personalized Video: Manually recorded for each prospect. → Personalized Voice Note
This approach prioritizes manual, personal, and authentic engagement over automated mass outreach.
I'll be testing two distinct messaging paths:
Connection Request → Generic Video Invite to Interview: This video will outline what I've learned so far and provide a quick walkthrough of the prototypes.
Connection Request → Personalized Video (without prototype walkthrough): If this yields a higher response rate in a specific niche (e.g., solar panel installation or HVAC companies), I'll then "vibe-code" personalized prototypes for those segments.
My thought model in a diagram:
The path that I’m taking marked in green:
To all the GTM, Sales, and Growth professionals reading this, your advice is invaluable. Feel free to reach out to me on LinkedIn. I’m all ears.
Priority #2 Opportunity Solution Tree; Choosing my #1 opportunity
What does my OST look like today, and what #1 opportunity do I want to learn more about?
A quick run-through in video form:
Priority #3: Competitive Analysis; Differentiator
Important notes upfront
First things first: the word ‘competitor’ is misleading’. Instead, I prefer to think of the alternatives my ICP is using or considering to get their job done. Excuse the confusing headline, but I figured the term ‘Alternative Analysis’ wouldn’t ring a bell.
1. Thinking solely within your product category is a trap.
In my case, there are plenty of software tools in the “Dedicated Field Service Management software solution” category, but these haven’t found much adoption with my ICP (field services companies with fewer than 20 technicians).
2. The alternatives you’re up against depend on your ICP
The "alternatives" you're genuinely up against hinge entirely on your Ideal Customer Profile (ICP). Most startups inflate their competitor lists because they're blurry on who they're actually serving. If I’m building for enterprise field services companies, I’m up against Salesforce. But Salesforce is far too heavy-handed, complex and expensive for my ICP. Want to uncover your true adversaries? Ask your ICP directly: "What are you using now, and what else have you tried to get the job done?"
In reality, I’m primarily up against manual flows:
Engineers write service notes on paper/ key them into their phones, and enter them into spreadsheets or forms in the office after finishing their jobs
Engineers need to manipulate Google docs or PDFs on mobile and send these via mobile on the job site
Engineers send service notes & pictures to the back office via email or whatsapp
3. Understanding unmet needs > competitive research
Final note on competitive research: Don't get lost in the competitor rabbit hole. Far too much energy is squandered meticulously tracking rivals, and not enough on getting intimately close to the customer.
I learned this the hard way in my first FinTech product role. I obsessed over our incumbent competitors, going to absurd lengths to unearth their hidden, constantly shifting pricing, just so we could undercut them by a few measly cents. It was pointless. Prospects rarely jump ship from a trusted incumbent to a young startup unless you're offering something significantly better or significantly cheaper. A few cents won't move the needle. I'm convinced I would have delivered far more value by spending that time interviewing and shadowing customers instead
Nonetheless, I need a high level overview of the competition within my product category, since my solution won’t exist in a vacuum. This will also shape my thinking around competitive differentiation. Where do I, and where don’t I want to play?
Competitive/alternative research is where AI really shines, especially if you’re looking to get a quick directional read and don’t want to invest hours for a ‘perfect’ result
Using AI for competitor research. Step by Step prompting guide
Tool: Gemini 2.5 Pro
Desired output
A high level tabular overview of dedicated software tools (I’m building my understanding of the manual processes from audience interviews, for this step I want to focus on tools within my product category)
Putting competitors into meaningful categories
Opportunities for competitive differentiation
Step 1: Identify competitor categories
I want to identify meaningful categories for my competitors that will help me decide where to play. I already have an assumption for meaningful competitor categories and am using Gemini 2.5 flash as a sparring partner to poke holes and identify what I’ve missed.
Find my prompt on Notion, under Competitor research - Step 1: Identify competitor categories
My categories:
Horizontal Enterprise Platforms with FSM Modules (Salesforce (Field Service Lightning), NetSuite, ServiceNow (Field Service Management) (Assumption - too heavy for my ICP)
Generic Form Builders (Jotform, Typeform, SurveyMonkey) (Assumption - they are used to create mobile forms that engineers can use to capture service notes, but they don’t offer any capabilities for storage and searchability of reports
Dedicated Vertical FSM Suites (SMB & SME Focused) (Commusoft, Jobber, ServiceM8, Housecall Pro, ServiceTitan (for larger SMBs). (Offer broader FSM capabilities, but some might still be overkill or too expensive for the very smallest teams, and their documentation might not be their strongest suit).
Dedicated Vertical Point Solutions (Documentation/Reporting Focused) (GoAudits; Fulcrum; Truecontext; SafetyAuditor) (This is where my solution would fit)
Step 2: Listing competitor data in a spreadsheet
I want the AI to only look at the websites of the competitors (not at third party websites or forums), to avoid hearsay. I don’t want the AI to spit out a ready-made spreadsheet. Instead I want to manually review each competitor and copy/paste the relevant information into a spreadsheet. This manual step forces me to read through and check the provided information, and make a conscious decision of what is relevant information, and which competitors might be more interesting than others.
Step 3: Pick the 5 most interesting (‘dangerous’) competitors
I asked Gemini to think about all the competitors it’s reviewed, and to choose the 5 most interesting ones, given my value proposition and ICP. I asked it to provide for each suggestion: reasoning for suggesting; strengths of this competitor; gaps/differentiation potential.
Step 4: Propose competitive differentiators
Now that AI has reviewed all the competitors, with special attention to (the ‘winning competitors’ from the previous step and manual review), I asked it to develop 7 potential competitive differentiation strategies tailored to my niche, including pros; cons; and moat potential.
My take aways: Competitive differentiation by offering best-in-class notes capture first (engineers), report search second (managers)
Make it easier than ever for field engineers to capture service notes. Think: intuitive, guided forms, AI-powered quality checks (e.g., photo quality, completeness), Picture-to-text (e.g. to capture serial numbers and readings), Jargon trained voice-to-text, autocomplete/pre-fill based on historic data, etc. Like CoPilot, but for field engineers.
Make it easier than ever for field service managers/back office staff to search historic reports. Think: Search by date range; machine; customer; engineer; job id; job name. Advanced, custom-filterable search capabilities that go beyond simple job IDs to actual content (e.g., "show all reports where pump model X was serviced," "find all jobs with critical safety notes").
Vision: Enable cutting edge knowledge capture and knowledge sharing for field service teams.
Conclusions
I’m keeping the ICP (for now), but changing the GTM/channel. I’ll explore in-person events, partnerships, inbound marketing (SEA & paid campaigns with fake door)
Most interesting customer opportunity: “It needs to be extremely simple and low overhead for field technicians to capture service notes”. This opportunity came up several times during interviews and in Userinterviews.com screener surveys, it was echoed by the ‘experts’, and fits my competitive differentiation strategy.
I want to focus my time on understanding my ICP, not on competitive research. At the same time, my solution won’t exist in a vacuum. Gemini 2.5 Pro drastically sped up and simplified high-level competitive research. Not perfect, but enough for a directional read.
Next week I will ideate solution ideas to this opportunity and devise a test strategy.
Else thanks for sharing your experience! Which program do you use for videos?