The problem
When traction stalls, founders often tell themselves a comforting lie: we just need one more feature.
It sounds rational. It keeps the team busy. It avoids the more painful possibility that the real issue is not the product at all — it is that nobody has enough clarity about the buyer, the timing, the pricing, or the channel.
That is the lucidity gap: the distance between what your team says it knows and what the market has actually confirmed.
Why it fails
Lucidity loses because shipping is easier to organize than truth.
Tickets, sprints, and releases give the team visible progress. Buyer clarity, pricing confidence, and ICP discipline are messier. They require uncomfortable conversations, explicit trade-offs, and sometimes admitting that months of roadmap work were pointed at the wrong target.
So companies default to backlog therapy: add features, polish UX, ship faster — and quietly avoid the strategic question underneath.
A concrete method
Method: lucidity contract per initiative
Adopt a simple rule: no major initiative without a lucidity line. A line contains: market belief; signal that would strengthen it; signal that would destroy it; deadline; owner. If you cannot fill those five cells, you do not have a product spec—you have an aspiration.
Lucidity reviews (45–60 minutes, weekly or biweekly): walk active initiatives. For each, ask “What do we know now that we did not fifteen days ago?” If the answer is empty, velocity is likely noise. Then: “Which buyer decision does this release enable?” Not “make life easier”—decision: sign, renew, expand, internal referral.
ICP matrix: list candidate segments with three columns: observed pain (verbatim), purchasing power (evidence), urgency (trigger event). Pick one primary segment for the next four weeks. Others go to a cold queue with honest messaging. That choice is politically hard; that is where lucidity differs from pessimism.
Price and value: separate “price asked” and “price accepted.” Run a micro-experiment: minimal commitment offer, or two packages to see where silence appears. Log objections verbatim.
Channel: one channel hypothesis per sprint—not twelve tactics at once. Criterion: reach N qualified conversations with a stable script.
Belief / evidence board: columns date, belief, evidence for, evidence against, next decision. Shared with sales and product. When a belief survives three cycles without new evidence, it is demoted to named risk—no more “obvious” status.
This frame does not replace vision; it grounds vision in falsifiable market touchpoints. It becomes acceptable to say: “We stop this feature because our lucidity line says the wrong segment—not because the team is slow.”
Example
Example: six months of “customer asks”
A SaaS hears asks from three different verticals. Each AE reports a “critical need.” Product ships a string of releases stacking checkboxes: exports, roles, partial integrations, reports. Demos lengthen; complexity rises. Yet pilot conversion flatlines and engineering morale dips.
A lucidity review shows the three verticals do not share buyer, substitute, or cycle. The “missing feature” was three different stories compressed into one backlog. The lucid decision: freeze a release, remove two verticals from outbound talk for six weeks, and concentrate interviews on one decision chain. Outcome: fewer features shipped, but higher pilot-to-paid conversion in that segment—because message and product finally cohere on one problem.
The counterpoint: a team refuses the cut “not to upset” lukewarm leads. Multi-segment drags on. The next six months repeat the same plateau—proof lucidity is not a communication exercise; it is an informed choice to renounce. Renouncing hurts; refusing to renounce costs more.
The example also shows verbatim power. The decision shifted when the team pinned ten contradictory client quotes on a wall and had to choose which represented the primary ICP. While insights live only in heads, the product stays soup. Lucidity is often collaborative: sales brings real phrases, product translates to scope, leadership holds the segment line. Without that triad, “product problem” stays a ghost.
What to do now
Freeze one feature that lacks a written market belief behind it.
Then answer, in plain language:
- who is the buyer,
- why now,
- what they pay today,
- what would make them switch,
- and what signal would prove you are wrong.
If that exercise feels harder than writing the feature spec, you found the real problem.
Related reading
Lumor helps teams replace backlog fog with explicit pressure: 13 AI roles challenge the buyer, the economics, the timing, and the roadmap until the real problem becomes visible.