This month marks Woozle's 10th anniversary. A decade of running investment-grade primary research for hedge funds and private equity. Thousands of calls. Thousands of decisions with real capital behind them. Here's what I've learned about what actually builds conviction—and builds it fast.

1. Think forward, reason backwards.

Before you pick up the phone or send a single email, get crystal clear on the decision you're trying to make. Are you looking to buy a company? Short a stock over earnings? Build conviction on a long-term hold?

Start with that decision and reason backwards. What has to happen for this stock to miss? China slowdown, inventory pileups, new product underperformance. Now you know what you're actually looking for. Now you know who to talk to and what to ask.

Most people start with "I want to learn about Company X" and wonder why they end up with interesting but useless information. Interesting doesn't generate returns. You need to know what decision you're making before you can know what information matters.

2. Start at the top. No assumptions.

Map out the stakeholders for that particular decision. Who has the most purview? Who sees the full picture versus just their corner of it?

The most important screening question is always: who knows what, and when do they know it? Not who you assume knows based on job title. Actually figure out who has visibility into the thing you care about.

This sounds obvious but almost nobody does it properly. They start with whoever's easiest to reach, or whoever they've spoken to before. That's backwards. Map the landscape first, without assumptions, and work through it methodically.

3. Be upfront about what you want.

Tell them what you're trying to learn. Tell them why you're doing this research. Tell them why their expertise matters and how it fits into the bigger picture.

This isn't just about being decent—although it is that too. It's practical.

Most of the questions you're going to ask are detailed, specific, and time-period focused. If you ambush someone with "what were your Q3 inventory levels compared to Q2" without any context, they're going to give you a poor answer because they haven't had time to think about it.

Give them time to prepare. Let them understand the ask. People respond to transparency. You'll get better information from someone who's had twenty minutes to gather their thoughts than from someone you've caught off guard.

4. Go in with deep knowledge.

Every meeting, every call—walk in with as much intel as possible on the consensus narrative. What does the market think about this company? What's the bull case? The bear case? What's priced in?

This matters for two reasons.

First, it helps you separate signal from noise. When someone tells you something, you need context to evaluate it. Is this new information or what everyone already thinks?

Second, the edge lives in the gap between what the market believes and what's actually happening. You can't spot that gap if you don't know what the market believes.

People share more when they feel understood. When you can talk at their level of detail, when you clearly know the space, they open up. They stop giving you the sanitised version and start telling you what they actually think.

5. Be fluid.

Have a base outline of topics you want to cover, why they matter, and what you expect to hear. But don't treat it like a checklist you need to get through.

Getting answers to questions you didn't even think to ask—that's where the real insight lives. If someone starts going down a path you hadn't considered, follow them. Don't interrupt to get back to your script.

Classic Pareto applies: 80% of the insights come from 20% of the questions. If you're getting gold on one topic, stay there. Drill down. Don't move on just because you've got other questions on your list.

The list is a guide, not a contract.

6. Ask why. Then ask why again.

Every insight worth having comes from deeper and deeper why questions. Don't accept the first answer. Don't assume you understand.

Why did you make that decision? Why did that happen? Why do you think that changed? Why does that matter?

You're trying to uncover real motivations, real reasoning, real decision-making. The trends behind the trends. The logic that drives the numbers.

Most people are too polite to keep pushing. Don't be. The gold is always three or four whys deeper than where most people stop.

7. Trust but verify.

People aren't lying to you—usually. But they misremember. They forget details. They think they do one thing but actually do another. They tell you what they believe is true, which isn't always what's actually true.

Don't take anything for granted. Fact check. Cross-reference.

One person telling you something is an anecdote. Five people independently telling you the same thing? Now you're onto something.

This is the difference between primary research that makes money and primary research that just makes you feel informed. Anyone can collect opinions. The skill is in triangulating them, verifying them, and figuring out which ones actually reflect reality.

8. Don't be afraid to ask direct questions.

Most people worry about biasing the sample. That's a real concern if you do it carelessly. But done correctly, direct questions are where you accelerate speed to insight.

Here's how I think about it: ask your open questions first. Let them tell you what they see, what they think, in their own words. Then, once you've covered that ground, ask directly about the specific thing you're trying to figure out.

"We've been hearing X from others—does that match what you're seeing?"

Often they agree and then go into more detail precisely because you mentioned it. You've given them permission to talk about something they might not have thought to bring up.

9. Ask what you're missing.

Near the end of every conversation, ask direct questions about what might surprise you. What's new or different lately? What are they watching that you haven't asked about? Are they optimistic or pessimistic overall, and why?

You don't know what you don't know. They might have the single most important piece of information in their head and never mention it because you didn't ask the right question.

"What would you be paying attention to if you were in my shoes?" is a powerful question. So is "what surprised you recently?"

These open-ended prompts often surface the stuff that actually matters—the things that weren't on your list because you didn't know to put them there.

10. Spend time on the human stuff.

You need to spend time getting to know this person. What are they working on? What motivates them? What's going on in their world?

This isn't small talk you rush through to get to the real questions. This is the real work.

People open up to other people. That connection matters more than your question list. The researchers who treat every call like a transaction get transactional answers. The researchers who treat every call like a conversation with someone they're genuinely curious about? They get the good stuff. The things people don't tell just anyone.

11. Logic over numbers.

The most important factor in good investment decisions isn't hard numbers. It's the logic that drives those trends.

You're never going to sample a full population. Time and compliance don't allow it, and frankly it's not necessary. But if five or six people independently tell you the same story, you've got something.

The question then becomes: is this credible? Is it likely happening more broadly than just these people? Is it significant enough to impact the business?

We've had our best calls with zero quantitative questions—because they aren't always needed to make the right call. Sometimes you just need to understand the dynamics. The reasoning. How pervasive the trend is.

That's often more valuable than a precise number that's probably wrong anyway.

12. Build your source book.

You need to know who you're speaking to, speak to them regularly, and keep the conversation going over time. But more importantly, you need to track how they perform.

Score everyone on what they told you and how it played out. Were they right? Were they reliable? How often do they share useful information?

If someone is consistently reliable and consistently shares valuable insight, you've found gold. Protect that relationship. Go back to them first.

Start analysing what makes them tick. What do your best sources have in common? How do you spot them early? That meta-learning compounds over years and becomes one of your biggest advantages.

13. Most people don't have the nerve to just ask.

Here's the thing: people like to talk. For all sorts of reasons. They like being seen as experts. They like being consulted. They're bored. They've got opinions they've never been asked about.

You don't need to be a master sleuth to get valuable information. You need to pick up the phone, ask questions, share your own reasoning, invite them to poke holes in it.

The barrier isn't access. It's nerve.

Most people in this industry sit behind their terminals hoping the answer will appear in the data. Meanwhile, the person who could tell them exactly what's happening is sitting in an office somewhere waiting for someone to ask.

Go ask. You'd be surprised what people will tell you.

14. Record everything.

Always keep a record of what you're doing. Record calls. Track notes. Store them properly.

What worked? What didn't? What did that source tell you two years ago and how did it play out? You can't improve if you're not tracking. You can't spot patterns if you don't have the data.

No textbook teaches this. No market research doctrine is built for investment decisions. Statistical significance, p-values, margin of error—none of it correlates with whether you'll actually make the right call. I'd argue the opposite. The people most obsessed with methodological purity are often the worst at generating actionable insight.

This is lived experience. Trial and error. Learning what works by doing it thousands of times and paying attention to the results.