Every Interview Starting from Zero Is Costing You More Than You Think
June 30, 2026 · 7 min read
By Ric Garcia, Co-founder of Mayetik
Your organization has been doing discovery for years.
Customer interviews. User research. Stakeholder sessions. Expert panels. Hundreds — maybe thousands — of conversations, each one containing genuine intelligence about the problems you're solving, the people you're serving, and the world you're operating in.
Now answer honestly: where is all of that?
For many organizations, the answer is somewhere between "scattered" and "gone." A folder of transcripts no one opens. Notes in a former employee's Notion workspace. Summaries that summarized the summaries until the original signal was unrecognizable. Institutional memory that walked out the door with the last person who ran the research.
You've been paying for discovery for years. You've been keeping far less of it than you'd expect.
The Treadmill Nobody Talks About
There's a particular kind of organizational waste that never shows up on a budget line: the cost of relearning what you already know.
A new product manager joins and spends their first three months doing discovery that their predecessor did eighteen months ago. A sales team develops customer intelligence through hard-won experience, and when half the team turns over, that intelligence goes with them. A consulting engagement produces a rich findings report that gets presented, filed, and never referenced again — until the next engagement, when the same questions get asked of the same stakeholders.
Every time a conversation starts from zero, you're not just spending time. You're paying a compounding tax on your failure to capture what came before. In organizations running regular discovery cycles, that tax is rarely visible on any single project — but across a year, teams commonly spend a meaningful portion of their research capacity rediscovering what prior rounds already surfaced. The hours are real. The opportunity cost is real. Neither shows up on a budget line.
The problem isn't that organizations don't learn. It's that they learn and then forget — systematically, structurally, at scale. And then they learn again. And forget again. The treadmill keeps moving. The organization stays in place.
The Asset Nobody Manages
Most organizations have systems for the things they've decided are assets. Documents live in repositories. Transactions live in databases. Code lives in version control. Each of these is managed deliberately — captured, organized, retrievable, and built on over time.
There is one significant organizational asset that receives almost none of this treatment: the structured qualitative intelligence generated through interviews, stakeholder conversations, user research, and expert discovery.
This is not a knowledge management problem in the generic sense. It's a specific structural gap. Structured qualitative intelligence — the accumulated understanding derived from designed conversations over time — is not a document. It's not a database record. It can't be dropped into a repository and retrieved the way a specification or a contract can. It requires a different kind of infrastructure: one that captures conversations in a structured enough form to compare them, synthesizes patterns across them, and accumulates that synthesis into something that grows more useful over time.
Many organizations still lack that infrastructure. So every round of discovery begins fresh, adds to the pile of transcripts and notes that no one will systematically review, and expires — leaving the organization exactly as informed as it was before, minus the time it just spent.
This isn't a new problem. For most of organizational history, it was simply prohibitively expensive to solve at scale. Accumulating structured qualitative intelligence at scale required analytical labor that was prohibitively expensive — synthesizing across dozens of interviews, maintaining consistency across rounds, making findings queryable over time. The bottleneck wasn't capture. It was everything that had to happen after capture. That constraint has changed. The work of synthesis, pattern recognition, and accumulation that once required weeks of analyst time can often happen in a fraction of the time. Which means the gap between organizations that treat structured qualitative intelligence as infrastructure and those that don't is about to widen faster than most people expect.
Why Existing Tools Don't Solve This
This is where organizations typically reach for something familiar. None of them fully solve it.
Meeting recorders and transcript tools primarily focus on capture and retrieval rather than long-term accumulation of organizational intelligence. A perfect transcript of a poorly structured conversation is still a poorly structured conversation — and a searchable folder of those transcripts is not organizational intelligence. It's a very well-organized starting-over problem.
Research repositories and knowledge bases preserve artifacts but don't create synthesis. When someone needs to understand what customers said about a specific problem across twenty conversations last quarter, a repository asks them to do all the analytical work themselves. The information is technically accessible. The structured qualitative intelligence is still locked — or more precisely, it was never created in the first place.
Survey platforms optimize for structured data at scale but sacrifice the qualitative depth that makes discovery valuable. They produce comparable inputs, but inputs of the wrong kind. Closed questions and rating scales can't surface the unexpected thing you weren't asking about — which is often exactly where the insight lives.
AI summarization tools can accelerate synthesis but are limited by unstructured inputs. Applied to inconsistently designed conversations, they return sophisticated summaries of inconsistency. The problem isn't the back end. It's the front end — and no amount of synthesis capability rescues discovery that was never designed to be synthesized.
What's missing is infrastructure that begins upstream: structured question design that makes conversations comparable, individual synthesis that makes each session retrievable, and accumulation logic that makes each round of discovery build on the last.
What It Looks Like in Practice
The following are composite scenarios based on patterns common across discovery-heavy organizations. They are illustrative, not case studies.
A B2B SaaS company had been doing customer discovery consistently for two years — quarterly interview rounds, diligent researchers, good participation rates. When a new Head of Product joined, she asked a simple question: what do we know about why customers churn in the first ninety days? The research team surfaced fourteen relevant interviews from the past eighteen months. But they had been run by four different researchers using four different question structures — some about onboarding, some about value realization, some about competitive alternatives. None could be directly compared. The team synthesized what they could, a process that consumed two researchers for most of a week. Six months later, when a new round of churn research was commissioned, the team largely repeated the same questions, because no one was confident the previous findings were complete enough to build on. The cost wasn't the interviews. The cost was starting over — twice — on a question the organization had already paid to partially answer.
The pattern looks different in consulting but the loss is the same. A strategy firm sends a senior partner to lead a digital transformation engagement for a financial services client. The partner runs eighteen stakeholder interviews over three weeks. The findings are strong — a detailed picture of organizational resistance, capability gaps, and leadership misalignment. The engagement ends. The findings live in a deck. Two years later, a different partner leads a similar engagement for a similar client in the same sector. The firm has run dozens of transformations. Almost none of that accumulated intelligence is usable. The new partner starts where the last one did: from zero. The institutional knowledge exists somewhere across hundreds of old decks and departed partners' memory. It has never compounded into something the firm can actually use. Three weeks of senior partner time. Eighteen interviews. A strong set of findings. And the next engagement is no smarter for it.
Storage Preserves. Synthesis Creates.
This distinction matters more than it might seem.
Most organizations conflate retention with intelligence. They assume that if a conversation is recorded, transcribed, or filed somewhere, the insight has been preserved. It hasn't. The words have been preserved. The insight — the meaning that emerges when you put this conversation next to that one, when you see a theme appear across eight respondents that no single respondent articulated, when you notice that the pattern you saw in Q1 has shifted by Q3 — that insight doesn't exist in any transcript. It exists in synthesis.
Synthesis is not summarization. A summary tells you what happened in one session. Synthesis tells you what's true across many sessions — what patterns hold, what outliers deserve attention, what has evolved over time. Summarization is archival. Synthesis is generative. It creates structured qualitative intelligence that didn't exist before, rather than merely compressing what already did.
This is why no existing tool category consistently solves the problem end-to-end. Repositories store. Recorders capture. Summarizers compress. None of them synthesize across respondents, across rounds, across time — producing intelligence that accumulates rather than merely persists.
Structured Qualitative Intelligence That Doesn't Compound Decays
In investing, compounding is the mechanism that separates wealth from income. It's not how much you earn — it's how much you keep, and what that kept value earns over time. The principal generates returns. Those returns become principal. The asset grows.
The same principle applies to structured qualitative intelligence. The question isn't just what you learned from last quarter's discovery. It's whether last quarter's learning is still working for you today — whether it's informing this quarter's questions, sharpening this round's synthesis, and building toward a body of organizational understanding that gets more valuable the more it accumulates.
For most organizations, it isn't. Because most structured qualitative intelligence isn't compounding — it's decaying. Every week that passes without a system to capture, synthesize, and build on what was learned, the value of that intelligence erodes. Context is lost. People move on. The findings that should have informed this quarter's decisions are vague memories at best.
Non-compounding structured qualitative intelligence has a name: it's called starting over.
None of this is to say that fresh discovery is never warranted. Conditions change. Markets shift. A question worth asking in 2022 may need asking again in 2025. But there is a meaningful difference between deliberately re-examining a question with the benefit of what was previously learned and unknowingly repeating work because no system exists to carry that learning forward. The first is strategic. The second is waste.
And starting over — the unintentional kind — is expensive in ways that are almost impossible to measure precisely — which is exactly why it keeps happening.
What Compounding Looks Like
It starts with structure at the point of capture. Not just recording conversations, but designing them consistently — so that what gets asked in interview ten is comparable to what got asked in interview one, regardless of who ran it. Consistency is the precondition for pattern recognition. Without it, you have anecdotes. With it, you have something synthesizable.
It requires individual synthesis before collective synthesis. Each completed interview should produce a structured brief — a distillation of what that respondent said, in a form that can be compared against every other respondent's brief. That's the unit of accumulation: not the raw transcript, but the structured brief.
It requires cross-respondent synthesis that finds what individual interviews can't. What's true across five respondents that no single respondent said explicitly? What pattern only emerges at interview twelve? What theme was present in Q1 and returned in Q3? That intelligence doesn't exist in any single conversation. It exists in the accumulation — and it only becomes queryable when the accumulation is structured.
And it requires that new discovery builds on old discovery. Discovery that doesn't learn from itself is not a process. It's a recurring expense.
This is what the model looks like when it works: Question. Capture. Synthesize. Compound. Each stage makes the next one more valuable. The structured qualitative intelligence doesn't evaporate — it accumulates.
The Infrastructure That's Missing
Most organizations treat discovery as an activity. Run the interviews. Synthesize the findings. Act on the results. Repeat.
What that model misses is that structured qualitative intelligence, managed as infrastructure, behaves like a completely different asset. Structured capture makes conversations comparable. Individual briefs make each session retrievable. Cross-respondent synthesis makes patterns visible that no single interview could reveal. A queryable knowledge layer means that what was learned last year can inform the decision being made today — without anyone needing to remember it or track down whoever ran the research.
Infrastructure removes the dependency on memory, availability, and luck. It turns discovery from a recurring expense into an appreciating asset — one that gets more valuable the more it accumulates, rather than decaying the moment the debrief ends.
The Competitive Implication
Most organizations are optimized for capturing conversations, not accumulating structured qualitative intelligence. The distinction matters: capturing is an activity, accumulating is a system. And systems compound in ways that activities never can.
This means the playing field in most industries is surprisingly level — not because everyone is good at building organizational intelligence, but because relatively few organizations have built that infrastructure effectively. Most competitors are on the same treadmill: running discovery, losing what they learn, and starting over the next time the question comes up.
That creates an asymmetric opportunity. The organizations that build compounding intelligence infrastructure aren't just doing better research — they are, over time, operating from a different information base than their competitors. Every client engagement, every product discovery round, every stakeholder session adds to a body of structured knowledge that informs the next decision. Competitors doing the same volume of discovery but starting from zero each time are not building toward the same place. They are running harder on a treadmill while the compounding organization is climbing a slope.
The organization that builds the infrastructure first doesn't just get incrementally better. It can become structurally better positioned — with a compounding organizational memory that competitors, still starting from zero, are unlikely to close the gap on simply by doing more interviews.
The Question Worth Asking
How much has your organization spent on discovery in the last three years?
Add up the research hours, the consultant fees, the tool subscriptions, the time of every senior person who sat in on a customer call or a stakeholder interview or a user research session. It's a real number. In most organizations, it's a significant one.
Now ask: what percentage of that investment is still generating value today?
If the answer is uncomfortable, the problem isn't that discovery is wasteful. Discovery is one of the highest-leverage activities an organization can do. The problem is that the structured qualitative intelligence it generates has no infrastructure to live in. It gets created and immediately begins to decay, instead of compounding into something that grows more valuable over time.
The fix isn't doing less discovery. It's building the infrastructure that makes discovery stick.
Stop Starting Over
Structured qualitative intelligence deserves its own infrastructure layer — one designed from the ground up for how conversation-derived knowledge actually works: built at the question, accumulated through synthesis, compounded over time, and queryable when it matters.
That's what Mayetik was built to provide. Not a better place to store interviews, but a system that turns interviews into intelligence — structured capture that makes conversations comparable, individual knowledge briefs that make each session retrievable, synthesis that builds across respondents and rounds rather than resetting after each one, and a queryable knowledge layer that makes what was learned last year available to the decision being made today.
Discovery is one of the highest-leverage activities an organization can do. The organizations that get the most from it aren't the ones running more interviews. They're the ones building the infrastructure that makes each interview compound.
The organizations that learn fastest aren't the ones doing more discovery. They're the ones forgetting less of it.
Mayetik helps teams design better questions, capture structured conversations, and synthesize intelligence that compounds over time. If your organization runs on discovery, we'd love to talk.
Start capturing knowledge today
Mayetik helps teams design better questions, capture structured conversations, and synthesize intelligence that compounds over time.
Next in Part 2
The Art of the Question: Why Asking Well Is a Competitive Skill
Many organizations spend millions making decisions and almost nothing improving the quality of the questions those decisions are based on. That's not a knowledge problem. It's an operational one.