How Long Does a Web Design Project Actually Take (And What Makes It Run Late)?
Every web design conversation I have ever had reaches the same question about four minutes in. Not what it costs. How long it takes.
It is the right question, and it usually gets a bad answer. The bad answer is "four to six weeks," delivered confidently, which is technically the design time and has almost nothing to do with when the site goes live. Then week nine arrives, the site is not up, and everybody is annoyed at each other over something that was predictable in week one.
So here is the real version, including the parts that are your job rather than mine.
What is the honest timeline for a small business website?
For a typical Western PA service business, meaning eight to fifteen pages, a contact form, a gallery, and no online store: six to ten weeks from signed to live.
Roughly how that splits:
Week 1: discovery. A long conversation about the business, not the website. What you sell, what is profitable, what you want more of, which jobs you want fewer of, who actually calls you and what they ask first. This is also where we look at what you have now and what it currently ranks for, because throwing out pages that quietly bring in customers is the most expensive mistake in a redesign.
Weeks 2 to 3: structure and design. What pages exist, what each one is for, what happens on a phone. You see real screens, not a mood board.
Weeks 3 to 6: content and build. These overlap, and this is where projects live or die. Writing, photos, the actual building.
Weeks 6 to 8: review and revisions. You go through it, we fix things, you go through it again.
Week 8 to 10: launch and the week after. Domain and hosting, redirects from old pages, forms tested from a real phone, analytics, Google Business Profile pointed at the right place. Then a week of watching for anything odd.
If somebody quotes you two weeks, they are either using a template with your logo dropped in, or they are quoting the build alone and the other six weeks will still happen, just uncounted and unmanaged. If somebody quotes six months for a fifteen-page site, they have a queue problem and you are in it.
What actually makes a web design project run late?
In my experience it is almost never the design work. It is three things, in this order, and two of them are on the client's side of the table.
Photos. This is number one and it is not close. A site is designed around real pictures of your work, your building, your crew, your trucks. Then the photos do not come, because taking them means somebody has to remember a phone at a job site for two weeks. The project sits. Everybody says "we are waiting on the designer." Nobody is waiting on the designer.
Content approval. Not writing the content, approving it. We draft it. Somebody at the business has to read it and say yes, and that person is usually the owner, and the owner is out on jobs until seven. Copy sits in an inbox for eleven days and that is eleven days of the project.
Decision by committee with no decider. Three family members in the business, three opinions on the homepage, no one authorised to end the argument. This turns a two-day revision round into a three-week one. Name one person who decides before the project starts. It matters more than any design choice you will make.
The pattern behind all three: the parts of a website project that need the business owner are the parts that slip, because the owner has a business to run. Which is not a criticism, it is just something to plan around instead of pretending it will not happen.
How do you actually make it go faster?
Front-load your side of it. If you do these four things in week one, you will land at six weeks instead of ten, and it is genuinely that simple.
Take the photos before you sign anything. Twenty to thirty pictures. Finished work, work in progress, your truck, your crew, your building, you. A recent phone is fine, better than fine. This single item is worth two weeks of schedule.
Write down your ten most common customer questions. The ones you answer on the phone every week. Those questions are your site's content, and they double as the questions Google and ChatGPT pull answers from when somebody searches. Nobody knows them but you.
Find your logins now. Domain registrar, current hosting, Google Business Profile, whatever email is attached to them. Half of all launch delays in this area are a domain registered in 2011 by a nephew who has since moved to Arizona. Finding that out in week one is a phone call. Finding it out on launch day is a two-week transfer.
Name the decider. One person. Everybody else advises.
Do those four and the project moves at the speed of the build, which is the only part that is actually mine.
Why does the cheap fast option usually take longer in the end?
Because "live" and "working" are different dates, and the cheap version only optimises the first one.
A template site with your logo on it can be live in a week. Genuinely. It will also, in a fair number of cases, be rebuilt within eighteen months, because it was never structured for what the business actually needed. Add the original build, the year and a half of it underperforming, and the rebuild, and the fast option took considerably longer than doing it once.
The other version of this is a redesign done without checking what the old site ranked for. Pages get renamed, old addresses are not redirected, and traffic that took four years to earn goes to zero on launch day. That recovery, when it is possible at all, takes months. We wrote the full decision framework on that in redesign or rebuild, and it is worth reading before you agree to anything, including with us.
There is a real question of what a website is for underneath all this. A site that looks good and a site that brings in work are not automatically the same object, and we covered what a website designer actually does at length. Our web design page shows the process, and the finished ones are on our work page.
When should you start if you want the site working by your busy season?
Count backwards, and count generously, because a website does not start working the day it launches.
Take the six to ten weeks to build. Then add another eight to twelve weeks before search results really settle, since a new site has to be crawled, indexed, and trusted, and pages that changed addresses take a while to sort themselves out.
So from signing to actually pulling its weight in search: four to five months.
Which means, practically, for a Western PA business:
- Want it working for spring (March, when the phone starts ringing for exterior work)? Start in October or November.
- Want it working for fall and the pre-winter rush? Start in May.
- Want it working for the Christmas retail season? You needed to start in August, so start now.
Seasonal businesses feel this hardest, and there is a whole planning rhythm to it that we laid out for one trade in particular in the seasonal SEO playbook for Western PA landscapers. The same shape applies to roofing, HVAC, paving, and pools.
The mistake I watch owners make every year is starting the website in the first week of their busy season, when they finally feel the pain of not having one. That is the exact eight weeks they have no time to take photos or approve copy. The project drags to January and they lose the season they were trying to save.
What does the timeline have to do with the SEO side?
They are separate clocks and it is worth knowing that going in.
The website build is the six to ten weeks above. The search work runs on its own schedule alongside it, and for the local search piece we put a number on it:
We get your business into the top 3 of the Google Maps pack for your core service queries within 90 days, or you stop paying the monthly fee until we do. We keep working. You keep your number.
Ninety days from when the search work starts, which for most clients is at launch, not at signing. So it is not a promise that a site signed today is in the map pack by Thanksgiving. It is a promise that once we start, the clock is on us rather than on you.
On price, so you can plan around it: Foundation is $499 a month with a $500 setup, Growth is $998 a month with a $1,000 setup, and Dominate is $1,497 a month with a $1,250 setup. The setup covers the website build described in this post. Paying annually cuts the setup in half and gives you two months free.
If you have a season you are aiming at, tell me which month it is and I will tell you whether there is time, including if the honest answer is that you should wait and start after it instead of rushing something out in the middle of it. Andrew Hersh, andrew@padigitalstudio.com, 724-638-7754.