Pricing is the hardest part of running a web design agency. Not designing, not delivering, not closing — pricing. Price too low and you can't scale out of the grind. Price too high and you lose pitches to cheaper competitors who promise the same deliverable. Most agencies spend years oscillating between the two extremes before they find a model that works.
This post is about that model. It's not a magic number, and it's not a calculator. It's a framework for deciding how much a specific project should cost this specific client — based on what the work is worth to them, not what it costs you to do. Once you internalize it, quoting stops being stressful.
Why hourly pricing caps your income
Hourly pricing locks your income to hours worked. That's not a philosophy problem — it's a math problem. If your rate is $100/hour and you bill 25 productive hours a week (realistic for a solo agency), you top out at $125K/year. You can raise your rate, sure. But every dollar over $125/hour requires an increasingly hard pitch, because clients benchmark against "the freelancers I saw on Upwork for $40/hour."
Hourly also punishes efficiency. The better you get, the less you bill. A designer who rebuilds a site in 20 hours because she's done it 50 times makes less than a slower designer who takes 40 hours. That's a direct disincentive to improve, which is insane.
And clients hate it too. "It'll be about $4,500, depending on hours" is a quote that both of you will regret. They want a number. You want to deliver value. Hourly creates a conflict that flat pricing resolves.
Why flat project pricing still underwater-prices you
Most agencies who leave hourly go to flat project pricing: "$5,000 for a full rebuild." This is better, but usually leaves money on the table.
The problem is that "full rebuild" means something very different to a 2-person startup and a 40-person regional firm. Both can benefit from the same service, but a new site for the startup is worth maybe $10-20K in lifetime value (better positioning for a few early sales). The same site for the regional firm might drive $250K+ in new closed deals annually.
If you quote both at $5K, you're either overcharging the startup (losing the deal) or undercharging the firm (leaving $15-20K on the table). Flat pricing can't distinguish.
The client-revenue-based framework
The fix is to price based on how much value the work is worth to that specific client, not how much work it takes. You estimate the client's annual new-client value and charge a fraction of that.
The formula, roughly:
Project price = 30-70% of the client's annual new-client value from improved web presence
You don't need to be exact. You need to be in the right order of magnitude. Here's how to estimate the annual value:
- What does one new client pay them? Ask on the discovery call: "What's your typical engagement size for a new customer?" Get an average.
- How many new clients could the website reasonably drive in year one? Usually 1-10 for a small service business with a better site. Be conservative — estimate 3.
- Multiply. Customer value × new clients = annual incremental revenue from your work.
- Price at 30-70% of that figure. 30% when the prospect is price-sensitive; 70% when they're sold on the ROI and you want to capture value.
Worked example
A local law firm bills $3,000 average per new case. They close about 20 new cases a year, almost all via referral. After your rebuild (including SEO basics + clear conversion paths), it's realistic to estimate they close 5 additional cases per year from web leads.
- Annual incremental revenue: $3,000 × 5 = $15,000
- Project range: 30% = $4,500. 70% = $10,500.
Anchor your proposal at $10,500. Be willing to negotiate to $7,500. Refuse below $6,000.
Compare that to "flat project pricing" of $5,000. You just doubled the ceiling for the same work — and the client still nets $4,500/year in year one alone, and recurring thereafter.
Three pricing tiers in one proposal
Always present three options. The reason is the "anchoring" effect in negotiation: when people see three options, they overwhelmingly pick the middle one. This is called the compromise effect, and it's one of the most reliable behavioral findings in consumer research.
A good three-tier structure for a web project:
- Essential (60% of middle price): just the rebuild, same sitemap, basic SEO, no new content or integrations.
- Recommended (the anchor you want them to pick): full rebuild + content strategy + 3 months post-launch support + analytics setup.
- Growth (140% of middle price): everything in Recommended + SEO retainer for 6 months + monthly reporting.
Set the Recommended tier at the 70% figure from the client-revenue formula. Set Essential at 60% of that. Set Growth at 140% of that.
For the law firm example: $6.3K / $10.5K / $14.7K. Most clients pick the middle. Some pick Growth. Essential exists to give the price-sensitive client a way to say yes without negotiating on the Recommended tier, which protects your margin.
When to quote on a call vs in a proposal
The decision is about buying signal, not about you.
Quote on the call if:
- They've described the project clearly
- They've given you a budget range or at least confirmed a rebuild is in scope
- They're asking "how much" — not "send me a proposal and I'll think about it"
Send a proposal if:
- Multiple decision-makers need to see it
- The scope is ambiguous and needs to be documented
- They're comparing quotes (in which case your proposal needs to differentiate on substance, not just price)
When in doubt, quote a range on the call ("most of our rebuilds come in between $7-12K depending on complexity") and follow up with a detailed proposal. Never commit to a number without having thought through scope.
Discount tactics that protect margin
"Can you do anything on price?" is the most common negotiation question. How you answer shapes the rest of the project.
Bad: drop the number. "$10,500 — actually, I can do $8,500." You've just told them the original number was fake. Everything from here is a negotiation, not a decision.
Better: trade scope for discount. "I can meet $8,500 if we remove [specific scope item]." Now the discount is a real tradeoff, not a concession. They get to decide if the savings is worth the loss.
Best: add a constraint. "I can meet $8,500 if we lock the delivery window and you sign by Friday." Now they're trading on time, not scope. You get the deal and the project starts when you want it to.
Nuclear: discount with an exchange. "$8,500 + a video testimonial at project end + permission to list you in the portfolio + a referral introduction." This is how portfolio-stage agencies survive discounting — they extract future value to offset present loss.
If none of these land and the prospect still demands the original discount, walk. A client who negotiated your fee down is a client who will negotiate every request, every revision, every deliverable from here on. The project stops being profitable in week three.
When to walk on price
There's a floor below which you should never quote, no matter how "in demand" you are for cash flow. It's the price below which the project can't land well — the client hates the result, you hate the process, both of you wonder why you got involved.
For most agencies that floor is around $3-4K for a real rebuild. Below that you're doing a template implementation, which is a different business entirely. Don't pretend a $1,500 rebuild is a rebuild. Price it as what it is, take the job on those terms, or refuse.
Walking away from underpriced work is a skill. It feels terrible the first few times. It gets easier once you realize every underpriced project you take on costs you a properly-priced project you don't have bandwidth to pursue.
Related reading
- The 6-section proposal that closes 30%+ of cold leads — the proposal structure built around three-tier pricing
- Handling "it's too expensive" on a discovery call — deep dive on the #1 price pushback
- How to qualify a web design lead in 60 seconds — filter for prospects whose project is worth pricing in the first place