Home / When Architect Referrals Dry Up
Almost every established practice was built on referral. It is the highest quality source of work there is: pre-qualified, price-tolerant and warm. It is also completely outside your control, and that is the problem nobody addresses until it becomes urgent.
This page is about what actually works when you decide to build a second channel, written for practice owners rather than for marketers.
| Source | What it is genuinely good at | The catch |
|---|---|---|
| Referral from past clients | Highest trust, least price resistance, shortest sales cycle | Entirely outside your control and it tracks your last few years of project types |
| Repeat developers and builders | Predictable volume, fewer proposals | Concentration risk. One relationship ending can remove a large share of revenue |
| Directories such as Houzz | Works immediately, needs no content | Rented demand. Stops the day you stop paying, and you compete on the platform's terms |
| Awards, press and peer recognition | Real credibility with peers and juries | Rarely how a homeowner finds you. It supports a decision rather than starting one |
| Instagram and Pinterest | Excellent for aspiration and for staying visible to past contacts | Poor at capturing someone actively deciding. Discovery, not intent |
| Local search | Reaches people at the moment of intent, and you own it permanently | Slow to build, and most firms publish the wrong content for it |
Referrals lag your portfolio. They bring you more of what you have already done, for people who resemble your past clients. That is fine while your market is stable. It becomes a problem in three specific situations, and every practice hits at least one of them eventually.
By the time somebody searches "architect near me", they have usually already decided to build, already spoken to two firms, and are comparing. That search is late and everyone is fighting for it.
The search that happens first looks nothing like it: "do I need a permit to add a second storey", "can I build an ADU on my lot", "how close to the property line can I build", "how long does entitlement take". Those happen six to twelve months earlier, while the homeowner is still deciding whether the project is possible at all.
Now the part most agencies selling this idea will not tell you: the generic version of it is already lost. Search "do I need a permit for a home addition" and you are competing with your own city's planning department, with Angi and HomeAdvisor, with every ADU builder in your metro, and in California with Symbium, which returns an instant address-level feasibility answer with the zoning already encoded. A blog post does not beat a parcel-level answer.
What actually works is narrower and it is the version individual firms already prove out rather than the version agencies sell. It is jurisdiction-specific and it is tied to an offer: a real feasibility assessment for a specific address, in the municipalities you actually practise in, that produces a written answer the homeowner cannot get anywhere else. Firms doing this sell it as a paid feasibility study and it becomes the front end of the design commission.
The genuinely open ground is that no agency serving architects is positioned on this at all. They sell brand, website, social and Houzz. That is the gap, and it is a gap in positioning rather than in keywords.
Not philosophy. Not another portfolio entry. These, in roughly this order:
You can do all of the above yourself, and if you have someone in the office who can write, you probably should. The reason firms hire someone is not that the work is difficult, it is that it never becomes urgent enough to beat billable work, so it does not happen.
If you want it done rather than planned, that is what I do. Architecture firm marketing explains the approach, and you can book a 30 minute call. No contract, and everything I build stays yours.
It should, and that is a reasonable reaction. The people who own this category have trained you to expect a pitch with no substance. The difference I can offer is checkable: before any call I look at your site and your search visibility and tell you what I would fix first, whether or not you hire me. If that conversation is not specific to your business, you will know inside five minutes and you should end it.
I am based in India and I work with clients in the United States, the UK and Israel. Saying that plainly is better than letting you find it out later. What it means in practice: my afternoon and evening covers the whole UK working day and the US morning, so you get same-day replies rather than overnight ones. What it does not mean is a team of juniors. You get me, which is also the limit, because I take a small number of clients at once.
This is the most common thing I hear and usually the story is the same: a long contract, a monthly report full of numbers that went up while the phone did not ring, and no way to leave. So there is no contract here. You can stop at the end of any month, and the site, the content, the tracking and the ad accounts are yours from day one, in your name, not mine.
Because traffic is not what gets reported. The events that matter are phone taps, form submits and booking clicks, tracked from the first week so that by month two we can both see which pages produce enquiries and which produce visits. If a page brings people and no enquiries, that is a finding, not something to bury under a sessions chart.
You should not have to. What I need from you is roughly an hour a month, plus answers when something needs a decision only you can make. If a project needs more of your time than that, I will say so before it starts.
Written out because the fear with any agency is not the price, it is not knowing what happens after you pay.
| When | What happens | What you have to do |
|---|---|---|
| Before the call | I look at your site, your Search Console if you will share it, and the two or three competitors who actually take work from you. | Nothing. |
| The call, 30 minutes | I tell you what I would fix first and roughly what it is worth. If search is the wrong spend for your situation, I say so here. | Answer questions about how enquiries reach you now. |
| Week 1 | Conversion tracking goes in first, so phone taps, form submits and booking clicks are measured from the start. Full technical crawl. You get a written list of what is broken, ordered by what it is costing you. | Access to the site and the analytics. |
| Weeks 2 to 3 | The fixes that need nothing from you: the technical faults, the pages that rank but do not convert, the enquiry path itself. | About an hour, reviewing anything client-facing. |
| Week 4 | First written update. What was done, what moved, what did not, and what is next. Not a dashboard. | Read it and tell me if the priorities are wrong. |
| Month 2 onward | The slower work: content, local visibility, links. Monthly written update in the same format. | An hour a month. |
Who you deal with: me, every time. There is no account manager and no handover to a junior, which is the advantage and also the reason I keep the client list short.
What happens if it is not working: you leave at the end of any month and keep everything. I would rather you did that in month three than stay unhappy for a year, because the referrals I get from people who left on good terms are worth more than one uncomfortable retainer.
I publish this because almost nobody in this field does. Search for marketing help for an architecture practice and you will find a dozen firms asking you to book a call before they will tell you anything. That is not confidentiality, it is a qualification filter, and you are the one paying for it in time.
Where you land above that depends on the job. A practice wanting two more projects a year and one rebuilding its pipeline after losing a key developer need very different work, so the figure is scoped to that rather than to a package.
What is fixed regardless of scope: it is a monthly retainer starting at $800, there is no contract, you can cancel any month, and everything I build stays yours including the site, the content and the ad accounts. If we run ads, that spend goes to Google in your own account and I do not mark it up.
These are not packages and there is no tier you get upsold into. It is simply what the work looks like at different sizes of practice, so you can see where you probably sit before we speak.
| If your practice looks like this | The work is mostly |
|---|---|
| One or two people, one municipality, residential | Google Business Profile, the enquiry path, and permit and feasibility answers for your single jurisdiction. This is where $800 a month sits. |
| Established practice, several municipalities, mixed residential | The above across each jurisdiction you actually work in, plus project pages rebuilt as case studies and a steady content cadence. More municipalities is more work, and that is the honest driver of the number. |
| Larger firm, commercial or mixed, multiple project types | A different problem entirely. Commercial work is won on relationships and reputation rather than local search, so the honest answer is usually less SEO and more positioning, case studies and the site itself. If that is you, say so on the call and I will tell you whether search is even the right spend. |
For context on the top of that range: a practice running several municipalities with an ongoing content programme and a feasibility offer built for it is a low four figures a month, not $800. I am telling you that rather than letting you find out on the call, because the opposite approach is what the whole field does and it wastes everybody's time.
The reason the floor is published at all is that on a field of seventeen agencies serving this profession, only two publish any number whatsoever. Everyone else requires a call before they will tell you what anything costs. That is not confidentiality. It is a qualification filter, and you are the one paying for it in time.
What the first call is: 30 minutes, no deck. I will have looked at your site and your search visibility before we speak, so I can tell you what I would do first and roughly what it is worth. If the honest answer is that you do not need me, I will say so on that call rather than three months in.
I work with one architecture practice, in the Pacific Northwest. One client is not a pattern and I will not dress it up as one.
The transferable work is diagnostic and not sector specific: a local search rescue that moved a service business from position seven to position two in the map pack over 90 days, and a form rebuild that took conversion from 1.4% to 12.3% on the same traffic. Both are written up lever by lever on my case studies page, including the levers that did nothing.
My public Upwork record is the part I cannot edit: 225 jobs at a 100% Job Success Score.
Before you talk to anyone, including me, it helps to know what you are actually buying. These are free and there is no signup:
Referral from past clients and repeat relationships with developers and builders account for the majority of work at most established practices. Directories such as Houzz and local search make up most of the rest. The risk is that referral is entirely outside your control and it tends to bring you more of the work you have already done.
Usually one of three things: a key developer or contractor relationship ended, the market turned and discretionary residential work got postponed, or your past clients have simply run out of people to refer. None of them are a reflection on the work. All of them are why a second channel matters before you need it.
It works and it is rent. You are buying enquiries you do not own, on a platform that sets the terms, and it stops the day you stop paying. Reasonable as a stopgap while you build your own visibility, poor as a permanent strategy.
It is genuinely good for aspiration, for staying visible to past contacts, and for attracting staff. It is weak at capturing someone who has actively decided to build, because that person goes to Google. Treat it as brand maintenance rather than lead generation.
Permit and zoning answers for your specific jurisdiction, honest cost ranges for the project types you want, and a plain explanation of what working with an architect involves. Those reach the homeowner while they still have an open question, which is months before they search for a firm.
Local search typically shows first movement in roughly 60 to 90 days, and architecture has an unusually long consideration period on top of that, so plan in quarters rather than weeks. Anyone promising enquiries next month is describing something they do not control.
Yes, but not marketing. If you are booked out, the lever is pricing, not demand. The time to build a second channel is while the referrals are still flowing, because building it after they stop takes months you will not have.
Thirty minutes. Tell me where your last few customers came from and what you could handle more of. I will tell you whether marketing is your constraint or whether something else is, and I will say so plainly if it is not.