Case studies
40 case studies, with the caveats attached
Every engagement here is a real business. Client identities are kept private, every number comes from the engagement itself, and where a result was never measured we say so instead of filling the gap.
How should you read a marketing agency's case studies?
Look for a baseline, a metric, a timeframe and a definition. A percentage without a starting number, a growth claim without a unit, or a lead count without a definition of a lead can all be true and still tell you almost nothing.
These 40 pages carry the numbers the engagements actually produced, and each one states what its number does not prove. Several report no figure at all, because none was ever measured, and those are published as they are.
How to read a case study, including ours
Marketing case studies are the least regulated writing in business, and it shows. A short checklist protects you from most of it.
Almost every case study you will read, including from firms doing genuinely good work, is constructed to be impressive rather than to be checkable. That is not usually dishonesty. It is that the impressive version is easier to write, nobody asks for the rest, and the agency has no incentive to volunteer the parts that complicate the story.
Read the full breakdown: How to read a case study, including oursHide the full breakdown: How to read a case study, including ours
The result is a genre with a predictable set of moves. Percentages without baselines. Growth without units. Business events that happened during an engagement presented as its outcomes. Lead counts with no definition of a lead. Ratings printed as though they were permanent. Each of these can be technically accurate and still leave a reader with an impression the evidence does not support.
We have written these forty pages to be read by somebody applying the opposite standard. Every number on every page came from the engagement it describes. Where a figure was published without a baseline, the page says so. Where a claim on the original page was not measurable, we have restated it and explained why. Where a business achievement happened during an engagement but was not caused by it, the page declines to claim it. Several of these pages are less impressive as a result, and a couple of them carry no figures at all.
The trade is deliberate. Forty case studies where every single one carries a striking number is not a record of forty engagements, it is a selection from a larger set. What follows is the whole set.
Five questions worth asking of any case study
Apply these to the pages on this site, and to every agency you are considering. They take about a minute each and they eliminate most of the noise.
What was the baseline?
A multiple or a percentage means nothing without a starting point. Four times growth from a base of five is twenty. Four times growth from a base of fifty is two hundred. The same words describe two completely different businesses, and the baseline is the most commonly omitted number in the genre.
- Ask what the figure was before
- Ask over what period the comparison was made
- Treat a percentage with no baseline as a direction, not a magnitude
What exactly was measured?
Growth in what? Traffic, impressions, enquiries, bookings and revenue are five different claims and they behave nothing alike. A site can triple its impressions and receive no additional enquiries. Two of the pages on this site report a growth percentage without a stated unit, and both say so rather than picking the flattering interpretation.
Who defined the terms?
Qualified lead is the most elastic phrase in marketing and it is almost always defined by whoever benefits from the definition. A figure based on it is not comparable between two businesses, let alone between two agencies. Ask for the written definition, and ask when it was written, because one set afterwards is not a definition, it is a description.
See the remaining steps: Five questions worth asking of any case studyHide the remaining steps: Five questions worth asking of any case study
What else was happening?
Businesses grow for reasons that have nothing to do with marketing: a competitor closing, a seasonal peak, a price change, a new location, a referral relationship, a good year in the category. An honest case study names the confounds it can. A dishonest one takes credit for the weather.
What is being claimed, and what merely happened?
This is the one that separates careful firms from the rest. Three of the engagements on this site coincided with a significant business event: an acquisition, a practice expanding into a surgical centre, a brand winning retail distribution. All three are on the original case pages. None of the three is claimed here as a marketing outcome, because none of them was caused by marketing.
If an agency cannot answer these five about its own work, that is itself an answer. If it can, the conversation gets considerably more useful very quickly.
Strong evidence and weak evidence
Not every claim on this site is equally well supported, and it would be misleading to present them as though they were. Here is roughly how we would rank the kinds of evidence you will find.
| Item | Stronger | Weaker |
|---|---|---|
| The claim | A metric both parties can see in a platform | A description of how things felt |
| The comparison | A stated before and after | A percentage with no baseline |
| The unit | Enquiries, bookings, patients, jobs | Growth, presence, visibility |
| The period | A named window | Unspecified |
| The definition | Written before measurement began | Applied afterwards |
| The attribution | One channel, tracked | Several channels, blended |
Several pages on this site sit in the right-hand column. They are published anyway, clearly labelled, because leaving them out would make the set look better than the work was.
What the pattern actually is
Read end to end, these forty engagements say something fairly consistent, and it is not what a marketing site would normally lead with. The winning move is almost never a clever campaign. It is fixing something ordinary that nobody owned.
In the local businesses, the single most common finding is a Google Business Profile that had never been configured properly: the wrong primary category, hours that were out of date, no service area set, no photographs of real work. That is a settings problem. It costs nothing to fix, it is the highest-leverage surface in local search, and it was wrong in the large majority of these engagements before anyone touched it.
Read the full breakdown: What the pattern actually isHide the full breakdown: What the pattern actually is
The second most common is a website that describes the business rather than serving the visitor. Services listed the way the company is organised internally, rather than pages answering what somebody actually typed. In the clinics it was treatment menus instead of condition pages. In the trades it was one combined services page instead of separate pages for separate jobs. In the manufacturer it was capability statements instead of specification answers. Same error, five industries.
The third is measurement. A striking number of these businesses could not say where their work came from. Calls arrived on a number that had been in use for years, forms went to an inbox, and nothing carried a source. In several of these engagements the first month's apparent improvement was partly the effect of counting things that had always been happening, and we have said so on the pages where that is likely.
The fourth pattern is sequence. The engagements that reduced cost did it by building earned visibility first and cutting spend afterwards, never the other way around. The ones that produced results in weeks were conversion work on traffic that already existed. The ones that took months were visibility work that created new traffic. Those are two different clocks, and most of the disappointment in this industry comes from confusing them.
The last pattern is the least comfortable. In a meaningful number of these cases the constraint turned out not to be marketing at all. It was a phone nobody answered, a front desk with no process for a high-value enquiry, a form that asked for insurance details before it had earned the right to, or a business at capacity that did not need more demand. More advertising would have made several of these businesses worse.
The environment all of this happens in
SparkToro's 2026 analysis of Similarweb clickstream data measured how many US Google searches end without anyone clicking through to a website.
SourceSparkToro with Similarweb clickstream data, 2026
That shift is the reason a case study from five years ago would read differently. A ranking is worth less than it was, a profile is worth more, and being the source an answer draws on has become part of the job rather than an experiment.
Where the numbers on this page come from
Every result on these forty pages came from the engagement it describes, reported by the client from their own records. We did not audit anyone's accounts, and where a figure is the client's own we say so on the page rather than presenting it as an independent measurement.
The industry statistics are different and are held to a stricter standard. Each one names a real, public, linkable publisher with a stated methodology, and is quoted exactly as that publisher states it. Vendor research is attributed by name rather than laundered into studies show, and figures from older studies carry their year in the visible sentence.
Read the full breakdown: Where the numbers on this page come fromHide the full breakdown: Where the numbers on this page come from
A number of widely circulated statistics did not survive that check and do not appear anywhere on this site. The dental cost per new patient figures, the personal injury cost per lead and cost per case figures, the Google Business Profile photo statistics, the claim that a specific share of searches have local intent, and the small business marketing budget percentage commonly attributed to a federal agency were all traced to blog-to-blog citation loops, withdrawn sources or vendor estimates with no disclosed sample. Several of them would have made these pages look better. They are not here.
About the firm itself, three confirmed facts, stated plainly: 20+ years in the industry, 100+ businesses and clinics, and 92% client retention in 2025. The first of those is experience in the industry, not the age of the company, and we are careful about the difference.
You will not find a star rating or a review count printed anywhere on this site, including on our own pages. Ratings are live values that change every time somebody posts, so a number typed into a page is true on the day it is written and quietly becomes false afterwards. Where a client's rating changed during an engagement, the relevant case study reports it as a historical outcome at the time, not as a current figure.
What the headline figures below do and do not mean
Everything from here down is a grid of engagements, and each card leads with a number. Those numbers are real and they came from the business that produced them, but a headline is the least informative part of any case study, which is why the qualification belongs here rather than at the bottom of the page.
Results are from a specific client engagement and vary by market, budget, competition and other factors. They are examples of past outcomes, not a guarantee of future results.
Read the full breakdown: What the headline figures below do and do not meanHide the full breakdown: What the headline figures below do and do not mean
Two of those variables do more work than the rest. Competition in the specific market decides how hard the same technique is, and it is the variable that almost never appears in a case study. Budget decides what was possible at all, and the engagements below range from a few hundred dollars a month to five figures. Why your result will be different, further down, sets out the rest of them.
So read the cards as a set rather than as a shortlist of best results. Several of these pages report no figure, because none was measured, and they are published in the same grids as the ones that do.
Healthcare and clinics
Eighteen of the forty. Clinics are the hardest local category to move and the most instructive, because the constraint is almost never traffic. It is capacity, intake and reputation. The engagements that worked here fixed the profile and the phone before they spent on reach. Each headline below is one engagement in one market, not a typical result. What made it possible, and what it does not prove, is on the case study itself.
Vein Clinic · Chicago
Monthly ad spend came down from about $20,000 to about $7,000 while qualified leads kept arriving. SEO, AEO and GEO layered under a tightened Google and Meta Ads programme, with CRO on the existing traffic and automated lead capture and qualification.
Read the case studyHair Transplant Clinic · Chicago
Qualified leads from paid advertising increased, on an ad budget of about $6,000 a month. Google and Meta Ads managed toward patient enquiries rather than clicks, with an AI assistant qualifying and supporting prospective patients.
Read the case studyOphthalmology Clinic
Tripled organic website traffic across a year, and doubled traffic from paid ads. A full year of SEO built alongside managed Google Ads and Meta Ads, so organic and paid grew against the same demand rather than competing for it.
Read the case studyOptometry Clinic
More LASIK consults and cataract cases, as reported by the practice. Google Ads aimed at high-value procedures, hands-on consulting on the intake process, and AI automation behind the enquiry flow.
Read the case studyUrgent Care · Chicagoland
Average daily patient count went from 15 to 28. Ongoing local marketing aimed squarely at daily visit volume rather than at rankings or impressions.
Read the case studyHigh-Output Urgent Care
An additional 40 patients a month, on a deliberately small ad budget. A new website that answered the three questions urgent care patients actually ask, plus lean Google Ads against immediate-need searches.
Read the case studyUrgent Care
120 new patients a month by month three, from the organic website and SEO. A new urgent care website, organic visibility work, and brand visibility through Branding Signal, with no dependence on paid search.
Read the case studyMulti-Location Urgent Care · Texas
A 23% gain in local traffic by month two. Healthcare consulting, Google and Meta Ads, website management, patient-experience automations and an AI agent handling routine contact.
Read the case studyClinic · near Tampa, Florida
The Google rating moved from 3.5 to 5 stars, alongside a new website. A new website and a review programme built inside platform policy, with no incentives and no gating.
Read the case studyClinic · Sun City, California
Average Google rating moved from 3.0 to 4.2 stars. Reviews and listings handled through Bird Local, with Branding Signal working on how the clinic appeared everywhere else.
Read the case studyDental Clinic · Florida
Went from not appearing in local results to ranking consistently across roughly a five-mile radius. Local SEO and Google Business Profile optimisation, aimed at the map pack rather than at the blue links.
Read the case studyDental Offices · Texas
Four times growth in new patients, as reported by the practices. A local marketing programme run across the practices, so each office competed in its own area instead of against the others.
Read the case studyMed Spa · Chicagoland
15 new patients a month through SEO. SEO and local SEO built up over time around treatment demand, not around the brand name.
Read the case studyDermatology Clinic · Beverly Hills
Organic growth increased. No figure was published, so none is claimed here. SEO and content in one of the most competitive aesthetic markets in the country.
Read the case studyPediatric Clinic · Chicagoland
More organic presence at a lower ad cost, both reported directionally. Organic visibility built up so the paid budget could be cut back rather than defended.
Read the case studyOccupational and Physical Therapy Clinic · Michigan
45% new-patient growth after three months, reported by the clinic. An SEO campaign scoped specifically to new-patient acquisition, with UI and UX work and conversion rate optimisation on the enquiry path.
Read the case studySpeech Therapy and ABA Practice · Chicago, IL
Enough patient calls from organic search that the practice paused SEO in month four, with no ad spend. An SEO-focused website redesign, with Bird Local handling reviews and listings.
Read the case studyMulti-Location Primary Care Clinic · Chicago
Grew to 60+ reviews at a 4.6-star average, up from a single 1-star review. A review profile rebuilt through Bird Local, the existing website managed and expanded with SEO, and Branding Signal on brand visibility.
Read the case studyHome and local services
Twelve engagements across trades, restoration and cleaning. Speed matters more here than anywhere else on this page: the customer is standing in a wet basement or looking at a broken door, and the first business that answers usually wins the job. Each headline below is one engagement in one market, not a typical result. What made it possible, and what it does not prove, is on the case study itself.
Junk Removal Company
Qualified leads inside the first month, and four new customers. A MyClickPage website built around the quote request, because in junk removal the quote is the product.
Read the case studyAppliance Repair · Chicago
10 to 15% growth in qualified leads within the first three months. Lead-generation infrastructure: the site, the local search footprint and the tracking that shows which of them produced the call.
Read the case studyConstruction Company · Mundelein, IL
Two new customers through the Google Business Profile and the website. A website and an optimised Google Business Profile for a contractor with a small, specific service radius.
Read the case studyFireplace Company · Boston
20% growth reported by the company after a redesign and Google Ads. A website redesign and Google Ads pointed at a business with a hard seasonal peak.
Read the case studyWater Damage Restoration
Organic calls started arriving through SEO. A new website and organic visibility work for a category where the customer is in an emergency and will not scroll.
Read the case studyCarpet Cleaning Company · Indiana
Leads started arriving where there had been almost no online presence. A first real online presence: a website and the local search footprint underneath it.
Read the case studyCleaning Company · near Dallas
Leads started immediately after the website launched. A new website with local SEO for a recurring-revenue service where the first clean is the only hard sale.
Read the case studyRenovation Company · Chicago
41 organic leads in five months, and $200 a month off the running cost. A new SEO website, with Branding Signal and Bird Local handling brand visibility and the local record.
Read the case studyRoofing Company · Bolingbrook, IL
A next-day website the company kept, and $300 a month off the running cost. A MyClickPage website delivered the next day, for a roofer who needed to be online before the next storm cycle.
Read the case studyRoofing Company · Willowbrook, IL
Top three in local search for 14 keywords by month three, with no ad spend. An SEO website launched inside 24 hours, then managed. No separate SEO campaign and no ads.
Read the case studyElectrician · Chicago
The electrician reported being completely booked after the Local Services Ads campaign went live. A Local Services Ads campaign, which is Google's pay-per-lead format for licensed trades.
Read the case studyGarage Door Company · Bridgeview, IL
The company reported being completely booked with new customers. A new website, SEO growth and Google Ads, run together for a same-day repair business.
Read the case studyProfessional services
Two engagements, both of them a lesson in how small a professional services market actually is. A handful of the right enquiries is a good month. Volume metrics mislead badly in this category. Each headline below is one engagement in one market, not a typical result. What made it possible, and what it does not prove, is on the case study itself.
Accounting Firms · Los Angeles and Chicago Suburbs
12 leads and 8 new clients in the first month for the Los Angeles firm. A new website, foundational SEO and Google Ads, plus a second accounting website delivered for a firm in the Chicago suburbs.
Read the case studyInsurance Broker · Illinois
30 times more search impressions in Google Search Console, and $200 a month saved. A new MyClickPage SEO website with ongoing management, replacing a site that cost more and was seen by almost nobody.
Read the case studyLegal
Two firms, two very different economics. Bankruptcy work is high volume and cost sensitive. Personal injury is low volume and the cost per qualified lead is the only number that means anything. Each headline below is one engagement in one market, not a typical result. What made it possible, and what it does not prove, is on the case study itself.
Bankruptcy Law Firm
$900 a month off overall expenses, with qualified organic leads by month two. A custom website and programmatic SEO across the situations people actually search before they call a bankruptcy attorney.
Read the case studyMulti-Location Personal Injury Law Firm
36 qualified leads a month, with leads from ads at close to $105 each. Google Ads and SEO run as one programme at a lower management cost than the firm had been paying.
Read the case studySoftware and SaaS
One engagement, included because it is the clearest example on this page of an outcome that marketing contributed to but cannot claim. Each headline below is one engagement in one market, not a typical result. What made it possible, and what it does not prove, is on the case study itself.
Agencies and B2B
One engagement, and the only case here where the reported outcome is revenue rather than enquiries. Read it with that in mind. Each headline below is one engagement in one market, not a typical result. What made it possible, and what it does not prove, is on the case study itself.
Retail, brands and enterprise
Four engagements that share nothing except scale of catalogue or footprint. What carries across them is that at volume, the work is systems and consistency, not campaigns. Each headline below is one engagement in one market, not a typical result. What made it possible, and what it does not prove, is on the case study itself.
Pre-Owned Medical Equipment Company · National
Four high-ticket qualified leads in month two. An SEO ecommerce build with API integrations pushing the same inventory to other selling platforms.
Read the case studyLabel Manufacturer · Chicagoland
15% organic growth for a B2B manufacturer. SEO and content built around the specification language buyers actually search, not around the company's own product names.
Read the case studyManufactured Consumer Brand
A brand identity and a live ecommerce store. The brand went on to launch in retail stores. Branding from the ground up plus an ecommerce website, for a product that had no market presence at all.
Read the case studyMulti-Location Enterprise · 68 locations
At least 20% local growth across 68 stores, not broken out by measure. Local marketing managed across all 68 locations as one system, with each store treated as its own local market.
Read the case studyWhy your result will be different
The single largest determinant of what any of this work produces is the competition inside your own area, and it is the variable that almost never appears in a case study. A roofing company reaching the local top three for fourteen terms in a suburb with a handful of serious competitors is a real achievement and it is not the same achievement in a dense metropolitan market. The technique does not change. The difficulty does, enormously.
Budget is the second. Several of these engagements were run on deliberately small advertising budgets, and a couple involved reducing a large one. A programme built for a five-figure monthly spend behaves nothing like one built for a few hundred dollars, and both appear on this page.
Read the full breakdown: Why your result will be differentHide the full breakdown: Why your result will be different
Category economics are the third. Four qualified enquiries a month is an excellent result for a business selling capital equipment and a poor one for a business selling a recurring cleaning service. Two new customers is a good quarter for a construction company and a bad week for an urgent care. Comparing raw counts across these pages without holding the economics in mind will mislead you every time.
Then there is what happens after the enquiry. Several of these businesses improved because they changed how they answered the phone, how quickly they replied, or what their front desk said to a high-value caller. Where that was part of the engagement, the pages credit the client's own team, because the work was theirs.
And finally, timing. Urgent care volume moves with the season. Roofing moves with the weather. Hearth retail concentrates most of a year into a few months. Restoration is driven by events nobody controls. A percentage measured across one season in a weather-driven category is indicative rather than conclusive, and the relevant pages say so.
Results are from a specific client engagement and vary by market, budget, competition and other factors. They are examples of past outcomes, not a guarantee of future results.
What we will not do, on this page or anywhere else
These are standing rules rather than preferences, and they apply to our own marketing as strictly as to a client's.
- Publish a number we cannot trace to the engagement or to a named, linkable public source
- Convert a client's directional statement into a percentage
- Claim a business event, such as an acquisition or a retail listing, as a marketing outcome
- Print a star rating or a review count as static text, including our own
- Guarantee a ranking, a lead volume, a timeline or a return
See the full checklist: What we will not do, on this page or anywhere elseHide the full checklist: What we will not do, on this page or anywhere else
- Name a client who has not asked to be named
- Publish review or rating markup built from data we did not collect first-hand
- Use a statistic that failed verification, however familiar it sounds
Want to know what is actually limiting you?
Most of these engagements started by finding the constraint rather than by choosing a channel. That conversation is free and usually short.
Straight answers.
Why are client names not published?
Because they have not asked to be named, and describing an engagement by category and location is enough for a reader to judge whether it resembles their own situation.
It is also the convention the original case pages used, and changing it unilaterally would mean publishing something about a client that they did not agree to.
Some of these case studies have no numbers. Why publish them?
Because they are part of the set. Forty case studies where every single one carries an impressive figure is a selection rather than a record, and a reader has no way of knowing how large the set it was selected from actually was.
The pages without figures still describe the method in full, and the method is the part that transfers. Several of them are among the more useful pages here for exactly that reason.
Did you change any of the original claims?
We restated several that were not measurable. Tripled SEO became the measurable claim underneath it, which was that organic website traffic tripled. Dominating searches within a five-mile radius became ranking consistently across that radius. Loved it and kept it became the two things that can be observed, which are retention and a cost saving.
Nothing was strengthened, nothing was added, and each restatement is explained on the page where it appears.
How long do results usually take?
It depends entirely on which lever is being pulled. Conversion work takes effect as soon as it ships, because it acts on traffic that already exists. Visibility work takes months, because discovery, indexing and ranking take the time they take.
We do not give timelines. What we will do is tell you which of the two your situation calls for, which determines the shape of everything that follows.
Can you guarantee a result?
No, and we would be wary of anyone who does. Rankings depend on competitors who are also working, advertising costs depend on an auction, and enquiry volume depends on demand nobody controls.
What can be committed to is the work: what gets built, what gets measured, how often you hear from us and what the report actually shows.
What if the honest assessment is that we do not need marketing?
Then that is what we will tell you, and it happens. Several businesses on this page were constrained by capacity, by how the phone was answered or by an enquiry process that lost people, and in each case more demand would have made things worse rather than better.
Finding the actual constraint is the first conversation. Sometimes it is not a marketing constraint at all.
Talk to the team
A short call, a look at how the business currently shows up, and a straight answer on what we would do first.