The cost curve, not the calendar, sets the clock
The instinct to treat timing as a date on a calendar is the first error. Entry timing is not a question of when a category will be large; it is a question of when the cost of teaching the market falls below the margin a buyer will pay. Every category carries an education debt — the cumulative spend required to make a buyer understand the problem, trust the solution, and change a workflow. Pioneers do not lose because they are early in time. They lose because they personally finance the entire education debt and then watch a fast follower acquire educated buyers at a fraction of the cost. The relevant variable is who pays to convert ignorance into demand, and whether they can recoup it before the knowledge becomes a public good.
This reframes the famous failures. The first movers in tablets, web video, and social networking were not wrong about the destination; they were wrong about the carrying cost of arriving before the enabling conditions — bandwidth, devices, payment rails — had socialized the expense of adoption. Too early means the education debt exceeds any one firm's balance sheet. Now means an exogenous force is paying down that debt on your behalf.
Reading the signals that separate too early from now
Because the cost curve is invisible, operators substitute the wrong proxies: press attention, competitor funding, their own conviction. The reliable signals are quieter and external. They share one property — they are evidence that someone other than you is now absorbing the cost of changing buyer behavior.
- Adjacent infrastructure crosses a threshold: a platform, regulation, or price point flips a workaround into a default, so you no longer fund the precondition.
- Unprompted workarounds appear: buyers are hacking together your solution from spreadsheets and duct tape, which means the problem is felt and the education debt is already partly paid.
- The vocabulary stabilizes: when buyers name the category without prompting, demand has shifted from latent to articulated, and sales cycles compress.
The absence of all three, paired with high internal conviction, is the signature of too early. Conviction is not a signal; it is the thing the signals are meant to discipline. The most dangerous position is a true belief about the destination held by the only party willing to pay for the journey.
The trade-off is learning versus the right to be copied
Moving first buys proprietary learning — real contact with buyers that no analysis replicates — but it forfeits the right to copy. The follower watches your funnel, inherits your category vocabulary, and skips your dead ends. The decision, then, is not early-versus-late but whether your learning compounds into something a follower cannot lift: a data advantage that improves with use, a supply relationship that locks, switching costs that deepen. Where the learning is durable and accretive, paying down the education debt yourself is an investment. Where the learning is generic and observable, that same spend is a subsidy to whoever enters next with a cleaner balance sheet.
The decision: enter on the second derivative
The practical rule is to commit capital when the rate of cost decline is accelerating but the category is not yet named by the median buyer — the second derivative of adoption cost, not its level. Enter too far before that inflection and you fund the curve for the industry; enter after the vocabulary stabilizes and you pay incumbent-class acquisition costs without incumbent-class assets. The narrow window between those two failures is what now actually means, and it is identified by external evidence of falling education cost rather than by internal certainty about the prize. The entry decision is therefore less an act of vision than an act of measurement: instrument the cost of changing a buyer's mind, watch its slope, and move when the slope turns in your favor and not a quarter before.