On who creates value, who captures it, and who quietly pays for it
Product strategy often talks about value creation as if value were enough. It is not. Every product also determines who captures that value, who performs the unpaid or underpaid work that makes it possible, and who absorbs the externalities the business model leaves outside the transaction.
That is not a late-stage ethics overlay. It is part of the product system itself. If a product depends on invisible labor, hidden harm, or cost being pushed off-platform, then that logic belongs inside strategy from the beginning.
In other words, every product has a value flow whether the team maps it or not. Someone creates value. Someone captures it. Someone else often pays for the neatness of that arrangement.
If a product looks healthy at the company level, teams tend to assume the system is healthy overall. But a dashboard can only tell you what the dashboard was designed to count.
Who is creating value here? Who is capturing it? Who is carrying the cost? I find it useful to name that explicitly as a model: create value, capture value, absorb cost. If those answers point to three different groups, then the product is not only enabling behavior. It is organizing a distribution of gain and loss.
This is easiest to see in platform businesses. Users create attention, content, and data. Creators or workers often provide labor. The company captures revenue. The cost can land somewhere else entirely: moderators exposed to traumatic material, gig workers carrying risk without security, users paying in privacy, mental load, or degraded trust.
Ride-share platforms make this legible quickly. The product looks frictionless to riders because a large share of the uncertainty has been pushed onto drivers. Social platforms can look free because users and moderators absorb the emotional and civic cost. AI products can look magical because data work and content review have been pushed out of frame.
The missing fourth question is usually the one that matters most: who absorbs the cost when the dashboard looks clean? The product can appear elegant precisely because somebody else is holding the mess.
People provide labor, attention, content, trust, and coordination.
Ranking, matching, policy, pricing, and workflow decisions determine where value goes.
Revenue, efficiency, growth, and convenience become visible to the firm and the market.
Because it pulls the conversation away from screens and into incentives, ownership, labor, governance, and extraction. That can feel abstract until trust collapses, workers revolt, regulators intervene, or users start naming the cost that had previously been hidden.
It also feels harder because most available product tools are local tools. You can redesign a screen. You can test a message. It is much harder to redesign the value flow underneath the business. So teams default to the parts they can move quickly, even when the real issue lives deeper in the model.
The page view was never the whole transaction. The roadmap only looked complete because part of the system had been pushed offstage.
That is why this frame often arrives late, in the form of PR crisis, regulation, or backlash. The underlying system was there from day one. The company only started calling it a product problem once the off-platform cost became impossible to ignore.
The smoothest products are often the best at hiding who is holding the friction. Convenience is real. But convenience can also mean that risk, uncertainty, or cleanup has been shifted to someone with less leverage. That is why "frictionless" is not always the same as well-designed. Sometimes it just means the cost has moved somewhere the buyer does not see.
This is one reason value-flow analysis belongs near the center of strategy. The neatest user experience can be built on the messiest underlying arrangement.
It shows up when a marketplace looks efficient because worker precarity is not counted as product friction. It shows up when a social platform looks healthy because the emotional cost of participation is being absorbed by users rather than reflected in the KPI. It shows up when AI products look magical because the human labor cleaning data or reviewing harmful content has been made invisible.
Amazon Mechanical Turk makes the pattern hard to ignore. The product looks like seamless machine intelligence from the outside, but much of the system is held together by hidden human labor paid at extremely low rates. One widely cited analysis found a median hourly wage of about $1.77, and the broader world of ghost work exists precisely because the interface feels automated while the labor remains out of frame.
The common pattern is simple: the company captures the upside while someone else carries the instability. That arrangement can look sustainable for a while, especially when the hidden cost sits outside the main dashboard. It rarely stays hidden forever.
The strategic mistake is thinking this belongs to ethics, policy, or comms after the fact. In practice, it shapes growth, trust, retention, defensibility, and regulation exposure. It belongs much closer to the center of product thinking than most teams currently place it.
Teams can start asking better strategic questions. Whose labor is currently invisible? Which costs are being treated as someone else's problem? Would the product still work if those costs had to be held inside the model rather than pushed outward?
Another useful question is whether the product would still look healthy if the people creating the value had more power to refuse the terms. If the answer is no, then the strategy is more fragile than it appears.
A simple exercise helps here. Take any major initiative and force three columns onto the page: who creates the value, who captures the value, who absorbs the cost. If the third column is blank, it usually means the cost has not been named yet, not that it does not exist.
This is not an argument for moral purity. It is an argument for clarity. Systems that rely on hidden extraction rarely stay stable forever, and product strategy gets stronger when it names the full transaction instead of only the part the company captures.
I think of this as the structural companion to the other Sociology x Product essays. Why Useful Products Still Fail focuses on social risk at the point of adoption. The Product Funnel Starts Too Late focuses on what standard dashboards miss. This essay asks a different question: even if the product is growing, who is the system actually working for?
Whenever a roadmap item looks obviously good, it is worth adding four questions to the review. Who creates the value? Who captures the value? Who absorbs the cost? Who has the least power to refuse the arrangement? That last question is often where the product's real fragility lives.