For service businesses that already spend on marketing

Before anything changes, the current setup gets written down.

Collister Computing can take on the website, the search work, the reporting and the request queue for a service company that already has all of it running. The engagement is quoted and scoped in writing, and it opens with about ten business days of read-only work. It is not limited to the Treasure Valley.

No guarantee of rankings, leads or revenue is offered here, and none will be written into an engagement.

Who this is for

Who the engagement is for.

  • The spend is already real. A multi-crew service company with a website, tracked phone numbers, advertising accounts, and a marketing invoice arriving every month.
  • Nobody can produce the record. No one can say on request who owns each account, what shipped last week, or what a completed job cost to buy.
  • Three people have to be served. The owner who wants requests to move, the CFO who wants spend tied to booked and completed jobs, and the in-house marketer who needs access to the systems.
  • Nobody is replaced first. The first engagement is read-only and runs alongside whoever does the work today.
  • Looking for one clean page instead? The fixed monthly one-page website plan for Treasure Valley businesses is on the website service page.

The starting point

A paid, fixed-scope proof sprint of about ten business days.

The sprint is read-only. It begins after a signed engagement and written authorization. It changes nothing that is live, and it does not ask anyone to end a current agreement, unlink an account, or move a phone number.

What it establishes is what is actually running, who controls it, whether the business could recover every account and asset on its own, and whether spend reconciles to booked and completed jobs.

  1. Days 1–2 — Inventory

    Every account listed with who owns it, who can recover it, and what stops working if a vendor’s access ends. Every published phone number mapped to the surface it appears on and the routing behind it. Definitions written down, from lead through gross profit.

  2. Days 3–4 — Baseline

    A full files-and-database backup, with a restore demonstrated on staging. A complete crawl of the site. Exports from the analytics, search, advertising, business-profile and CRM systems already in place.

  3. Days 5–6 — Reconcile

    A first scorecard from spend through to gross profit. Unmatched calls, forms and jobs counted rather than smoothed over. A URL and redirect map. Service lines and territories ranked against crew capacity.

  4. Days 7–8 — Three real requests

    Three requests that have been waiting get built as bounded work packets, in staging or as proposals only, each with before-and-after evidence and a way back.

  5. Days 9–10 — Decision memo

    What the current arrangement should keep doing, what could transition, what needs a specialist, and a proposed scope for the next ninety days.

What exists at the end

  • Control and dependency matrix. Every account, who owns it, who can recover it, and what stops working if access ends. Some accounts have no second administrator who can recover them on the business’s own side. The matrix records each of those gaps.
  • Definition registry. Written definitions from lead through qualified lead, booked job, completed job, collected revenue and gross profit. The owner and the CFO approve them before anything is reported.
  • Verified backup and a demonstrated restore. Files and database together, because a content export leaves the database behind.
  • Baseline manifest and freshness report. Every imported figure names its source, its period, its definition and how complete it is.
  • Scorecard and attribution-gap report. Spend separated into fee, software and media, then followed through the funnel. Unmatched calls, forms and jobs are counted and explained.
  • Migration and opportunity map. Old and current URLs, redirects, canonicals and search performance, set against service line, ZIP code, crew capacity and margin.
  • Three reviewed work packets. Three real delayed requests, built in staging or as proposals, with before-and-after evidence, a recorded approval and a way back.
  • Decision memo and a proposed scope. What should continue, what could transition, what needs a specialist, and a 30/60/90-day scope for the work that follows.

The memo can recommend keeping the current arrangement exactly as it is. That is one of the results it is allowed to reach.

Out of scope

What the sprint does not touch.

  • No production change to a website, redirect, DNS record, business profile, campaign or tracking number.
  • No advertising budget or bidding change.
  • No customer outreach and no review solicitation.
  • No call recording, transcript or customer record placed into an AI tool.
  • No notice of termination to an existing provider, and no public criticism of one.
  • No full website rebuild.
  • No guarantee of rankings, leads or revenue.

How the work runs

Inside the engagement, every request becomes a tracked change.

A tracked change carries an owner, a deadline, the evidence behind it, a staging preview, a recorded approval, automated checks, a deployment record, a measurement plan and a way back. Requests can arrive by email, text or chat and land in one queue. The queue spans the website, search, paid media, local listings, content and attribution, and it is reviewed with the owner on a cadence set in the engagement. The queue is open-ended. The engineering capacity behind it is not, so priority and fair-use terms are set in the engagement agreement before work starts, rather than promising unlimited simultaneous work. Outside the sprint the same limits hold: no customer records in an AI tool, no public criticism of a provider, and no change to a budget or a live system without a recorded approval.

  • Staging before production. The engagement puts code, configuration, redirects, schema and tag-manager exports under version control, and runs automated checks on links, sitemaps, schema, forms, phone numbers, analytics, accessibility and performance, with screenshot comparison on the pages that earn revenue.
  • A restore demonstrated before targets are set. Several recent backups in separate locations, and a restore performed before any delivery target is agreed.
  • A second reviewer on material changes. Where a material change has a named reviewer in the engagement, that reviewer is never the person who proposed the change. No automated step records a person’s approval.
  • Access granted per task. Permission is granted for the task at hand and does not widen because a credential happens to be available.
  • Named reviewers. Austin Ross leads the work. Named specialist reviewers for paid media, editorial, design, and legal or privacy questions are brought in per engagement. Each one sees only the evidence inside the decision being reviewed.
  • How the tooling is used. Automation drafts, researches and checks, with sources recorded and a person approving. It does not publish, change a budget, answer a public review, or take in call recordings, transcripts or customer records.

Ownership and reporting

The accounts stay in the business’s own name.

Accounts and data

Hosting, domain, analytics, tag manager, advertising and business-profile accounts stay owned by the business. So does the creative produced for it: ad copy, images and video. The advertising platforms bill the business directly. Manager access is granted and can be withdrawn. Business Profile ownership stays with the business, which is also what Google requires. Export and transition assistance on termination goes into the engagement agreement before work starts.

Reporting

Reporting separates fee, software and media spend, then follows spend through qualified lead, booked job, completed job, collected revenue and gross profit. Every indicator carries a formula, a source system, an owner, a refresh cadence and a reconciliation rule. Where two systems disagree, the difference is quantified and shown. Rankings appear as a diagnostic.

Response targets

Set in the engagement, once coverage is assigned.

Response and delivery targets are set in the written engagement. They are set once coverage for each function is assigned and recorded. Incidents that take a site, a form or call tracking offline are acknowledged first. Standard content changes come next. New service or location content runs on a longer clock. Around-the-clock response is not offered.

These targets belong to the quoted engagement described on this page. The one-page website plan has its own terms, and they are in writing on the website service page.

Transition

Functions move one at a time.

  • Observe first. Read-only. New reports run in parallel against at least thirty days of each platform’s own figures before anyone relies on them.
  • Then one low-risk change. A single change, deployed with an exact way back, to prove the workflow end to end.
  • Website and reporting next. These transition first, because their dependencies are the easiest to verify.
  • Then technical and content search work, then local listings. Old routing, tags and accounts stay in place through their attribution and porting windows.
  • Paid media last, after specialist review. Specialist review covers budget, bidding, match types, negative keywords, conversion setup and geography for the first sixty to ninety days. It happens before paid media moves.

No function moves while a critical dependency is unknown, a source is stale or unreconciled, or a restore has not been demonstrated. No function moves while the business would still need a departing provider’s cooperation to keep an asset.

Questions

What owners and CFOs ask first.

Will rankings or paid performance drop during a transition?

They can, and nobody can promise otherwise. The work is sequenced so that a drop is detectable and reversible. Old routing, tags and accounts stay live through their attribution windows. One function moves at a time, and a way back is demonstrated before each step.

What happens to the phone numbers?

Every published number is mapped before anything moves: the surface it appears on, the carrier of record, the forwarding hops, the after-hours routing, and what it reports into. Numbers printed on trucks and in old advertising are traced first, and porting authority is confirmed before a port is requested.

Who does the work, and what happens if the primary is unavailable?

Austin Ross leads the work. Named specialist reviewers for paid media, editorial, design, and legal or privacy questions are brought in per engagement, each receiving only the evidence inside their own scope. Coverage for each function is assigned and written into the engagement before any target for it is agreed. Where a function has no covering reviewer, the engagement says so in writing.

What does it cost?

The engagement is quoted after the sprint has shown what the work is. The structure is a defined core scope with written exclusions. Advertising and software are billed to the business directly by the platforms. There is no percentage of advertising spend and no compensation tied to raw lead count. The sprint itself is priced and agreed before it starts. The engagement’s length, its notice period and what happens at the end are set in the engagement agreement before work starts. The one-page website plan’s terms, linked in the footer, govern that plan only and do not apply here.

What is handed back if it ends?

An export another operator could pick up: accounts already in the business’s name, the definitions, the reporting history, the work packets, the backups and the access records. Transition assistance on termination goes into the engagement agreement before work starts.

Contact

Email Collister Computing.

Say what is running today, who touches it, and what you cannot see from where you sit. That is enough to scope the sprint.

austin@collistercomputing.com