Chasing The Truth: A Practical Guide to Source Verification and Investigative Research

I've spent years dealing with verification workflows, and most people approach it wrong from the start. Chasing The Truth isn't a catchy phrase — it's a working methodology for anyone who needs to verify claims, cross-reference sources, or build an unassailable research file. It originated in investigative journalism circles around 2017 when a small group of reporters realized that the standard fact-checking pipelines used by major outlets were consistently failing on complex claims involving financial records, geolocation data, and social media metadata. They built something better. This guide covers how to actually use the method in practice. At its simplest, Chasing The Truth is a three-phase verification loop: collect, triangulate, and pressure-test. Most people stop after collect and call it a day. That's why so many articles contain errors that could have been caught in ten minutes. Here's what each phase actually looks like. Collect means gathering every primary source you can find before you write a single word of analysis. This includes original documents, raw footage, unedited interviews, and the primary data behind any claim. Not the news article about the news article. The actual thing. When I started doing this properly, my verification turnaround time went from 3 hours per story to about 45 minutes. The initial collect takes longer, but it eliminates the back-and-forth later when sources contradict each other.

Triangulate means confirming any factual claim through at least two independent primary sources. If one source says a event happened on March 14th at 3 PM, you need a second source that independently corroborates that timestamp and date. This is where most people fail because they accept a second source that is actually just a derivative of the first one — a republished article quoting the same press release. I once spent two days investigating a claim that was built on three sources, all tracing back to a single corporate FAQ page. That happens constantly. Always ask: who originally said this, and what evidence do they have? Pressure-test is the phase that separates amateurs from people who actually produce reliable work. You take your verified claim and actively try to prove it wrong. Search for counter-evidence. Look for alternative explanations. Interview people who would benefit from the claim being false. I remember working on a story about a manufacturing defect where the company had already prepared a beautifully produced rebuttal video. The angle I found that broke their narrative wasn't in their press materials at all — it was in a customs declaration form filed two years earlier that showed the same defect pattern in a different country. The customs document was public record, buried under a government portal that required a specific search string to surface.

Tools and Resources

Chasing The Truth as a formal methodology is available as a free toolkit and reference guide at chasingthetruthmethod.com. It includes template checklists for each phase, a database of commonly manipulated source types, and the pressure-test framework I described above. The documentation is thorough enough to use as a standalone reference, though I'd recommend starting with Phase 1 and building from there rather than trying to implement everything at once. Beyond the methodology itself, a few tools come up repeatedly in practice. First, InVID for video verification — it handles frame extraction, reverse image searching, and metadata inspection in one interface. Second, Maltego for link analysis and entity relationship mapping, which is essential when you're trying to trace ownership structures or organizational connections. Third, Trove and newspaperarchive.com for historical document verification — these two databases alone have saved me from publishing at least a dozen incorrect claims over the years. If a story involves a quote or statistic, there's a non-zero chance it appeared in a publication decades ago with a completely different context.

Get the Full Details

Chasing The Truth Chapter Summary | Ruby Shamir
Chasing The Truth Chapter Summary | Ruby Shamir

Where the Method Fails and What to Do Instead

I need to be honest about the limitations. Chasing The Truth works extremely well for claims that have paper trails, digital footprints, or institutional records. It does not work well for oral history, undocumented events, or situations where all primary sources are controlled by a single actor. In those cases, the methodology can give you a false sense of confidence because you've followed the process correctly but the inputs are irredeemably biased. When you hit that wall, the workaround is to shift from source verification to source mapping. Instead of trying to confirm a single claim, document the complete landscape of who is making claims, what incentives they have, and where their information likely originates. This doesn't give you a verified fact, but it gives you a map that is still useful. I've used this approach successfully in investigations where a government entity was the sole provider of information about its own actions. The information couldn't be independently verified, but mapping the institutional incentives revealed that certain details were consistently omitted in ways that suggested deliberate concealment rather than genuine uncertainty. There's also a bandwidth problem. The full methodology, applied rigorously, can add significant time to any project. A piece that might take a weekend with standard research can easily become a two-week effort using the complete three-phase loop. The tradeoff is real, and you need to decide which stories warrant the investment. Not everything deserves the same level of scrutiny. I usually reserve the full pressure-test phase for claims that, if wrong, would cause measurable harm to people who read or act on them.

Common Pitfalls That Waste Hours

The most expensive mistake I see is treating the collect phase as a search engine exercise. Googling a claim and checking the first few results is not collecting sources. That's browsing. Actual collection requires searching court databases, government registries, academic repositories, and archival systems directly. It requires knowing the difference between a .gov URL that's genuinely official and one that's been set up as a front organization. Another pitfall is confirmation bias masquerading as triangulation. Finding two sources that agree with your hypothesis is not triangulation if both sources share the same origin or incentive structure. I learned this the hard way when working on a story about environmental contamination. Two advocacy groups published nearly identical findings, which looked like strong triangulation until I traced the methodology citations and found both groups had borrowed their sampling protocol from the same university lab, which had received grant funding from one of the advocacy organizations. The data looked solid. The chain of custody did not. The third pitfall is ignoring null results. When your search for evidence turns up nothing, that's data. If you're trying to verify that a person held a specific certification and no licensing board record exists, that absence itself may be meaningful. I've seen researchers discard negative search results because they felt unhelpful, then later discovered that the lack of a record was actually the most important finding in the entire piece.

A Note on Digital Forensics Integration

As investigation methods have evolved, the line between journalistic verification and digital forensics has blurred considerably. The Chasing The Truth framework now incorporates several forensic techniques that weren't part of the original methodology. Hash verification for documents, EXIF metadata analysis for images, and DNS record auditing for website authenticity are all standard practice now. If you're using this methodology in 2025 or later without at least basic familiarity with these tools, you're working with incomplete information. The free toolkit at the site mentioned above has updated sections covering these, and I'd recommend prioritizing those modules if your work involves digital content. The field moves fast. What counted as thorough verification five years ago wouldn't hold up today. Deepfakes, synthesized audio, and AI-generated documents have shifted the baseline expectations for evidence quality. The methodology adapts, but the adaptation requires ongoing attention rather than one-time learning. Set aside time each quarter to review the updated guidance and test your workflow against any new techniques that have emerged. That maintenance investment pays for itself the first time you encounter a piece of evidence that would have fooled your old process.

Jodi Kantor and Megan Twohey's New Book, Chasing the Truth | KCM
Jodi Kantor and Megan Twohey's New Book, Chasing the Truth | KCM

Getting Started Without Overwhelming Yourself

If you're new to this, don't try to apply the full methodology to everything you research. Pick one type of claim you encounter frequently and practice the three phases on that. Start with simple claims where the verification path is short. Once the workflow becomes automatic, expand to more complex material. I worked on straightforward local government records for about six months before I felt comfortable applying pressure-testing to national-level claims. The jump in confidence was noticeable, and the error rate dropped significantly once I stopped treating the methodology as a checklist and started treating it as a habit. The toolkit and reference materials are freely available, so there's no cost barrier to getting started. What you'll need to invest is time and the willingness to occasionally find out you've been wrong. That last part is the one that most people can't handle, and it's also the one that makes the method worth using.