Bay Area restaurant tenant improvement costs in 2026 land in the neighborhood of $120 to $320 per square foot in our practice and across publicly reported industry construction cost indexes — with cosmetic refreshes at the low end, full-service kitchens and full ground-up build-outs at the top. But the per-square-foot number isn't what determines your actual budget. Six factors — the existing kitchen infrastructure of the space you lease, building age and existing-conditions surprises, whether you trigger change-of-use, your finish level, the lease itself, and your jurisdiction — can move the same project up or down that range meaningfully, often by 40 to 60 percent. This guide is for restaurant owners, operators, and investors who want to land at the low end. It's written from an architect's chair after running TI projects across Walnut Creek, Oakland, San Francisco, and the Tri-Valley.
We get a version of the same call every few weeks. An operator has a concept, has a broker showing them three spaces, and wants to know what the build-out is going to cost so they can underwrite the deal. The honest answer is a range so wide that it's nearly useless without context. The useful work happens before that call: in how you read the space you're about to lease, in which lease terms you negotiate, and in which decisions you make in design before the GC ever bids. That's what this guide is about.
The short answer: the range is real, but it's the wrong question
Across publicly reported industry construction cost indexes and our own practice in 2026, restaurant tenant improvement in Bay Area submarkets falls into roughly four cost tiers, with most projects landing in the middle two:
- Cosmetic refresh (existing restaurant, finishes only): $40–$90/sf
- Partial build-out (existing kitchen, new front-of-house): $120–$200/sf
- Full build-out (new kitchen + full FOH from cold shell or non-restaurant use): $200–$320/sf
- Adaptive reuse / heritage-building conversion: $200–$350/sf
Those are the headline numbers. They're also the numbers every broker, every cost consultant, and every other content site quotes back to you. What they don't tell you: the same 2,000-square-foot space can land at $130/sf or $280/sf based on six decisions, most of which are locked in before construction even starts. The architect's job — the actual job, not just drawing — is helping you make those six decisions in your favor.
Factor 1 — The kitchen that's already in the space you're about to lease
This is the single biggest line item you control, and you control it before you sign the lease. A space that comes with an existing Type I hood (sized to the equipment you intend to install), a code-compliant grease interceptor (sized correctly for your anticipated drain load and accessible for pumping), and an electrical panel with capacity for restaurant loads can save $60,000 to $120,000 against a cold shell or a retail conversion.
What 'existing' actually has to mean to count:
- Type I hood that's already permitted and inspected — not just present. If a previous tenant installed it without permits, your jurisdiction will likely require you to bring it into compliance (which can mean removal and replacement).
- Grease interceptor sized for your equipment, not the previous tenant's. A 750-gallon interceptor sized for a small café won't pass plan check for a full-service kitchen with multiple sinks and a wok station — and replacing or upsizing an in-slab interceptor can mean $15,000–$40,000 plus a multi-week site shutdown.
- Electrical service that supports your load. Restaurants commonly need 400A or larger; many older retail bays come with 200A. The panel upgrade plus PG&E coordination is typically $8,000–$25,000.
- Existing make-up air system and exhaust duct routing through the roof or building shaft. If the building doesn't already have a roof-mounted exhaust fan or a shaft accommodating one, you may need a building-wide structural review or a landlord conversation.
The single best money-saving move available to a restaurant operator is leasing a former restaurant space rather than a retail or office shell. Even if the previous concept failed, the infrastructure investment that failed concept paid for is real, transferable value.
Factor 2 — Building age and the existing-conditions surprises that come with it
Buildings constructed before 1980 in California carry meaningful hidden cost risk: lead-based paint, asbestos in plaster and pipe lagging, and undersized structural framing relative to modern restaurant equipment loads. A pre-construction hazardous-materials survey runs $1,500–$4,500 depending on building size, and the abatement (if needed) can run $8,000–$40,000 depending on scope. Plan for both, even if you ultimately don't need them.
Pre-1965 buildings in particular often need structural reinforcement for the equipment loads of modern commercial kitchens — wok stations, char-broilers, ovens, walk-ins. Walnut Creek, San Francisco, and Oakland each have stocks of older mixed-use buildings where this comes into play. Berkeley and Alameda likewise. The fix is rarely dramatic, but it's almost never priced into the GC's first bid because the engineer hasn't been brought in yet.
How to manage this: commission an architect-led existing-conditions survey before you sign the lease, or at minimum in the contingency period. The survey is $2,500–$6,000 — vastly cheaper than discovering structural inadequacy after demolition.
Factor 3 — Whether your project triggers 'change of use'
This is the biggest single budget driver that nobody talks about until the permit comes back. If your project changes a space's occupancy classification — for instance, going from M (Mercantile / retail) to A-2 (Assembly / restaurant) — you trigger a long list of code upgrades that don't apply to a like-for-like restaurant-to-restaurant TI. The same applies to going from B (Business / office) to A-2.
What 'change of use' typically triggers:
- Full ADA accessibility upgrade across the entire space, not just your TI scope — including the path of travel from the public sidewalk to your front door, the front door itself, all restrooms, all interior routes, and counter heights
- Title 24 envelope and mechanical compliance to current code (not just to the TI scope) for the full leased premises
- Fire-sprinkler upgrades, often including kitchen-hood-area fire suppression beyond what was previously installed
- Possible occupant-load recalculation and exit-capacity review, which can trigger an additional exit door (typically $8,000–$20,000)
- CASp (Certified Access Specialist) inspection or self-certification — required in many California jurisdictions and effectively required for restaurants
A like-for-like restaurant TI (former restaurant to new restaurant, same occupancy classification) often clears plan check in 4–6 weeks. A change-of-use restaurant TI can take 10–16 weeks and adds $40,000–$80,000 in code-driven scope. That's the single biggest reason 'find a former restaurant space' is the highest-leverage move in this entire guide.
Factor 4 — Finish level (the obvious one, but it's not what you think)
Yes, fine-dining finishes cost more than fast-casual. The non-obvious part: finish level decisions during design — not during construction — are what determine whether you stay at the low end of your range or drift up. Three places where this happens:
Specifying once vs. specifying twice. If your architect specifies all finishes during design development and gets pricing before construction documents are stamped, you can value-engineer in design — at architect-time-cost ($1,500–$5,000 of redrawing). If you wait and value-engineer during construction, every change is a change order at the GC's overhead and profit markup (typically 15–25%) plus the underlying labor and material price increase from the contractor's already-locked subcontractor bids.
Custom millwork vs. modified stock. A custom-built front bar in solid walnut runs $18,000–$45,000. The same form factor in a stock species (white oak, maple) with custom face panels can hit $9,000–$18,000 and read 80% the same from across the room. The decision is made on a Tuesday in design development; the cost is locked in on Friday when you sign the GC contract.
Equipment selection cadence. Commercial kitchen equipment has long lead times (6–14 weeks for mid-range fryers, woks, and walk-in coolers in 2026, longer for specialty items). Equipment selected in design and ordered when CDs are issued arrives on time. Equipment selected in CA (construction administration) drives schedule slip — and every week your space sits idle is rent, utilities, and lost opening revenue you don't recover.
Factor 5 — The lease itself
The lease determines whether your TI budget is $250K out of pocket or $250K minus a $100K landlord contribution. The lease also determines whether your project clock starts on the lease commencement date (you're paying rent while you build) or on a delivery-date trigger (you're not). These terms are negotiable. Most first-time restaurant operators don't know to negotiate them.
Lease terms that materially affect your TI economics:
- TI allowance (TIA) — a landlord contribution to your build-out, typically $20–$80/sf in Bay Area Class A and B retail in 2026. Strong-credit national tenants get more; first-time operators get less. The TIA is almost always reimbursable on lien-waiver and final-inspection completion, so plan to front the cash.
- Rent commencement — calendar date vs. delivery date vs. opening date. 'Open and operating' clauses meaningfully shift risk to the landlord. The longer you can defer rent commencement, the more your TI budget effectively grows.
- Free rent / abated rent period — common to negotiate 3–6 months of free or abated rent during build-out. On a $7,500/month lease, six months of free rent is $45,000 of effective budget added to your project.
- Permitting contingency — a clause letting you exit the lease (or extend rent abatement) if permits take longer than X weeks. Critical if the space is in a slow jurisdiction or if you're triggering change-of-use.
- Existing-condition warranties — landlord representations that the existing roof, HVAC, electrical, and plumbing are in working order at delivery. Failing this, you absorb the cost of pre-existing system failures discovered during your TI.
An honest read: a 10-year lease with a $60/sf TI allowance and 6 months of free rent on a 2,000-sf space is worth roughly $165,000 of effective TI budget to the tenant ($120K TIA plus $45K rent abated). The same space at $0 TIA and zero free rent on a 5-year term is a fundamentally different project even at the same per-sf construction cost.
Factor 6 — The jurisdiction
Same project, same finishes, same kitchen — three different Bay Area jurisdictions, three meaningfully different cost outcomes. The variables are permit fees (small but real), plan-check timelines (large impact on rent during construction), and the specific code interpretations local plan reviewers apply (Title 24 energy compliance, hood and grease requirements, accessibility).
Rough order-of-magnitude differences we see across our active project map in 2026:
- San Francisco DBI — two routes that matter. Like-for-like restaurant TIs with simple scope (no change of use, no expansion, no Section 311 trigger) can qualify for DBI's over-the-counter (OTC) review and clear in days to a few weeks. Projects triggering change of use, occupant-load increase, or full ADA path-of-travel review go to standard in-house plan check, typically 8–14 weeks, sometimes with Section 311 neighbor notification adding more. Knowing which track your scope qualifies for is itself a design decision worth making early. Labor rates inside city limits run roughly 10–15% above the East Bay for the same trades.
- Oakland — moderate fees, moderate plan-check (5–8 weeks for like-for-like, longer for change-of-use), front-loaded process (energy and accessibility forms required at intake, not as later corrections).
- Walnut Creek — moderate fees, generally cooperative Building Department, plan check often clears in 4–6 weeks for complete applications. Coordination with Contra Costa Environmental Health for kitchen permits adds time but is generally predictable.
- Berkeley, Alameda, Emeryville — similar to Oakland in cost; Berkeley specifically tends to enforce energy and accessibility more strictly than its East Bay neighbors.
- Tri-Valley (Pleasanton, Dublin, Danville, San Ramon) — generally fast plan-check, moderate fees, suburban-context buildings often newer and easier to renovate, but ABC license geographic constraints can be a factor for bars.
Seven hidden costs the GC bid doesn't include
Your general contractor's bid covers their scope of construction — labor, materials, equipment installation, permit fees they administer. It does not cover an entire 'second budget' that catches first-time restaurant operators by surprise. The list:
- Owner-furnished equipment (OFE) — small wares, glassware, china, cooking vessels, knives, mixers. Typical: $25,000–$60,000 for a full-service restaurant.
- FF&E — tables, chairs, banquettes, decorative lighting, art. Even modest furniture for a 50-seat dining room runs $20,000–$50,000.
- Signage and exterior identity — channel-letter signage with permit, blade signs, awnings, painted exterior. Often $8,000–$25,000 with permits.
- ABC license consultant — for beer/wine ($475 state filing fee) or full liquor ($13,800+ for transfer of an existing Type 47). Most operators retain a consultant for $1,500–$4,000 to manage the process.
- Health Department plan review and permitting (separate from building permits) — Contra Costa, Alameda County, or SF DPH plan check fees run $500–$2,500.
- Pre-opening utilities, hood cleaning, initial inventory of consumables, deep clean — typically $4,000–$10,000.
- AV, POS, surveillance, music licensing, Wi-Fi infrastructure — most restaurants spend $8,000–$25,000 here. PCI-compliant POS hardware alone runs $4,000–$12,000.
Combined, this 'second budget' typically lands between $80,000 and $200,000 for a small-to-mid Bay Area restaurant — meaningful budget that should be planned alongside (not after) the construction estimate. Most lenders and most pro formas leave it out.
Pre-lease checklist: eight questions to ask before signing
Every question on this list saves real money relative to discovering the answer after lease execution. Walk the space with your architect and your GC (or a project manager) before you sign:
- 1. Was the previous tenant a permitted restaurant? (Verify with the city's permit history, not just visual inspection.)
- 2. Is the existing Type I hood sized for the equipment we plan to install, and is it currently permitted and inspected?
- 3. Is the existing grease interceptor sized for our anticipated drain load, and where is the cleanout access?
- 4. What is the existing electrical service amperage, and is there space in the panel for our anticipated loads?
- 5. Does the building shell support exhaust ducts to the roof, and is there an existing roof-mounted exhaust fan we can reuse?
- 6. Will our project trigger change-of-use? (If currently classified as B or M and we'll be A-2, yes. Confirm with the local Building Department, not the broker.)
- 7. What's the building's age, and has a hazardous-materials survey been done in the past 10 years?
- 8. What is the TI allowance, the free-rent period, and the rent commencement trigger in the LOI? Are they negotiable?
When to walk away from a space
The decision to pass on a space is often worth more than any decision made during construction. A few signals that a space will land at the top of the cost range no matter what you do:
- Previous use was retail, office, or warehouse — and lease terms don't compensate with strong TIA and rent abatement
- Pre-1965 building with no documented seismic retrofit and no hazardous-materials survey on file
- Electrical panel at 200A or below, with no clear path to upgrade (no available capacity from PG&E at this address)
- No existing roof exhaust pathway and a landlord unwilling to permit structural penetrations
- Jurisdiction is San Francisco, change-of-use is required, and your runway can't absorb 14–16 weeks of permit review plus 12–16 weeks of construction
- The landlord won't represent existing conditions of roof, HVAC, plumbing, and electrical as being in working order at delivery
Walking away costs you brokerage time and emotional energy. Building in the wrong space costs $80,000 to $200,000 in extra construction, plus months of additional rent during a longer build. The math is rarely close.
What an architect actually does before construction starts
The pattern across every project where we've helped a client land at the low end of the cost range is the same: we get involved before lease execution, walk the space, identify the cost drivers from this list, and either negotiate them into the lease, value-engineer them out of the design, or recommend the client pass on the space. The fee for pre-lease feasibility work is typically $2,500–$8,000 — a fraction of one round of mid-construction change orders.
After the lease is signed, the same pattern continues: we lock finishes and equipment in design development (not in CA), coordinate Title 24 compliance early enough to avoid plan-check resubmittals, and stage permitting through the right agencies in the right order. None of this is exotic. All of it compounds.
If you're at the broker-tour stage on a Bay Area restaurant space, we can usually run a pre-lease walk-through in a week. The shorter and earlier this conversation, the more leverage you have on the lease itself.