Roasted Beanery: Growth Plan & Website Brief

The plan, number by number

Each lever below shows the recommended setting, the expected result, and why it matters. Baselines are confirmed as of July 21, 2026. Where a number is an estimate rather than sourced data, it says so. That honesty is deliberate and it's how the whole plan is built.

1.1: The scoreboard (90-day goals)

1.2: Followers & the local reality check

We researched every comparable independent coffee shop in the region. Local followings grow in tens-to-low-hundreds of followers per month, linearly, not by viral percentages: Black Crow (Enfield) reached ~1,118 in 17 months; Vernon Coffee Roasters ~1,439 in 4 years; Cafe Barista (Suffield) ~4,571 in 4 years; Dreamscape (Woodstock VT, a tourist town) ~1,884 in 16 months.

1.3: Posting cadence

1.4: Content quality (the honest one)

Current output scores roughly 2 out of 5 on the rubric: hook in the first 2 seconds, lighting, audio/captions, brand consistency, call-to-action, copy. The goal is a sustained 4.0+. This matters more than volume: the algorithm decides distribution in the first seconds of a video, and production polish is not the fix: owner-shot phone video in good light with a real hook outperforms agency gloss for local discovery.

1.5: Ads: the personalized phased answer

Sized against two real things: your actual monthly residual (gross profit after real labor + assumed rent/overhead, before any owner draws, ~$7,976/mo at your July run-rate) and the real seasonality already in the console (US Census food-service data). This exact schedule is loaded into the console's ad-spend log, so the forecast tab's projections come from it directly; nothing here is a placeholder.

What $500/month buys at food-and-beverage benchmark rates ($7–15 CPMs, ~$2.05 restaurant CPC): roughly 23,000–50,000 Meta impressions reaching a large share of the ~80–85k people within 15 minutes, plus ~75 high-intent Google clicks.

1.6: Reviews: the highest-leverage free lever

Target: ~9 new Google reviews/month (register ask + QR card + SMS follow-up). That catches Black Crow's ~111 within a year. Comparable shops run 3–18 reviews/month, so 9 is ambitious but observed in the wild. Review count and velocity are primary map-pack ranking inputs; this is what wins "coffee near me."

1.7: Loyalty & SMS

Their Toast loyalty app already exists (with one rating; nobody knows about it). Relaunch it at the register with an SMS list. The evidence here is the strongest of any channel: loyalty enrollment lifts return rates ~7%→30% (Toast/Resy 2026), members spend 18–30% more per visit, birthday offers redeem at ~35%, and SMS opens at ~98%.

1.8: Forecast & financials (what the model says)

• Where you're actually starting from: real Toast data shows net sales of $42,796 in May, $35,928 in June, and a $32,700 run-rate in July, a real ~10%/month decline since the opening peak, driven almost entirely by fewer guests (average ticket has held steady near $12.50). That's the honest starting line for every projection below; the plan's job is to reverse this trend, not grow from an already-rising baseline.
• Scenario envelope: flat growth at 1% / 2.5% / 5% per month brackets the future; the driver model, which reads the ad calendar, the roadmap phases, and seasonality, currently lands month-12 revenue inside that envelope.
• Seasonality is real and data-backed (US Census food-service sales): January–February run ~9–10% below average, so plan cash accordingly; and the "holiday rush" is mostly myth for food service (November is actually below average; December only +4%).
• September 2026 is the anomaly: the Four Town Fair is in Somers, Sept 17–20; the roadmap builds the whole month around it.
• Financials: enter real revenue/COGS/overhead in the console and it computes break-even revenue, net margin against the 5–18% healthy band, and marketing as % of revenue against the 3–7% coffee-shop norm.

The website platform decision

You have Shopify today. Before defaulting to it, here's the honest comparison: cost, technical SEO ceiling, ease of use, and how far each can go. Prices verified July 2026.

*Share of real-user origins passing all three Core Web Vitals (CrUX/HTTPArchive, Nov 2025–Jan 2026 snapshot). Important honesty note: these measure each platform's population, not its ceiling. Next.js's 68% blends every dashboard and web app ever built with it; Shopify's 77% skews toward big optimized stores while a field test of 1,000 typical stores passed at 48%. The claim we stand behind: a purpose-built Next.js site controls every performance lever and clears the bar by design; Shopify's ceiling is capped no matter how much you spend on apps.

Why performance is a money question, not a vanity metric

• Deloitte/Google, 30M sessions: a 0.1-second speed improvement lifted retail conversions +8.4% and order values +9.2%.
• Portent, 27k pages: a page loading in 1s converts at ~3.05% vs ~1.12% at 3s: nearly 3× the customers from identical traffic.
• Google's own data: each 1-second delay costs up to 20% of conversions; the majority of mobile visits abandon past 3 seconds.
• The ads connection: every ad click lands on this site. Landing-page experience is a Google Ads Quality Score component, and faster pages convert more of the clicks you paid for; speed is effectively a discount on every ad dollar.
• The tracking connection: a custom site runs full server-side tracking (GA4 + Meta Conversions API). Advertisers using CAPI alongside the pixel see ~13% lower cost per result (widely attributed to a 2022 Meta study). Shopify's native pixels run in a locked sandbox.

Apps vs. built-in features

The same capabilities, two ways to pay for them. On Shopify, each need is a monthly subscription; on a custom build, most are simply how the site is made.

Design & sitemap

We'll design this live, together, so the brand stays yours. Below is the proposed structure: the client's requested pages merged with local-SEO best practice for a café.

Proposed sitemap

Local-SEO principles baked into the build

• Name–Address–Phone identical everywhere (site, GBP, Yelp, Facebook, Apple Maps), and always paired with "Somers, CT" to separate from the unrelated Texas "Roasted Beeanery."
• Title pattern: Page · Roasted Beanery · Somers, CT coffee shop; every page targets one local intent.
• Structured data throughout: CafeOrCoffeeShop, Menu, Event, FAQ, hand-authored, not app-generated.
• Static-generated pages, optimized images, sub-second loads: the speed argument from Part 2, applied.
• Reviews surfaced on-site; a /reviews link in every SMS receipt flow.

Design session: we'll build it live (Claude design). A working visual identity already exists from the planning tools: espresso neutrals, latte-foam surfaces, a signature teal pulled from their own Blue Magic spirulina refresher, and a roastery serif, which we can carry straight into the site or evolve with the owners in the room.

What to expect: timing and sequencing

The site is the final 30%

The website is not the marketing plan; it's what makes the rest of the marketing plan convert. Ads, social content, events, and promotions are the 70% that generate attention; the site is the 30% that catches it, turns it into visits, orders, subscriptions, and list signups, and feeds clean data back into the ads. Lead with ads, socials, and events. The site multiplies them; it doesn't replace them.

Good performance takes real backend work

If the goal is a site that makes ads and socials perform better, and it is, then the build includes work nobody sees: keyword and competitor research, information architecture, schema, image pipelines, tracking (GA4 + server-side events), copywriting, and testing on real phones. A pretty page can ship in a weekend; a page that ranks and converts cannot.

Advanced-but-light beats deep-but-beginner

The cheapest website is the one you never have to rebuild. Going deep on a beginner platform means every page, integration, and workflow eventually gets recreated from scratch when the platform's ceiling arrives: that's a paid-twice redesign. Building on the advanced foundation now, even if we launch light (a handful of excellent pages) means every future addition (subscriptions, wholesale, a second location, online classes) is an extension, not a migration. Start advanced and light. Grow deep over time.

Resources for your team

Alongside this plan, you'll get a short list of free, official resources: Google Business Profile help docs, Meta's own content courses, Toast's own training for the tools you already pay for, even a free small-business mentorship nonprofit (SCORE.org), so your team can build skills at your own pace between check-ins. It's in the full written plan (§13), organized by topic so whoever owns each piece can go straight to the relevant one.