Reputation analysis · Saskatoon, Saskatchewan
Everyone tells you to answer your reviews.
Nobody checks whether it works.
A shelterbelt is a line of trees planted deliberately across the prairie to reduce the force of the wind before it damages soil, crops, buildings or livestock. It does not eliminate the weather. It creates protection, stability, and time to respond.
Shelterbelt Reputation Analysis does the same for your business. Every review you have is read in full, not a sample of the recent ones, and the analysis says plainly what they add up to: what recurs, what is noise, and the one change worth making first. It starts with a free baseline.
A one-time reputation baseline for your business. You'll have it within five business days. No obligation, no sales spam. Your details are used only to prepare and reply about your baseline. Never sold or used for marketing.
The gap
You earned the reputation. We read it from the outside, where its strengths and weaknesses actually show.
You've done years of good work and it shows in your rating. What you may not be able to see is whether that rating is good for your industry, which of the things you're doing actually track with it, and which are efforts that might have been in vain. A number on its own doesn't tell you any of that.
Reading it properly means reading every review on your Google page — all of them, not the recent handful — and being willing to report the hard facts. That is the work we do.
Is an average rating of 4.8 good?
In some industries in this city 4.8 is typical. In others it's the top of the field. Nobody has told you which one you're in.
A blip, or a pattern?
One bad week could be noise. The same complaint three times is a signal, and it may not be the one you'd guess.
Effort that isn't tracking with your rating
The habit you're proudest of may not be the one moving your rating. That's worth knowing before you spend another year on it.
Measured here, not borrowed
Most businesses answer nearly every review. Some answer almost none. Hardly anyone is in the middle.
In August 2026, we directly counted the review responses posted by 44 Saskatoon businesses. Each line represents one business and shows its response rate. Only 4 of the 44 sit anywhere between 25% and 75% — there is no middle of the pack to blend into. A customer reading your page sees one of two kinds of business: the norm, where nearly everything is answered, or its absence.
Answering reviews is a habit a business either has or doesn't, and the ones who have it are not necessarily the ones with the best ratings — that is the part you may not be able to see. The habit is also the norm here, which is the finding most owners get backwards: not replying does not make you unremarkable, it makes you conspicuous.
The free baseline gives you your own figures, read the same way these were.
How it works
Read the right source. Weigh it with judgment. Say it plainly.
No dashboard to log into, no scores to interpret on your own. Your reviews are read in full, measured against the field around you, and the analysis is written up as something you can act on.
We read the listing your customers actually see
Every number comes from your own Google listing, and every listing you operate, if there is more than one. We never take a rating from a site that copies it from somewhere else, or from a search result, because a wrong number is worse than none. We do look at the other public places customers talk about you and tell you what is there, but that is reported as what it is as industry context, and never folded into your rating or your review count.
All of it is read, and the analysis decides what it means
Not a sample and not a summary: every review your business has, across every Google listing you operate, is collected and processed in full before anything is written. Software does the gathering, and an AI model helps read the tone of what people wrote, but the counts and ratings are computed, not guessed. Nothing is reported as a finding unless it survives the statistical checks that separate a real pattern from a run of ordinary variation — and where the numbers are too thin to support a claim, we say so instead of dressing it up. An analyst then reads the computed output and every flagged review, and signs the result. Judgment is the product: knowing which review matters, and which is just a quiet Tuesday.
You get one plain-English document, and drafted replies
Four sections, written for your business, ending in a single decision and the evidence that would change it. Any complaint still sitting unanswered comes with a reply drafted in your voice.
We draft the response — you always post it
Nothing goes out under your name until you have read it. Review each drafted reply, change anything you like, and post it yourself when you're ready.
If you want the deep dive, it's there to buy
The baseline is yours to keep either way, and it may be all you need. If you do want the rest — which issues recur most in your own review record, where you sit against the measured field, one named decision with the evidence that would prove it wrong, and a drafted reply for every open complaint — the full analysis is available as a one-time purchase at one fixed fee, told to you before you commit. One payment, nothing that renews, and it is the only thing for sale here.
Who it's for
Built for the owner whose name is on the sign.
We work in a small number of Saskatoon industries, and only the ones we have already read end to end. That limit is the point: the comparison in section 02 is worth something because the field behind it was actually measured, business by business, rather than assumed.
Home services
HVAC, plumbing and roofing, where one missed-appointment review can undo a month of good work.
Automotive
Collision, auto body and repair, where the work is expensive, infrequent, and researched before anyone phones.
Landscaping & outdoor
Seasonal work with low review volume, where a single bad month is a much larger share of the record.
Insurance
Brokerages, where ratings across the field sit lower than in the other industries we cover and most complaints go unanswered.
Electrical is open and has its own drafting rules, but we have not finished measuring the field, so a report for an electrical business would carry section 02 with a smaller bench behind it, and that is said up front rather than discovered afterwards. If your industry isn't here at all, ask anyway: we would rather turn the work down than sell you a comparison we can't stand behind.
What you receive
Four sections, in this order. The order is the argument.
This is the full analysis: one document, written for you, that you read once and act on. No portal to log into, no dashboard to check, nothing that needs you to remember it exists. The free baseline is a short extract of it — enough to show you what we can see and to prove it with your own numbers, and stopping short of what to do about it.
Your own numbers
Rating, volume, how fast reviews arrive, the quiet weeks, which complaints were never answered, measured across every Google listing your business operates, not just the one you know about. Where a business runs more than one and the reviews are split, we report each listing's figures and say where they don't reconcile, rather than merging them into a single number we can't stand behind. Every complaint still sitting unanswered is on the page a customer reads before deciding whether to phone.
Where you sit among the rest in your industry, in your city
Where your numbers sit against the full field of businesses in your industry we have actually measured in this city: the middle, the range, and where you fall in it. We do not provide competitor intelligence and we name no other business, here or anywhere else in the report. We choose which comparable firms your figures sit beside, not you — the full field we've measured, never a hand-picked few. No other business is named or reported on separately, and your agreement says so in writing.
And what it cannot
Not an average and not a market norm, because a field this size cannot carry one: a number, and what it is out of. How much of your own feedback you answer, how much of your own criticism you answer, and how far apart those two are. We won't set a field reply percentage beside yours. A lifetime reply rate records what was captured as much as what a business does. Some of what you've been told to prioritize may not survive that count.
And what would change it
A single named decision, how confident we are in it, and the specific evidence that would prove it wrong. If it doesn't get you to one thing you can act on, you don't pay for it.
- Paid once, not monthly
- Every figure computed, every claim testable
- We draft your reply — you post it, never us
Any illustrative review shown for format is clearly labelled as such; it isn't a real customer. The industry comparison in section 02 exists only for industries we have actually read end to end. If yours isn't one yet, we'll tell you that before you pay, not after.
Our standard of care
What we do — and what we deliberately won't.
The value is in the discipline. A clear line around what we touch is part of what you're buying, so here it is in plain terms.
- Non-invasive. We read what's public. We never ask for your logins or access to your accounts.
- We draft, you post — never us. When a reply is needed, we prepare one in your voice for your sign-off. You post it; we never post or send anything ourselves.
- There is exactly one thing to buy from this reputation analysis, and this is it. The baseline is free, and the full analysis is the single paid product — one report, one payment. Beyond that there is nothing else to buy here: no website, no SEO, no ad spend, no review generation, no subscription. Not as an upsell, not at any price. If the baseline shows you don't need the full report, that is what the baseline says. Any other work under the Shelterbelt name is sold separately, under its own written terms.
- We do not provide competitor intelligence: no other business is named, anywhere in the report, including at your request. Industry comparisons are anonymized, and your agreement says you can't ask us to read a particular business.
- Accurate by rule. Numbers only come from your own listing, never from a site that copies them.
- Every figure is computed, and every claim is testable. No number in your report is typed by hand or estimated — each one is regenerated from the source data and checked against what the document says. Every conclusion is stated with how confident we are in it and the specific evidence that would prove it wrong, so you can argue with it rather than take it on faith.
- An analyst is accountable for it. Every report is read and signed off before it reaches you, by an analyst who is answerable for its conclusions, and you can reach them directly.
- No guarantees, no theatre. We can't promise a rating, and we won't pretend to be watching your reviews after the report is delivered. Nobody is. What we can promise is that on the day we read it, you will know exactly where you stand and what to do first.
The question we'd rather answer here
"Are you selling this same analysis to my competitors?"
For a one-time read, the honest answer is: possibly, yes. Buying it doesn't reserve your industry, and we're not going to imply it does. What you're paying for is the analysis of you: your position, your patterns, the one decision that follows for your business. Two firms in the same industry get two different reads, because the reading is of them.
We don't sell exclusivity, and we won't offer it. What you're buying is the quality of the analysis you get, and that stands on its own.
Asked and answered up frontPricing
One fixed fee, told to you before you commit. No subscription to cancel.
The baseline is free and it is genuinely useful on its own: it tells you the clearest thing your numbers say, proved with a number. What it deliberately does not tell you is what to do about it. That is the full analysis, and it is the one thing on sale here — a single payment, no monthly plan, nothing that renews. You can answer nearly every review and still be putting the effort in the wrong place; the full analysis counts what share of your complaints you answer against what share of everything else, each shown with what it is out of.
What stands out, proved with a number. A short document, yours to keep whether or not you go further.
- Your rating, review count and how fast reviews arrive
- Every Google listing your business operates, each one counted
- Which complaints are still sitting unanswered
- The clearest thing your numbers say
- Where you sit against the businesses in your industry we've measured
- One worked reply — to your most serious open complaint where one can be answered publicly, otherwise a positive review — so you can see the standard
- Which issues carry the most weight, or what to do about the rest
Which issues carry the most weight in your review record, and exactly what to do — in the order to do it.
- Everything in the baseline, in full depth
- Year by year: how your rating and review volume have actually moved, and the month anything broke
- Every recurring theme your customers raise, sorted by how much weight it carries in your record
- Where you sit in the measured field for your industry — the middle, the range, and your place in it, with no business named
- How much of your criticism you answer against how much of everything else you answer — both counted on your own page, and each one shown with what it is out of
- A reply drafted in your voice for every open complaint — all of them, not a sample. Two exceptions, and we tell you before you pay rather than after: where a complaint is about a named member of your crew, we count it and will not quote it, so it gets a conversation with you instead of a public reply; and a review that is a star rating with no written comment has nothing to reply to, so it is counted and listed rather than drafted for.
- Reusable pattern replies for the positive reviews you never got to
- One decision, stated with our confidence in it and the evidence that would prove it wrong
- Delivered as a PDF and walked through with you on a call
- Twenty-one days from delivery to reject it in writing, and you don't pay for it
Every figure is recomputed — and corrected if it's wrong
Each number is regenerated from the source data and checked against what the document says. If a figure is wrong, tell us and we reissue the corrected report without charging you for it. Your contractual remedy is broader: the twenty-one day rejection right below.
If it doesn't give you at least one thing to act on
Tell us in writing and you don't pay for it. You have twenty-one days from delivery to say so. If you've already paid, the fee comes back in full; if not, the invoice is cancelled. You delete your copies and we're done. You've never seen this before you buy it; that risk shouldn't be yours.
Taxes are extra. One payment for one piece of work. Nothing renews, and there is nothing to cancel.
Privacy & your information
We're deliberate about personal information.
Reputation work touches people's names and words, so we keep a tight, explicit line around what we collect and what we don't, both for you and for the public whose reviews we read.
When you contact us
The only information we collect through this site is the email address and the business name you enter. Our server also notes a couple of technical details your browser sends automatically, only to stop spam and abuse of the form, never to build a picture of you or follow you around the web. Those are cleared after 90 days; the enquiry itself we keep until you ask us to remove it, or up to 24 months after our last contact, whichever comes first. We use them solely to reply to you and to prepare what you asked us for. We never sell or rent them and we don't add you to a marketing list.
When we analyze reviews for clients
We read only publicly available reputation information: the reviews and business listings anyone can see. We never ask for your account logins and we don't access private systems. Reviewers' names are cut back the moment we collect them, before anything is saved: the name becomes a single initial — "Jane Doe" is kept as "J......" — and phone numbers, emails and links inside the review text are stripped at the same moment. A local filter also removes other people's names from the review text, but that filter is not a guarantee. An unusual or ordinary-word first name can slip through, so a name can occasionally appear inside a quoted review. We analyze patterns, not people: we don't build or sell a profile of any individual reviewer, and the reviewer's name field never reaches you.
No cookies. No analytics. No trackers.
This site sets no cookies, stores nothing on your device, runs no analytics, and loads nothing from anywhere else: no outside fonts, no scripts, no tracking pixels, nothing embedded. That's why there's no consent banner: there is nothing to consent to. Every typeface here is one already on your device, and the light/dark switch changes this page only and remembers nothing. We send no analytics, advertising or tracking data from your browser — only the ordinary requests needed to load the page, and anything you deliberately submit through the form.
Erasure — including if you're a reviewer
If you enquired: write to us and we'll delete your email and enquiry, and confirm when it's done. No conditions, no retention "for our records".
If you're a member of the public whose review we hold: the same right applies to you, even though you never contacted us. Write to privacy@shelterbeltintel.com, tell us where the review is and roughly what it said, and once we're satisfied it is yours to ask about we'll remove our copy and tell you when we have. We'll ask only for what we need to find the right record, never for ID; and if we can't confirm a review is yours, we'll explain why rather than risk deleting someone else's words.
Doing this work means passing information through a small number of established third-party services: collecting the reviews, storing them with reviewers' names already cut back, and the AI model that reads review text and writes the first draft of a reply. Our handling of personal information is described in full in our privacy policy, which names every one of these suppliers and what each handles for us. This is a plain-English summary, not legal advice, and not a substitute for the full policy or our terms of use.
Start with a free baseline.
Tell us your business name and we'll prepare your baseline — the same one every client starts with, read from your own Google page the same way the figures above were read. You'll have it within five business days. No obligation, and a reply from a real person rather than an autoresponder.
Your details are used only to prepare and reply about your baseline. Never sold or used for marketing. See our privacy note. Or write to us directly: hello@shelterbeltintel.com