Worked pre-mortem (Titan). "It is three years later; Titan was a disaster. What happened?"
| Disaster path (the story) | Prob × severity | Early-warning trigger to monitor |
|---|---|---|
| Lab-grown diamonds commoditized the studded segment; the margin engine compressed structurally, group EBIT fell toward ~10%, and the premium multiple de-rated hard, a double hit to earnings and multiple | High severity, medium prob; the thesis-breaker | Studded ratio + studded segment margin, tracked quarterly over 2yr; natural-vs-LGD price spread; Tanishq diamond same-store trend |
| Competition (Reliance/Kalyan) forced sustained discounting; Titan held share but only by giving up margin; ROCE drifted down | Medium × medium | Group EBIT margin trend; Titan's share of organized jewellery; competitor store-add pace and promotional intensity |
| A gold-price shock + a slowdown created a multi-quarter demand air-pocket in big-ticket jewellery, and a hedging lapse turned it into an inventory/margin hit | Medium × medium (cyclical, survivable) | Same-store & buyer-count growth through gold spikes; hedging-disclosure quality; inventory days |
| Governance drift (brand-fee creep, a value-destructive acquisition, or a TIDCO-driven payout distortion) quietly transferred value away from minorities | Low × high (the veto path) | RPT/brand-fee % of sales trend; acquisition pricing & fairness opinions; payout vs FCF and vs runway |
| Reinvestment runway exhausted: the base got so large that new stores and studded penetration stopped earning the hurdle, and retained cash piled up at low returns (the M3.07 §4.6 Trap-1 fate) | Low-medium × medium | Incremental ROIC on 3yr windows; store-productivity (revenue/sq ft) of new vintages; rising idle cash |
Advertisement
What do you do with the triggers?
The payoff is the right-hand column. Those five triggers are the monitoring dashboard for your permanent Titan file, and they map one-to-one onto the sell criteria in the one-page thesis. You have converted "there are risks" into "here are the five specific numbers I will watch, the thresholds that mean the thesis is breaking, and what I will do when each trips." That is the difference between an investor who is surprised by a disaster and one who saw the trigger months before and had already decided what to do. The pre-mortem is also, quietly, the last confirmation-bias corrective of the day: it is very hard to stay in love with a thesis right after you have written, in detail, the five ways it kills you.
How often should you re-run a pre-mortem?
A note on cadence (Phase 9 will formalize it): the dashboard is not a one-time exercise. Re-run the triggers on every annual report, every quarter's concall, and any thesis-relevant event, and update the company file. The deep dive produced the thesis; the dashboard is how you own it responsibly over the years that decide whether you were right.