February 10, 2026
In partnership with

You’ve probably been told to validate your idea by talking to people. If you’ve been reading for a while, you’ll remember that I’ve given you this advice.

I still believe it to be solid advice. What’s better, it helps a lot of us get moving.

But it isn’t perfect. And that’s for one simple reason: conversations produce opinions. And opinions can feel like evidence, but they aren’t. Not really.

The market is constantly producing something more useful. Search queries. Competitor reviews. Workarounds people have already built for themselves. Maybe I watched too much Mantracker growing up (IYKYK), but I call this process “reading the trail,” because demand leaves footprints if you know where to look.

This week, I’m walking you through a framework for finding those footprints. Three tiers of demand evidence, from surface-level opinions all the way to the strongest signal: willingness to pay.

You’ll also get a quick & easy four-step audit you can run this weekend on any idea, pivot, new service line, etc. The goal is faster decisions with less guesswork.

In this newsletter:

Key takeaways:

  • Zero-stakes conversations produce opinions, not evidence. The stronger signals are already in the market. You just have to go find them.

  • There’s a hierarchy to demand evidence. Potential customer conversations at the base, search and social signals in the middle, existing workarounds above that, and willingness-to-pay tests at the top. Treat them accordingly.

  • You can audit demand in a weekend. Search the landscape, mine competitor weakness, spot workarounds, then test with real stakes.

Your copilot for compliance is here

Ever been tired, stuck, and miserably documenting something it feels like AI was born to do? AI finally delivered.

Delve's AI copilot understands your entire tech stack and tells you exactly what to do next. Just ask it “What's my next step? Take me there” and watch Copilot read your screen and start doing tasks for you. Congrats, you've found the light at the end of the tunnel. Join Wisprflow, 11x, Bland, and more in the future of compliance with Delve.

Book a demo through the link below to get $1,500 off compliance and see Copilot in action.

Weekly Insight

Ruben Gamez was a software developer at a billion-dollar payroll company in South Florida. He was known amongst his friends as business-savvy and always happy to help, so naturally, someone asked for just that. Specifically, they wanted his help writing a client proposal.

Simple problem. Being systems-minded, Gamez searched for proposal-writing software.

But all he found was an MS Word plugin that felt like it belonged in 2003.

Most people would stop here, thinking “eh, that sounds like an annoying problem for my friend to figure out.”

Gamez went the other way. Straight into opportunity assessment mode. To Google Keyword Planner he went.

The search volume for “proposal software” and “proposal-writing software” was nearly inexistent. Nobody was searching for those keywords.

But “proposal templates” had consistent volume. So did “web design proposal template.” People were leaving footprints everywhere, just under a different name.

Gamez continued down the opportunity assessment process. Instead of building a product right away, he validated demand by building blog posts targeting those template keywords. Then he placed a simple landing page with an email signup.

Before long, three hundred people had signed up. Three hundred people who were actively trying to solve this problem with templates.

He built Bidsketch around that signal. The company grew to over 1,500 paying customers, allowing him to quit his day job roughly 18 months after launch.

Fast forward to 2019. He noticed DocuSign was overpriced and overly complex for small teams. He searched up alternatives. Looked at reviews. Found the pattern amongst the complaints.

Then, boom. SignWell was created to address that gap. By 2024, it was doing $5M annual revenue with a 7-person team, without any external funding.

There’s a clear pattern here.

  1. Gamez felt tension in a workflow, whether by experiencing it or by hearing from someone who had.

  2. Before doing anything, he “read the trail.” That is, he went to the market, looking for footprints (search volume, reviews, workarounds).

  3. Launched validation tests and interviewed his ICP from a base position of hard evidence.

We touched on the initial layers of this process in our Startup Phases newsletter. Your phase determines whether you’re solving the right problem or solving it the right way.

Today’s newsletter adds another layer: once you know what you’re solving, you need to know if people are actively looking for solutions right now.

The demand evidence hierarchy looks like this.

Layer 1: Search and social signals

What are people typing into Google, Reddit, and X (Twitter)? This gives you the words people are using to describe the issue.

Layer 2: Competitor feedback

Read reviews on G2, Trustpilot, and AppSumo. Skip the “this software sucks” stuff. Look for specific friction: “Too expensive for small teams.” “The onboarding is confusing.”

These are demand signals.

Layer 3: Existing workarounds

What are people currently using as a band-aid? Maybe it’s spreadsheets. Hired help. Automated, but maintenance-intensive, workflows. If people are doing something manually but would rather not be, you’ve found the footprint of demand.

If people are happy with what they’ve got, it may not be a big enough problem to warrant your time.

Layer 4: Willingness-to-pay tests and conversations

Finally, you have your market test step. By this point, you’ve got context. Now you’re looking to confirm the details.

We covered a powerful set of tools for that last layer in our willingness-to-pay newsletter. This week adds more context to the foundational steps that precede those tools, frameworks, and conversations.

Intent to Action

You can systematize your search, and know if your idea is a good one within a single weekend. Here’s how.

Step 1: Search the landscape

Open Google Keyword Planner. Type variations of your problem statement: “proposal writing,” “proposal templates,” “proposal software.” Watch the volume. Watch the related searches.

If you can’t find anything but truly believe it’s a problem, go to social media, forums, etc. Wherever your ICP hangs out. You’re looking for the kind of gaps Gamez found: people searching for workarounds, even if they aren’t looking for the product category itself.

Spend enough time here that you have a cohesive thesis, you can get this done in an hour or so. Document what you find. The specificity matters later.

Step 2: Mine competitor/market weaknesses

Find the people solving this issue (aka your future competitors), even if they’re solving it differently. Go to G2 and other software review platforms. Read the two- and three-star reviews. Remember: you don’t care about the one-star people that just came to vent. You’re looking for patterns in the friction.

Is everyone complaining about onboarding? That’s a signal you could compete on simplicity.

Is everyone saying it’s too expensive? That’s a lower-tier market waiting.

Write down the three to five most repeated complaints. These are footprints; you’ll follow them ’til you have a trail.

Step 3: Spot the workarounds

Look for places where people have detailed their solutions. If those don’t exist, elicit them yourself. Go to the places your ICP is hanging out and simply ask: “How do you currently solve this?”

Just like Goldilocks, you need to word your question just right.

  • “Do you have this problem?” is too abstract. You know they do.

  • “Would you use this tool? (+ describing your imagined solution)” is too specific. At this point, you want to know what they think.

  • “How are you solving (specific issue)?” is just right. You want to see what they’ve come up with.

If someone describes using Airtable + manual emails + Slack reminders, you’re onto something. They’ve validated the problem through their behavior. It’s the equivalent of Gamez finding template searches.

Follow up your initial question by asking how happy they are with their workaround. If people don’t feel noticeable friction, they won’t change what they’re doing.

For example, you could say, “I saw three reviews on G2 saying (your current solution) doesn’t integrate with Slack. Is that a big friction point?” instead of “Do you think an integration would be useful?”

Workarounds come in three flavors:

  1. Manual processes (time cost)

  2. Third-party tools used in unintended ways (friction cost)

  3. Outsourced work (money cost)

The higher the cost, the stronger the signal.

Step 4: Test with real stakes

Once you’ve got signals, build a simple landing page with a real price, and drive traffic to it by following up with earlier interviewees, leveraging SEO, posting on social and niche communities, etc.

Your goal is to collect high intent signals (like email sign-ups) from the same people you initially found searching for solutions.

From those emails, you start having more in-depth conversations.

From those conversations, you’re off to the races.

Sponsored

97% of finance leaders say their budgeting process is collaborative.
92% also say they face collaboration challenges.

So… what’s really going on?

Back-to-back Zooms and prolific Slack threads may look like collaboration, but it’s busywork dressed as teamwork.

The real thing requires structure: shared assumptions, sequenced inputs, and connected models that can be easily shared across teams.

We asked 250 finance pros what’s working and what isn’t this budgeting season. See how you stack up in Aleph’s 2026 annual budgeting & planning benchmark report.

Just follow what will make users happy, the way a scientist follows the truth.

Paul Graham

Toolbox 🧰

Shows where your target audience hangs out online, what they read, and what they search for. The recon layer that makes everything else sharper.

Dig into keyword volume, competitor traffic, and the exact search terms people use to describe the problem you’re solving..

See whether search interest in your solution is growing, flat, or dying.

What did you think of this week's newsletter?

Login or Subscribe to participate

Reply

Avatar

or to participate

Keep Reading