The same problem looks different through different lenses
These are not proprietary frameworks and Dan does not claim to have invented any of them. They are established ways of reasoning — first principles, inversion, root-cause analysis, systems thinking, and the rest — that change which evidence you notice and which conclusions you reach. The value is in knowing which model to reach for in which situation, and in not forcing eight different conclusions just because eight models are on the table. Every claim below is labeled so you can tell what is established, what is a hypothesis, what is an assumption, and what is a mathematical illustration.
Northgate Distribution (fictional example)
Northgate Distribution is a constructed example, not a real company. The facts below are fictional facts established within the scenario — they are not verified facts about any real business, and no model below adds a new fact unless it is labeled a hypothesis or assumption.
- Annual revenue ≈ $24M; a B2B industrial distributor.
- Revenue flat for two consecutive quarters (Q3 and Q4).
- Sales activity — calls, demos, outreach — up roughly 15% over the same two quarters.
- Pipeline appears full: meeting count is up.
- Closed-deal count is not increasing.
- No sales-rep turnover during the period.
- No change to the product offer or published pricing.
- No new competitor entered the core segment.
- Marketing switched ad platforms roughly 60 days before the flat quarter began.
Every model below works from this same fact pattern. A claim labeled Established is established within this scenario — it is not a verified fact about any real company.
First Principles
Strip away assumptionsBreak a problem down to the fundamental truths you can verify, then rebuild upward from there instead of reasoning by analogy. Most 'obvious' answers are inherited assumptions, not facts.
Applied to Northgate: the team is told to 'improve the script.' First-principles thinking asks what we are actually trying to produce — a closed deal (a committed next step), not a call. The script is one input among several: lead quality, timing, offer fit, and the rep's diagnostic skill. Activity rose 15% but closes stayed flat, so the binding constraint is downstream of raw activity. The real question becomes: which input, changed first, moves the most demos toward a close?
When conventional answers feel recycled, when the team keeps solving the same symptom, or when a 'best practice' from another industry is being copied without understanding why it worked there.
Inversion
Ask the oppositeInstead of asking how to succeed, ask how to guarantee failure — then avoid those things. The path to a good outcome is often clearer when you first describe the surest way to destroy it.
Applied to Northgate: what would guarantee flat revenue for two more quarters? Rewrite the script every month so reps never master one version, replace the top performer (the one reliable signal of what works), and ignore lead-source quality while selling harder into a thinner pool. Each of those is a reliable way to keep closes flat. Avoiding those three failures is more tractable than chasing growth directly.
When the goal is ambitious and the path is unclear, when risk management matters as much as upside, or when the team keeps adding effort without removing what's broken.
Five Whys & Root Cause
Look past the symptomAsk 'why' repeatedly until you reach a cause you can actually act on. A surface symptom (revenue is flat) usually sits on top of several layers of contributing causes; treating the symptom moves the problem, not the cause.
Applied to Northgate: Why are closes flat? Demos aren't converting. Why aren't demos converting? Pipeline looks full but may be thinner in intent. Why? Lead quality may have dropped after the ad-platform change ~60 days ago. Why? (The ad platform was switched.) Why would that lower quality? A new platform may serve a different, lower-intent audience — but that is a hypothesis, not yet established. The script is a candidate; the root cause is not yet proven.
When a metric is moving in the wrong direction and the team is debating which lever to pull, or when a fix has been applied and the problem returned.
Systems Thinking
See the whole loopMap the parts of a business as a connected system with feedback loops, delays, and side effects. Optimizing one part in isolation often degrades another. The question is not 'is this part good?' but 'how does this part affect the whole?'
Applied to Northgate: sales, marketing, and customer success are one revenue system, not three departments. A change in marketing's lead source can show up in sales' close rate 30–60 days later and in customer success's churn months after that. The ad-platform change ~60 days ago coincides with the flat quarter — consistent with a lead-quality lag, but the lag is an assumption and the coincidence is not proof. Optimizing the script in isolation would shift the problem, not solve it.
When a change in one department is being evaluated in isolation, when a metric improves but a related metric worsens, or when the same problem recurs in different parts of the business.
Second- & Third-Order Consequences
And then what happens?Every action produces a first-order result (intended), a second-order result (often delayed), and a third-order result (further out). Most planning stops at the first order. Good judgment requires tracing at least two steps beyond the obvious effect.
Applied to Northgate: first order — a discount lifts closes this month. Second order — buyers trained to wait for discounts purchase less at full price next quarter. Third order — the brand is perceived as 'always on sale,' eroding margin across channels. The discount looks profitable at step one and can destroy value by step three — which is exactly why Inversion argues against it before the root cause is known.
When evaluating a pricing change, a hiring decision, a cost cut, or any move with an obvious immediate benefit and a less-obvious delayed cost.
Probability-Weighted Reasoning
Expected value, not best caseList the plausible explanations, estimate each one's probability, and rank them. But the most probable explanation is not automatically the highest-value intervention. Before acting, compare expected benefit, implementation cost, investigation cost, reversibility, and downside. Decide on expected value — not on the single scenario you find most vivid.
Applied to Northgate, three plausible explanations: (A) lead quality dropped after the ad-platform change, (B) the offer no longer fits the new lead segment, (C) a sales-skill gap. Illustrative, unmeasured weights might rank A highest — but those weights are assumptions, not measurements. The highest-value next step is not necessarily 'fix the ad platform' (high cost, hard to reverse). It is the cheap, reversible cohort analysis that distinguishes the three — because knowing which cause is real changes which intervention is worth its cost and risk.
When a decision has a wide range of possible explanations, when the team is anchored to a single dramatic scenario, or when 'hope' is being used as a strategy.
Incentives & Game Theory
People respond to incentivesPredict behavior by mapping what each party is actually rewarded and penalized for — not what they are told to do. Misaligned incentives produce predictable failures regardless of talent, training, or exhortation.
Applied to Northgate: reps are rewarded on closed revenue this quarter; the owner wants durable revenue. If the real cause is a lead-quality change in marketing, the reps cannot fix it — yet they will be measured against the flat number. The predictable response is to discount to save the quarter, which Second-Order reasoning already flagged as value-destroying. The incentive, not the rep, is the mismatch.
When behavior consistently diverges from stated goals, when a policy 'isn't working' despite good people, or when one party's win looks like another party's loss.
Compounding & Geometric Improvement
Small, repeated gainsA small improvement repeated consistently compounds into a large result over time, just as a small loss does in reverse. But compounding is arithmetic on the binding constraint — it does not make every metric grow exponentially, and real systems have floors, ceilings, and slowing rates near the limit.
Applied to Northgate: a 1% weekly improvement in close rate compounds to about 67% over a year (1.01^52 ≈ 1.68) — a relative improvement, not a 1-percentage-point gain. But that compounding only bites if the binding constraint is addressed. If the constraint is lead quality, script practice compounds at ~0% because it doesn't touch the bottleneck. Compounding is real; unconstrained exponential growth is not. Find the binding constraint first — which is exactly what the investigation is for.
When the team is searching for one big move instead of many small ones, when progress feels slow, or when the value of consistency needs to be made visible.
Is the Problem Really Sales?
The same fictional scenario, examined through eight different mental models, produces overlapping diagnoses — and they do not all agree. Click each lens to see what evidence supports or challenges the obvious explanation, with every claim labeled by type.
Revenue has been flat for two quarters at Northgate Distribution. Sales activity is up 15%, the pipeline looks full, but closed deals are not increasing.
It's a sales problem — the team needs better scripts, more training, or a new comp plan.
The real output is a closed deal, not a call. The script is one of several inputs. Activity rose 15% while closes stayed flat, so the binding constraint is downstream of raw activity — the script may not be the lever at all.
Pipeline is full and activity is up, but closes are flat. If the script were the binding constraint, more activity would not have risen without lifting closes. The constraint sits downstream of activity, but which input it is remains an open question.
The point: Eight lenses do not require eight conclusions. Several models point the same way (lead quality), some raise the question without answering it, and others warn against the obvious intervention before the cause is known. Sound judgment is holding several models at once and letting the evidence — not the number of models — decide.
What would a good decision-maker do next?
The lenses converge toward a leading hypothesis — they do not reach a conclusion. The decision below separates what is known from what is still uncertain, names the most plausible explanation without declaring it proven, and picks a low-risk action that is worth taking before the investigation completes.
Revenue is flat for two quarters; sales activity is up ~15%; the pipeline looks full; closed deals are not increasing; no rep turnover; no change to the offer or pricing; no new competitor; and marketing switched ad platforms roughly 60 days before the flat quarter began. That is all established within the scenario.
Whether the ad-platform change actually lowered lead quality is not established — the 60-day coincidence is consistent with that hypothesis but does not prove it. It is also unproven whether the offer no longer fits the new lead segment, or whether a sales-skill gap contributes. We do not yet know which explanation is real, and more than one could be true at once.
Lead-quality decline following the ad-platform change is the leading hypothesis — not a conclusion. It is leading because the timing coincides and the other explanations (no offer change, no rep turnover, no new competitor) have less support within the scenario. A sales-skill gap is the least supported, given no turnover and (by assumption) stable close rates on prior-segment leads. Offer-fit drift is a serious runner-up if the new leads differ from the old.
Pull close rate by lead source and by lead-age cohort for the 60 days before versus after the ad-platform change, holding the script and the reps constant. If pre-change leads close at the old rate and post-change leads close lower, the input changed — supporting lead quality. If close rates are stable across both cohorts, the lead-quality hypothesis weakens and offer-fit or market explanations rise. Separately, read win/loss reasons on lost deals: if 'price' dominates after the change but not before, the offer may no longer fit the new lead segment. Either result could falsify the leading hypothesis.
The low-risk action justified now — before the investigation completes — is to stop changing the sales script and stop adding discounts. Both are supported by Inversion and Second-Order reasoning as likely to compound the problem, and both are cheap and fully reversible. In parallel, run the lead-source cohort analysis (one afternoon, reversible, low cost). Do not yet reallocate ad spend or restructure the comp plan — those are higher-cost, harder-to-reverse moves that should wait for the evidence. That sequencing is the expected-value correction: the cheap reversible investigation outranks the expensive intervention until the cause is known.
Success is the close rate recovering toward the pre-change baseline within 60 days of a lead-source fix, without an increase in discounting. If close rate recovers but discounting rose, the 'fix' traded margin for volume and did not solve the root cause. If the cohort analysis shows stable close rates across cohorts, the lead-quality hypothesis is falsified and the investigation pivots to offer-fit or the market — measured the same way: a defined baseline, a 60-day window, and discounting held flat.
Why this matters: the most probable explanation (lead quality) is not automatically the highest-value intervention. Reallocating ad spend or restructuring the comp plan costs more and is harder to reverse than a one-afternoon cohort analysis. The expected-value correction is to run the cheap, reversible investigation first, and let the evidence — not the illustration — decide which expensive, hard-to-reverse move is worth its cost and downside.
Thinking Is Only Valuable When It Leads to Action.
A mental model that never touches a decision is an academic exercise. Dan's 52-Week Value Creation System is where these models become operational: each week, a specific business area is diagnosed, a measurable action is taken, and the result is checked against a KPI. The models above tell you which lever to pull; the 52-week system tells you how to pull it, measure it, and compound it — without assuming the first hypothesis is the right one.
Diagnosis → Weekly action
Five Whys says the leading hypothesis is lead quality. The 52-week system turns that into: Week 3 — pull close rate by lead source and cohort; Week 4 — read win/loss reasons to test offer-fit; Week 5 — hold the script and discounting flat while the lead-source fix is measured.
Action → Measured result
Each weekly action is tied to a KPI (demo-to-close rate, lead-to-revenue, churn). The compounding model shows why one measured improvement per week, held for a year, beats one heroic initiative that fades by week six — but only when the improvement targets the binding constraint the investigation identified.
Could This Way of Thinking Be Valuable to Your Organization?
The same business problem can look very different through different mental models — and the most probable cause is not always the highest-value next move. If your team is debating which lever to pull, exploring how Dan reasons through problems may be a useful starting point — no predetermined position or contract required.
Explore Working With DanExpressions of interest, not binding agreements or guarantees of worker classification.