What does a real product strategy look like for an early-stage company?

Two cases in full. A team with solid discovery, a promising solution and mixed signals that never decided and died of the indecision. And the hospitality group's ladder, row by row: profile alone first, one restaurant reads it, check-in before payments, tiers on visits and not tokens, every row cited to the field. Then two public examples of bets a competitor would have made the other way.

What happens when a startup fails to decide between iterating and pivoting?

Two cases in full. A team with solid discovery, a promising solution and mixed signals that never decided and died of the indecision. And the hospitality group's ladder, row by row: profile alone first, one restaurant reads it, check-in before payments, tiers on visits and not tokens, every row cited to the field. Then two public examples of bets a competitor would have made the other way.

How do you go from discovery findings to a strategy and MVP?

Two cases in full. A team with solid discovery, a promising solution and mixed signals that never decided and died of the indecision. And the hospitality group's ladder, row by row: profile alone first, one restaurant reads it, check-in before payments, tiers on visits and not tokens, every row cited to the field. Then two public examples of bets a competitor would have made the other way.

Cases · Product Strategy · Product Thinking

§Two cases in full, and two in brief. The first is the failure this whitepaper most wants you to avoid. The second is working.

§ 14.1The team that stayed in limbo#

§A team in an accelerator, building for HR departments. Their discovery was solid: real conversations, a clear , a prototype that landed. Their solution was promising. And the signals were mixed in a way the framework makes visible: problem validation green, solution fit green, market timing yellow, team capability yellow with a missing sales skill, resource match red, because the first version needed roughly twice their budget and the timeline exceeded their runway.

§That grid has an answer. It says: to a first version that fits the resources, or pivot the segment to one that pays sooner, or raise against the green rows before the runway runs out. Any of the three is a move. The team chose none. They discussed. They refined the prototype. They took more meetings with the same prospects. Each month the grid read the same, and each month deciding felt slightly less urgent than the month before, because nothing had visibly changed.

§The company died of the indecision. Not of the , which was good, and not of the market, which was there. Of a red row that everyone could see and nobody put a date on.

§ 14.2The hospitality group's ladder#

§A multi-brand hospitality group with more than twenty brands, a lead named as decider, and a written position from the sprint: profile and reservations validated, loyalty unit and check-in gesture discarded, six open decisions with owners. The strategy work took the position and built the ladder, row by row.

§Mission. Written in silence, picked by the decider: every guest recognized across our brands without being asked for anything. The last clause cited six of six customers filling in profiles gladly and unpaid, and later forbade the point economy.

§Vision. From three seats, two years out. The guest is recognized on arrival at any brand without asking. The server knows a regular is coming before she sits and what she usually orders. Leadership sees, in one place, who its cross-brand regulars are. Payments appeared nowhere. That absence was the first correction to the team's original ranking, and it came from the vision row before the row was reached.

§Strategy. We will: the customer builds a profile alone; the operation reads it in one restaurant; with information flowing, check-in enters, and payments on top of it; tiers on real visits, not tokens. We will not: pay customers for their data; launch payments before check-in works; ask the operation to type during peak hours. Each line cited: the six-of-six profile completions, five of six operations staff naming the card, the six-of-six check-in failure, two operations interviewees who would not touch a screen in a quiet room.

§Goals. One per . Profile completion above sixty percent, unprompted, by day ninety. Servers in the pilot opening a regular's card before seating on half of shifts in month two. Unaided check-in at five of six in the next field test before it enters. Payments' goal deferred until check-in's holds.

§Roadmap. One month: profile and tier. Three: one restaurant reads, read-only. Six: check-in from the reservation, promoted only if the pilot's servers used the card. Twelve: payments, promoted only if check-in held. Every item as "can we do X to move Y," every jump with its criterion.

§Task. The guest builds a profile and sees her tier, on her own phone, from a reservation link or table card, one brand family, four weeks. Out: the economy, the check-in, the payments, the operations screen, each with its row of evidence.

§The room had entered the with payments as pain number one and left the strategy work with payments in month twelve. The product lead's sentence at the close was the tie-breaker and the summary: loyalty over payments to test the real value proposition, then payments, essential for the business case but not necessarily for the customer.

§ 14.3Two bets a competitor made the other way#

§Cursor forked the editor rather than extending it. The category's default was the extension, because it inherited the dominant editor's distribution. Cursor's founders have said the extension surface limited how much of the experience they could change, and their vision required changing most of it. "We will build the editor; we will not accept the extension's constraints." Several competitors chose the opposite, reasonably. That is a strategy: a choice with a cost, made from a vision, that someone else declined.

§Linear wrote its method down and made it public. The Linear Method is, read as a ladder, a strategy row with an unusually long "we will not": no endless configurability, no building for every team's process, an opinionated tool for the way its founders believed software teams should work. The incumbent's strategy was the mirror image, configurability for everyone. Both are strategies. Only one of them could have been written by a five-person team, because only one of them excluded enough to be buildable in a year.

§The HR team had evidence and no decision. The hospitality group had evidence and a ladder that turned it into a decision the room had not wanted. Cursor and Linear made bets that a competitor rejected, and wrote them where people could see. In every case that worked, something was left out on purpose and the reason was recorded. In the one that failed, nothing was.

§ 14.5What comes out#

§Your own ladder, filled and cited, with a date on every open row and a decision document behind it. Read it against the HR team's grid. If any row has been the same color for a quarter, that row has a date now.

14. Cases