A five-page website with one decisive owner takes me thirty hours. The same five-page website with a committee of three takes me ninety hours. Same scope. Three times the cost. I used to quote them the same price.
If you are estimating projects in hours of typing, you are estimating the wrong variable. The actual cost of any project is the friction surface around the work, not the work itself. Friction is decisions, stakeholders, content gathering, integrations, revisions, post-launch hand-holding, and the slow drain of a client who cannot make up their mind. None of that fits cleanly on a timesheet. All of it determines whether the project pays you eighty dollars an hour or fourteen dollars an hour.
I started tracking this seriously about six months ago and the data is brutal. The five projects I lost the most money on were not technically harder than the projects I made the most money on. They were friction-heavier. Every one of them had at least one of the following. Multiple co-owners who disagreed. A middleman agency between me and the actual decision-maker. A “we will send the copy soon” promise that turned into me writing the copy three weeks later. A late-stage scope change after I had already built. A client who needed a phone call before every email. None of those things show up in an hours estimate. All of them roughly double the time the project actually takes.
The friction multipliers I now apply
Here is the rough multiplier table I use when quoting. The base hours estimate assumes a single decisive owner, all assets and copy provided up front, no integrations, and one round of revisions. Everything else multiplies on top of that.
A single owner working alone, base hours times one. This is the dream client and they are rare. They reply within twenty-four hours, they make decisions without consulting anyone, and they trust you to make the small calls without checking in.
Two co-owners who are aligned, base times one point two. The slight overhead is just the second person reading the email and replying, but they do not slow anything down meaningfully. The discount on this multiplier is that you have to verify they are actually aligned before the project starts. Ask them what success looks like in separate conversations. If the answers match, multiplier one point two. If they do not match, you are in the next tier whether they realize it or not.
Two co-owners who are unaligned, base times one point eight. This shows up as the same email getting two contradictory replies four hours apart. It shows up as a revision request from one owner that the other owner never approved. It shows up as a meeting that gets rescheduled three times because one of them cannot commit. The work is the same. The orchestration is brutal.
A committee of three or more, base times two point five. Once you have three decision-makers you are no longer building a website. You are running a Slack channel. Every choice goes through consensus. Every revision requires approval from people who were not involved in the original brief. The client expects you to absorb this overhead because their committee structure is “just how we work.” They will not pay for it unless you priced for it from day one.
An agency middleman between you and the actual decision-maker, base times two. You will write everything twice. Once to the agency, who will translate it for the client. Once to clarify when the client misunderstood the translation. The agency will also have opinions, which is fine when they are right and expensive when they are wrong. Subcontracting through another agency can be profitable, but only if you priced for the translation tax.
A “we have a designer who will provide assets” promise, base times one point six. Sometimes this is true and the designer delivers clean assets on time. Most of the time the designer is overworked, the assets arrive late, the file formats are wrong, and you end up re-cutting them yourself. Price as if you will be doing the asset work. If you are not doing it, you got a bonus. If you are, you are not losing money.
A “we will send you the copy” promise, base times one point four. Almost nobody sends the copy on time. Most clients cannot write copy that converts even if they meant to. You will end up writing it. Build that into the estimate.
The questions that surface friction before you quote
Before I send any quote now I run a short conversation that surfaces friction signals. None of these questions feel like sales questions. They feel like project planning questions. They are both.
Who has final approval on design choices? If the answer is one name, friction is low. If the answer is “we will figure it out together,” friction is high.
How will we handle copy? If they have copy ready, friction is low. If they will send it during the build, friction is medium and probably high. If they want me to write it, friction is medium and at least it is contained.
What happened on your last website project? This is the most diagnostic question I ask. The way someone describes their last project tells you everything about how this one will go. If they say “it took longer than expected and we kept changing things,” that is a tell. If they say “we shipped it in six weeks and have not touched it since,” that is a different tell.
Who else needs to see drafts before approval? If the answer involves anyone outside the immediate owner team, friction goes up. Lawyers, board members, “my husband who has opinions about colors.” All of these are friction multipliers. None of them are bad people. They are just additional surface area.
What is the timeline tied to? If the timeline is tied to a real external event like a product launch or a fiscal year, friction is contained because the deadline is real. If the timeline is “as soon as possible,” friction is high because there is no forcing function and the project will drift.
What this changes about how I quote
I no longer give an hour estimate without first running through the friction questions. I no longer let clients push me to commit to a price before I know the friction surface. I no longer assume the cheapest version of the project, which is what most of us do when we want to win the deal.
The pricing change is straightforward. Take your honest hour estimate. Multiply by your hourly rate. Then multiply by the friction coefficient. That is the quote. If the client pushes back, you have a real conversation about which friction sources they can reduce. Maybe they can collapse the committee to one decision-maker. Maybe they can write the copy themselves. Maybe they can provide assets in a clean format. Each of those reductions earns them a lower number. They are also opting into the discipline that makes the project cheaper in reality, not just on paper.
The clients who will not negotiate friction down are telling you something important. They are telling you that the project will run hot and they will not absorb the cost. Walk away or quote at the high end of the multiplier. Both are correct answers.
The number I track now
I no longer track effective hourly rate by project total divided by hours. That number is a vanity metric because it does not separate friction from skill. I track effective hourly rate by friction-adjusted hours, where the adjustment is whether the project came in under or over my friction estimate. The skill part of my work is roughly constant. The variance is almost entirely in how well I read the friction surface up front. If I quoted for one point two and the project ran at one point eight, that is information about my estimating, not about the work.
After six months of tracking this, my variance has dropped to about fifteen percent. Before, my hourly rate ranged from fourteen dollars to one hundred ten dollars. Now it ranges from sixty-eight dollars to ninety-two dollars. Same work. Same skills. Better estimating of the variable I used to ignore.
If you are quoting in hours, you are pricing the wrong thing. Price the friction. The clients you want will pay for it. The clients you do not want will go to someone else who is undercharging because they have not figured this out yet. That is fine. Let them.