How small architecture firms can make more money in 2027

The answer first

There is no recovery to wait for. The August 2026 Architecture Billings Index came in at 47.2, project inquiries at 50.8 and design contracts at 48.3, and the share of firms expecting billings to fall rose from 21% at the end of the first quarter to 29% at the end of the third, with consumer prices up 3.4% on the year (AIA, 23 September 2026). The nonresidential forecast for 2027 is positive, and we show below why it does not reach a residential studio.

Growth for a small firm in 2027 therefore has to come from three things the firm controls: the mix of work, the price on it, and how much of the fee it keeps. Every number here names its source and date, and we say where a figure is a vendor’s or second-hand.

Where the demand is

The only residential segments AIA’s panel rates as improving are inside existing homes. In Q2 2026, additions and alterations scored a net +38% (from +28% a year earlier) and kitchen and bath remodelling +34% (from +26%), while first-time and affordable homes scored minus 25% (AIA Home Design Trends Survey, Q3 2026 report, 26 August 2026). The panel is a quarterly survey of hundreds of residential firms, and a net score is the share reporting improvement minus the share reporting weakening, so it measures momentum rather than volume. Renovation, retrofit, addition and preservation work already produced 45% of US firms’ gross billings in 2023 (AIA Firm Survey Report 2024, November 2024).

The strongest single signal is aging in place. Its net popularity rose to +65% in Q1 2026 from +49% a year earlier, and easier accessibility within the home to +60% from +31% (AIA Home Design Trends Survey, Q2 2026 report, 20 May 2026). In Q2 2026, accommodation for multiple generations scored +53% and first-floor primary suites +52% (Q3 2026 report). Owners aged 65 and over accounted for 27% of home improvement spending in 2023, up from 14% two decades earlier (Harvard Joint Center for Housing Studies, Improving America’s Housing 2025).

Two cautions. The remodelling market is flat in real terms: Harvard’s Leading Indicator of Remodeling Activity projects owner spending of $517 billion in Q2 2026 rising to $519 billion in Q2 2027, with growth slowing to 0.5% (JCHS release of 23 July 2026, read from republications because JCHS blocks automated access; the next release is 22 October 2026). Against 3.4% inflation that is a shrinking pie, so a small firm grows by taking a larger share at a better price. And accessory dwelling units are regional: permits rose about 1,100% nationally from 2015 to 2025 and California issued about 41,000 (Cotality, 1 July 2026; its state table carries no year label), yet the AIA panel’s net rating for accessory dwellings fell to +25% in Q1 2026 from +62% (HDTS Q2 2026 report). Lead with ADUs in California, and with additions and aging in place everywhere else.

Fee methods, and who finds which one profitable

Fixed fees dominate. In the AIA’s January 2026 special questions, 73% of firms used a stipulated sum regularly, and fixed fee was the method most often named most profitable over the past two years, by 30% of firm leaders (AIA Architecture Billings Index, January 2026 data, released 18 February 2026). Small firms were the exception: 33% named hourly their most profitable method. In the same survey, 56% of leaders said negotiating fees is harder than two years ago.

Our reading, labelled as inference, is that small residential scopes change more often than large commercial ones, so a fixed fee agreed at the first meeting carries risk the small firm cannot absorb. The structure that fits is hourly while the scope is being found, then a fixed fee or percentage once it is written down. The fee proposal playbook sets that out stage by stage.

Rate and multiplier beat utilisation

The widest gap in the benchmarks is price. In BQE’s 2026 architecture benchmarking data, the top tier of firms bills $290 an hour against $112 at the bottom, and the billing multiplier is the strongest single predictor of revenue growth (BQE, 6 August 2026). The same analysis found that utilisation and realisation show almost no correlation with profit margin, and that the firms with the highest backlogs wrote down nearly three times more billable work and ran margins 8.2 points lower. BQE sells the software that measures these things, so treat the direction as the finding and the decimals as its sample. Hours billed and hours collected describe how efficiently the work was done after the price was set; the price is where the margin was decided.

Realistic margins for a small firm

Published margins are higher than most small firms will recognise, and the reason is who gets surveyed. PSMJ reported a median operating profit of 20.5% on net revenue for year-end 2025, the highest in the survey’s history (PSMJ, 25 June 2026). Deltek’s 47th Clarity study reported 16.7%, down from 21.4%, across 896 A&E firms (Deltek, 12 May 2026). The AIA’s Firm Survey put profitability at 13.2% of net billings for 2023 (AIA Firm Survey Report 2024, November 2024). And a March 2024 ABI special question put firms billing under $250,000 at 9.0% against 11.5% for all firms, a figure we have only as a secondary summary of the AIA page.

The PSMJ, Deltek and BQE samples are firms that benchmark themselves and buy practice-management software. The 9.0% figure is the closer reference for a sole practitioner, and a five-person firm should expect to sit between the two. If you are in that range you are normal, and the question is which lever to pull first; we have written about where the margin leaks and what a fee discount really costs.

Pipeline: three months of work and falling win rates

Small firms carry the thinnest cushion. Firms billing under $250,000 saw backlog fall from 4.9 months to 3.1 months in Q2 2026, the sharpest drop of any size group, against 6.3 months for all firms (AIA, 22 July 2026). Three months of work is one lost project from an empty studio, which is why sending more proposals is so tempting, and the data argues against it. In Deltek’s 2025 study the median win rate across all A&E firms was 50% (Deltek Clarity, 46th study, 13 May 2025); the architecture-only figure in Deltek’s FY2025 data, which our calculators use, is 45%. A year later firms had submitted 32% more proposals while win rates declined (Deltek Clarity podcast recap, 9 July 2026). Qualifying harder, and declining the enquiries you would not want at full fee, is worth more than volume.

Scope and collections

Two leaks sit between the signed fee and the bank. The first is scope: in Deltek’s FY2025 data, 75% of projects finished on or under budget, so one in four did not (Deltek, 12 May 2026). Additional services absorbed without a fee are a discount applied after the contract; we have put numbers on what scope creep costs, and the scope and change orders playbook has the wording that stops it. The second is cash. BQE reports an average collection period of 49.4 days across firms using its software, and a 12.3% shorter period at firms taking electronic payment (BQE benchmarking page, edition year not shown; BQE is the vendor). Why architects get paid late explains what shortens it, and the getting paid on time playbook has the invoicing cadence and reminder sequence.

AI adds capacity, not fees

Small firms use AI less than large ones: 27% of small firms reported using AI in day-to-day work against 61% of large firms (AIA Firm Survey Report 2024, 2023 data). In Deltek’s 2026 study, 70% of A&E firms had adopted AI, yet only 38% reported a measurable positive business impact (Deltek, 12 May 2026). No survey we found shows AI changing what architects charge, so the honest claim is capacity: hours saved on enquiry replies, proposals, meeting notes and invoice chasing are hours a two-person studio can spend on qualification, site visits and the next client. Lowering the fee because the work took less time hands the gain to the client.

UAE and India notes

The method travels; the figures change. GlobalData forecasts real construction growth in the UAE of 4.8% in 2026, down from 11.1% in 2025, and about 4% a year from 2027 to 2030, with 48% of the federal 2026 budget going to infrastructure and construction (GlobalData, 10 August 2026). Public work holds up while private villa and fit-out demand softens, so expect more fee competition on private work; see what an architect costs in Dubai.

India is forecast to grow 6.4% in real terms in 2026, led by infrastructure (GlobalData release, 2026, an industry forecast). Fee pressure there is structural. The Council of Architecture publishes a Scale of Charges under its 1989 regulations, but a 2015 CPWD office memorandum says the scale is not approved by Parliament or the central government and should not be used to award consultancy work, and The Wire reports that registered architects routinely depart from it (both secondary sources; The Wire article is undated). Our COA article explains the scale. In both markets, positioning and a clear value case do more than a published scale.

What not to expect in 2027

The July 2026 Consensus Construction Forecast raised 2027 nonresidential spending growth to +3.0% from +2.2%, after cutting 2026 to minus 0.3% (AIA, 20 July 2026). The mix matters more than the headline: data centres +25% in 2027 after +33% in 2026, commercial work excluding data centres +1%, offices minus 4.5%. The forecast does not cover residential construction at all. A residential studio should plan for flat real demand and build its year on share and price.

What to do, in order

1. Name the work you want: additions, alterations, aging-in-place and multigenerational projects for owners staying put. Say so on the website, in the first reply and in the proposal.
2. Price for risk and scope clarity: hourly or a paid feasibility stage until the scope is written, then a fixed fee or percentage, with the rate set from the break-even calculator.
3. Qualify harder and decline well. With three months of backlog, every proposal has to carry the cost of the ones that lose.
4. Write the scope and charge for the change. One in four projects runs over budget.
5. Shorten the collection period before raising the fee. Moving from 49 days to 30 costs a well-funded client little and returns cash at once.
6. Measure AI before counting on it. Only 38% of firms can show an impact. Log hours for four weeks, then spend the saving on capacity.