GraceZero
What We Build

Software that gets measured

Five services, available now. Start with a fixed-fee review, automate the manual work between the systems you already run, or build the thing outright. Whatever we ship gets instrumented, and the same behavioral data drives everything after it.

When the build’s done, you decide how it runs. We can hand it off to your team, fully documented, walked through line by line, and yours to operate. Or we can keep it on our infrastructure and bill a flat monthly to run, measure, and improve it. Enterprise consultancies quote six figures for this scope. Our pricing is published.

Already Running Something?

Most of our work starts with software that already exists.

A website that stopped earning its keep. A portal your customers avoid. An app someone built three years ago that nobody has touched since. A process that works, barely, and only because people remember to do it. You don’t have to start over to find out what’s wrong.

We start from the outside, before you give us anything. We go through your customer flow the way your customer does, on a phone, and time it. That alone usually finds it. What you already have is the most honest brief there is: it shows us exactly where the work breaks down and what it actually needs to do.

What’s actually wrong

Not opinions about your design. The specific places your customer gives up: the booking that takes nine taps, the form that fails on a phone, the portal nobody logs into twice. Evidence, from the outside, before you spend anything.

Whether it’s worth rebuilding

Sometimes it isn’t, and we’ll say so. If the honest answer is an off-the-shelf tool you can buy for $40 a month, or a change to how the work flows that costs nothing at all, we’d rather tell you that than sell you a build.

Why we rebuild instead of patching

We don’t take over other people’s codebases. Not because we can’t read them, but because we can’t promise our standard on a foundation we didn’t lay. Backups, monitoring, security posture, test coverage: none of that bolts onto someone else’s shortcuts. If we’re going to stand behind it and keep it alive, we build it.

What it costs and what you get

A written plan with a number attached, not a proposal that hides the number until the third meeting. You keep the document either way, and every price is already published.

The paid version of that answer is the architecture review below: $2,500, fixed, and credited in full toward the build if you hire us within 90 days.

Start Here · Architecture Review

Two weeks. A written plan. Yours either way.

We audit your existing systems, workflows, and tooling, then hand you a written architecture recommendation: what to fix first, what to keep, what to retire, and where Ai actually helps versus where it’s being sold to you. You keep the document. Hire us to execute it or don’t.

00 · Architecture Review
Available now

Architecture Review.

Fixed fee, fixed scope, and credited toward the build.

Useful before signing a SaaS contract, before hiring, before committing to a build, and especially when something is clearly wrong and nobody senior has looked at the whole picture. It’s also how we scope a rebuild honestly: what you already run tells us what the replacement actually has to do.

Want to see the operations layer underneath all of this? See Hopper Ops → the dashboard your build would land on.

  • Your customer flow walked end to end, on a phone, the way a customer hits it
  • Discovery interviews with key stakeholders
  • Existing systems & tooling audit
  • Workflow & data-flow mapping
  • Measurement baseline: what you can and can’t currently see
  • Written recommendations & priority order
  • Live walkthrough of findings

$2,500 fixed. Credited in full toward the build within 90 days. Walkthrough before you pay anything beyond the deposit.

[ get a quote → ]
The Five Layers

Five services available now. Two more once there’s history to run them on.

Instrument the software once, and the same behavioral data drives every layer above it.

01 · Review
Available now

Review.

Find out what it costs before you commit to it.

A fixed-fee look at what you are running now and what it would take to replace it. You get a written answer on what is actually wrong, whether a rebuild is worth doing, what it would cost, and what you would get. Sometimes the answer is that off-the-shelf software already solves it, and we will tell you that instead of quoting you a build.

Who this is for

Anyone weighing a rebuild, a migration, or a first custom system, who wants a real number before committing. Also the right first step when nobody can agree on whether the current setup is the problem.

  • Written assessment of what you run today and where it actually breaks
  • Rebuild or keep, with the reasoning
  • Scoped cost and timeline you can take to a board or a partner
  • Off-the-shelf recommendation when that is genuinely the better answer

$2,500 fixed, credited toward the build if you hire us within 90 days.

[ get a quote → ]
02 · Automate
Available now

Automate.

Remove the manual work without rebuilding anything.

The work that happens between your systems: rekeying the same data into three places, the report someone rebuilds by hand every Monday, the follow-up that only happens when somebody remembers. Built on n8n, with Claude routed in only where reasoning actually helps and not where a rule is cheaper.

Who this is for

Operations losing hours a week to copy-paste between tools that do not talk. Often the right first project: it pays for itself before anyone commits to a full build.

  • Process audit of what actually happens today, not the version in the org chart
  • Workflow design and build on n8n
  • Models routed per job, so you are not paying premium rates for low-stakes calls
  • Tested against your real data, not a demo set
  • Documentation written for non-technical operators

Project. Smaller scopes close in 2 to 4 weeks.

[ get a quote → ]
03 · Build
Available now

Build.

The web app itself. The foundation everything else runs on.

A web app built for your organization’s specific needs. Not a template stretched to fit. Not a SaaS subscription you’ll outgrow. Customer and member portals, internal operations tools, and brochure sites rebuilt as working web apps. Automation on n8n with Claude routed in where reasoning actually helps, and Ai inside the product only where it removes work.

Who this is for

The shop, the practice, the studio, the site. Civic organizations, small firms, owner-operators, and trades running operations on spreadsheets, group texts, and workarounds that have gotten too large to manage by hand.

  • Requirements gathering & architecture
  • Full web app build & testing
  • Instrumentation wired in from day one, not bolted on later
  • Complete technical documentation
  • Live handoff walkthrough

Project. Scoped before we start, delivered when it’s right.

[ get a quote → ]
04 · Intelligence
Available now

Intelligence.

Because "I think it’s going well" isn’t a number.

Most software ships blind. Nobody can say which features carry the load, who quietly stopped showing up, or where people give up. Yours answers all three from day one. You stop paying for features nobody opens, you catch the customer drifting away while you can still call them, and when something breaks you hear it from us, not from the person it broke for.

Who this is for

Anyone who has ever asked "is anyone even using this?" and had no way to find out. Anyone with members or repeat customers, where somebody going quiet costs real money. And anyone about to spend on a redesign without a baseline to compare it against.

Setup, once
  • We decide together what actually matters in your operation, then name those actions so they mean the same thing forever
  • Every one of them wired into the app and recorded
  • Funnels and retention: where people start, where they finish, where they quit, who comes back
  • Session replay on the screens that matter, so you can watch a real person get stuck instead of guessing why
  • One dashboard built around your questions, not a template
  • A baseline snapshot, so every later change has something to be measured against
Every month
  • A written report in plain language: what people did, what they ignored, what moved
  • The three things we’d fix next, ranked by how many people they affect
  • Every new feature instrumented before it ships. Nothing launches blind
  • The dashboard stays live and yours to open whenever

Why monthly, not one-time: a single snapshot tells you almost nothing. The value is the trend, and every feature you add is a fresh blind spot unless someone wires it up. What it isn’t: Google Analytics, which counts visits. This tells you what happened inside the app. Runs on software we built.

$3,500 setup, then $500/mo. Priced alongside the build.

[ get a quote → ]
05 · Care
Available now

Care.

It lives here and we keep it alive.

Managed hosting on infrastructure we own and harden, watched by the same operations layer we check on our own systems every morning. Nightly backups that get restore-tested rather than just written, because a backup nobody has restored is a rumor. Incident response puts a person on the other end, not a ticket queue.

Who this is for

Anyone who doesn’t want to be the one holding the pager at 2am. Fully optional: you can take the whole build in-house, documented line by line, and run it yourself.

Every month
  • Managed hosting on infrastructure we own, your data isolated from every other tenant
  • Uptime, health, and backup-freshness monitoring
  • Nightly backups, verified by actually restoring them
  • Security patching, dependency updates, CVE watch
  • Ai model-deprecation alerts, so a retired model never takes a feature down without warning
  • Incident response under a written SLA

What Care covers: keeping what we built running, secure, backed up, and reachable, including fixing our own bugs. What it doesn’t: new features and scope changes, which get quoted before any work starts, like everything else. What moves it above the floor: number of tenants, traffic and data volume, and how fast you need someone on the phone when it breaks.

From $450/mo. Every build also ships with a 30-day support window before Care starts.

[ get a quote → ]
06 · Optimize
Next tier

Optimize.

We don’t guess what to fix. We prove the win.

The measurement layer tells you where people give up. This layer does something about it: a change shipped behind a controlled A/B test, measured against the version it replaced, and reported with the actual lift. When a change loses, we say so and revert it. That happens, and you’ll hear about it from us first.

Who this is for

Web apps with enough traffic for a test to reach significance. If yours isn’t there yet, we’ll say so rather than bill you for an experiment that can’t conclude.

  • Impact-ranked fix list built from the data
  • A/B tests against a success metric declared up front
  • Before and after reporting with real lift numbers
  • Losing variants reverted and reported honestly
  • Quarterly review against your operation’s goals

Retainer. Starts once there’s enough traffic to test against.

[ get a quote → ]
07 · Engage
Next tier

Engage.

We turn drifting users back into active ones.

Your data already knows who signed in once and never returned, who was a daily user and quietly stopped, and which group is three weeks behind everyone else. This layer turns that into action. We own the data and the segmentation; delivery goes out through your email or push tooling.

Who this is for

Operations with repeat customers or members, where someone going quiet costs you money. Memberships, portals, recurring services, seasonal programs.

  • Active, at-risk, and dormant segmentation
  • Automated nudges on the triggers that matter
  • Win-back sequences measured on recovery, not sends
  • Lifecycle reporting per segment
  • Your behavioral data stays on our infrastructure

Monthly. Starts once the measurement layer has history behind it.

[ get a quote → ]

Why two of those say next tier. Optimize and Engage both run on measurement history. An A/B test needs enough traffic to reach significance, and a win-back campaign needs to know what active looked like before someone drifted. Sold on day one, they’d be theater. They come online once your build has produced enough data to work from, and we’ll tell you when that is.

Every price is published.

Review $2,500 fixed · conversions from $9,000 · builds from $15,000

[ see pricing → ]
Website to Web App

Your brochure site, rebuilt as something that works.

The most common Build engagement we run. Your site stays your site, and it gains a login, a portal, and a memory. It installs on a phone from the browser with no App Store and no Google Play in the middle, and it starts producing behavioral data a static page can never generate.

What the conversion includes

Your existing content and brand carried across intact. Authentication and roles. The portal surfaces your customers actually need: booking, requests, documents, status, payments where it applies. Installable to the home screen, offline-tolerant on the screens that need it, and instrumented from the first day it’s live.

What you stop paying for

No App Store review queue and no Google Play review queue standing between you and a fix. No annual developer program fees. No second and third codebase for iOS and Android. We don’t build native mobile, and that’s deliberate. If your build genuinely needs it, we’ll say so and point you somewhere good rather than take the work.

The Terms, In Writing

How every engagement runs.

  1. Fixed quote before kickoff.The number is in writing before any work starts. Scope changes get a new quote, not a surprise invoice.
  2. No deposit until the scope is signed.You see scope, deliverables, and timeline in writing and agree first. Then the deposit happens, not before.
  3. You own the build.Code, documentation, credentials. Handed over at the walkthrough, yours to operate or move. Nothing held hostage.
  4. Post-launch support included.Every build ships with a support window after handoff. After that: run it yourself, or keep us on retainer. Both paths priced in the scope.
  5. Monitored from day one.Builds are held to the same Hopper Ops standard our own systems run on. Public view any time →
  6. Instrumented, not guessed.Measurement ships with the build, not as an upsell after it. You get to see what people actually do with the software, including the parts that disappoint us.
FAQ

Common questions before kickoff.

Q.What does custom software cost for a small business?
Custom builds start at $15,000, and most engagements land between $20,000 and $35,000 depending on scope. A brochure site rebuilt as a web app starts at $9,000, and a fixed-fee architecture review is $2,500, credited toward the build if you hire us within 90 days. Enterprise consultancies quote six figures for the same scope. Full pricing is published on the pricing page, and the Phase Zero call gives you a fixed quote in writing before anything starts.
Q.How do you measure whether the software is actually working?
We instrument it on day one. Every meaningful action is recorded, so within weeks we can tell you which features carry the load, which are ignored, who signed in once and never came back, and when during the day people actually work. On our own flagship build that review found 44 percent of activity happening between 6pm and 8pm, on the field. That one finding killed the desktop-first admin view, and we rebuilt for the sideline. That is the Intelligence layer, and it is why we do not guess what to fix.
Q.Will you work on the app or website I already have?
We will review it, and the review is genuinely useful on its own: you get a written plan with a number attached, and you keep it whether you hire us or not. What we do not do is take over someone else’s codebase and maintain it. That is deliberate. We promise backups, monitoring, restore-tested recovery, security posture, and test coverage, and none of that can be honestly promised on a foundation we did not lay. So if what you have needs to become something we will stand behind and keep alive, we rebuild it properly rather than patching around decisions we cannot vouch for. Sometimes the review concludes it is not worth rebuilding yet, and we say so.
Q.Why a web app instead of an iPhone or Android app?
A web app installs on a phone from the browser: your customer taps add to home screen and it is an icon like anything else. No App Store review queue, no Google Play review queue, no annual developer fees, and one codebase covering phone, tablet, and desktop. It also works offline for field use. We do not build native iOS or Android, and that is deliberate: native means two more codebases and two more places an urgent fix waits on someone else’s approval. If a build genuinely needs native, we will tell you and point you somewhere good.
Q.What is a fixed-fee architecture review?
A two-week engagement. We interview your team, audit your existing systems and tooling, map your workflows and data flows, then deliver a written set of recommendations with a priority order. The fee is fixed up front. You keep the document, hire us to execute or not.
Q.Do I need an Ai consultant, or can I use off-the-shelf tools?
Often the answer is off-the-shelf with one or two custom pieces wired in. If the only Ai work you need is summarizing emails, a $20-a-month SaaS does that. If you need Ai routed against your own data, your own workflows, or a private model, that is where custom work earns its money. The architecture review settles which side of the line you are on.
Q.How long does a custom web app take to build?
A typical scope runs 6 to 12 weeks from kickoff to handoff. Smaller workflow automation projects close in 2 to 4 weeks. Larger multi-tenant SaaS builds run 12 weeks or more. The scope is locked before kickoff, so the timeline is too.
Q.What is the difference between custom software and SaaS?
SaaS is a generic tool with a monthly bill that grows with usage. Custom software is built for your specific operation, owned by you, running on infrastructure you control. SaaS makes sense for commoditized work like email and accounting. Custom makes sense for the work that defines your operation.
Q.What is included in a workflow automation project?
Process audit, workflow design, the build on n8n with Claude routed in where reasoning actually helps, testing against your real data, and documentation written for non-technical operators. Models are routed per job (Opus, Sonnet, or Haiku) so you are not paying premium rates for low-stakes calls.
Q.Do you build on your infrastructure or mine?
Either. After delivery you decide. We can hand the build to your team, fully documented and walked through line by line, and it is yours to operate. Or we keep it on our infrastructure and bill a flat monthly to monitor and update it. The right call depends on what you want to own.
Q.What happens after the web app is delivered?
Every build ships with full technical documentation and a live walkthrough. Post-launch support is included for the window after handoff. After that, some clients run the web app independently, others keep us on retainer for the measurement, hosting, and monitoring layers. Either path is documented in the scope before kickoff.

Not sure which one fits? Tell us the problem.

We’ll tell you what we actually think, even if the answer is “you don’t need us.”

[ get a quote → ]