Real Estate Marketing Case Studies · Raleigh · Lake Norman · Remote
The work, the result, and the part the data cannot prove.
A record of what Oak & Algorithm changed for Realtors, lenders, and builders. Dates, starting conditions, delivered work, measured indicators, and the limits of every claim, all kept in the same place.
Results are not blended together. Completed work is labeled differently from search visibility, inquiries, and revenue.
The Short Answer
What has Oak & Algorithm actually changed?
Website, visibility, content, positioning, and marketing-system work for real estate and mortgage professionals. Two measurements are public and dated. A new Realtor website recorded 544 Google Search impressions and 13 organic clicks in its first four weeks. Oak & Algorithm’s own site recorded 57,767 impressions, 148 clicks, and 2,140 Google AI-feature impressions across three months.
Both are evidence of search visibility, not of leads, closings, or revenue. Separately, 27 people have contacted me since late July 2026, and 23 of them found me because an AI platform named my business. Nineteen of the 27 became clients. Those are inquiries, every attribution is self-reported, and they are labeled that way below.
Read the Labels
What does “result” mean on this page?
Four different evidence levels answer four different questions. They should never be collapsed into one success claim.
Completed work
A page, profile, workflow, campaign, or operating asset was built or implemented. Completion alone does not establish a business outcome.
Visibility indicator
Search impressions, clicks, indexed pages, or profile activity were observed in a dated source. Visibility is not the same as an inquiry.
Inquiry
A call, form, email, or booked conversation is tied to the work through a stated attribution method. Inquiry counts and one client-reported listing are claimed below, and the method behind each is stated plainly, including what it cannot verify.
Revenue
A closed transaction or collected revenue is documented and attributable under a defined method. Search exposure should not be presented as revenue.
Ashley Smith, REALTOR® · Compass North Carolina · Chapel Hill, NC
Can a brand-new Realtor website earn search visibility in its first four weeks?
I built Ashley’s Google presence from scratch, then turned it into a local answer engine. She went from no Google Business Profile to eleven five-star client reviews, and her profile impressions rose 53 percent from July to September 2026. Her site had never had search visibility either. Four weeks after launch, Google Search Console recorded ten pages earning impressions, thirteen organic clicks, and a click-through rate that climbed from 1.44 percent to 5.22 percent inside the same window.
Starting condition
A new domain with no performance history
No prior rankings, no indexed pages, no backlink profile. That makes this an early-visibility example, not a mature SEO trend and not proof of sustained rankings.
Work delivered
A full authority website, built to be found
- Realtor-focused site structure and page architecture
- Service, buyer, seller, and local-market content
- SEO titles, meta descriptions, and internal linking
- Search Console setup, measurement, and first-month reporting
Evidence source
Dated Google Search Console report
A four-page client report and a public case article identify Google Search Console as the source for July 23 through August 20, 2026. Figures below are as reported in that window.
| Page | Impressions | Clicks | What it indicates |
|---|---|---|---|
| Homepage | 397 | 9 | The brand and core service entry point is being surfaced first, which is what should happen in month one. |
| About page | 126 | 4 | People are checking who she is before they contact her. An About page that earns clicks is a trust signal, not filler. |
| Chapel Hill market article | 43 | 0 | Average position 7.0. Local content reached page one faster than the service pages did. |
| Query: “Ashley Smith Realtor” | 36 | 0 | Average position 7.6. Someone else is winning her own name. This is the next fix, not a win. |
Thirty-six people searched her name and nobody reached her site. A new domain does not automatically own its own brand query, and that gap is a specific, fixable problem: Google Business Profile alignment, consistent citations, and a name-matching page. It is listed here because leaving it out would make the report look better than it is.
Google Business Profile and local visibility · May to October 2026
From no profile to eleven five-star reviews and a 53 percent rise in visibility
Ashley had no Google Business Profile before this work. I built it from scratch in spring 2026, then kept it active every week with posts, review replies, and content written to answer what Triangle buyers and sellers actually ask. Google’s own profile data shows what followed.
| Month | Impressions | Change vs. July | What it indicates |
|---|---|---|---|
| July 2026 | 230 | – | The starting line for this comparison. |
| August 2026 | 332 | +44% | The answer-style posting program begins. |
| September 2026 | 353 | +53% | Her best month since the profile went live. |
| October 1 to 27, 2026 | 194 | In progress | Partial month at capture. Not compared, and not projected. |
Reviews
Eleven clients, eleven five-star reviews
All eleven were earned after the profile launched, and they say exactly what her positioning says: repeat clients who bought and sold with her more than once, investors, and buyers relocating to the Triangle. Nine already carry a personal reply from Ashley, which tells Google and every reader that someone is home.
Content shift
From introductions to answers
Launch posts introduced Ashley and the towns she serves. Once the retainer began, the posts switched to answering the real questions buyers and sellers search, written so Google and AI tools can lift them as direct answers:
- How NC due diligence fees and earnest money work
- Why NC closings use an attorney, and how to prevent wire fraud
- Chapel Hill’s creek buffer (RCD) rules before you buy
- Why property tax bills differ on the same Chapel Hill mailing address
- Orange County tax value vs. market value
Cadence
A month ahead, every month
A full month of posts is written and scheduled before it runs, so the profile never goes quiet when Ashley’s calendar fills up. Consistency is the signal. A profile that posts once and disappears tells Google nothing.
In early October, Ashley told me a seller found her and her website on Google and hired her to list the home. That is the step the first-month numbers could not show: a stranger searching, landing on a site that did not exist in June, and choosing her. She approved sharing it.
The Search Console figures and the profile impressions are visibility signals. They are not leads, listings, closings, or revenue, and they are not presented as any of those. The visibility growth and the content shift happened over the same months; no single post is claimed to have caused a specific result, and some of the movement may be seasonal. The listing above is client-reported: the seller told Ashley how they found her, and there is no form submission or tracking record behind it. The evidence does not establish AI citations, and it does not establish that a single page or tactic caused every impression. Average position is an aggregate across a date range, not a permanent rank.
Jennifer Cain, REALTOR® · LPT Realty · Raleigh & Cary, NC
What happened when Jennifer Cain’s first lead funnel went out with no ad spend?
Jennifer Cain is a listing agent in Raleigh and Cary, brokered by LPT Realty, and she had never had a website. I built her an AI-optimized one aimed at exactly the sellers she serves, then a lead funnel built around what those sellers were asking that agents in her area were not answering. The funnel went out as one unpaid post on her business Facebook page. Within hours, two people had asked for her guide and left their name, email, and mobile number. The post was boosted only after that. Here is the story in her words.
Starting condition
No website, and no idea where to start
Jennifer had never had a website. In her words, marketing herself was complete overwhelm: shiny objects, other agents’ YouTube channels and Facebook posts, and no way to tell what worked. She got into real estate to serve clients, not to become an online marketer.
Work delivered
A website for her niche, then a funnel
- An AI-optimized website written to her ideal client
- Pages she can send prospects, like her Google reviews page and guides
- A lead magnet built on questions local agents were not answering
- A funnel that notifies her the moment someone opts in
Attribution method
She described it herself, on camera
The inquiry count and timing come from Jennifer, in a recorded interview published November 2, 2026. The funnel form captured each person’s name, email, and mobile number and notified her immediately so she could follow up directly.
“Emily looked out there and saw what my clients are asking for that agents in the area aren’t addressing on their websites or in the content they’re posting.” Jennifer Cain, on why the funnel worked
Read the transcript
What was the hardest part of marketing yourself on your own?
The hardest part of marketing myself on my own was not knowing what to do. It is complete overwhelm out there. I don’t know what I don’t know. You get this shiny object syndrome, and you start watching everybody’s YouTube channels or seeing their Facebook posts. Then you’re following all these different rabbit holes, and you have no idea what works, what doesn’t work, how to get started, or how to optimize it. I did not get into real estate to be a master online marketer. So it was just complete overwhelm.
How do you feel about your marketing and systems now?
We’ve just started. It’s only been two or three months. I never had a website before, and now I have a website, and it’s not just a regular website. It is optimized for AI. Again, things I don’t understand how to do. And it is geared specifically to speak to my niche market, my ideal client. Again, not an area that I know how to do.
So now I have a website that I’m comfortable and confident to send out into the universe, to clients and prospective clients, and say, “Hey, here’s a little bit about me.” Or I’ll just send a page of that website. Maybe it’s the Google reviews page or a guide. Something that shows them, okay, this person is professional. She knows what she’s talking about.
And now we’re layering in the optimization of that. Again, not my technical area of expertise. And we’re just starting the lead magnets and funnels. So I feel like I’m finally getting a system in place that’s going to stop this roller coaster effect in my business, where I spend all this time cultivating clients, I serve those clients, I do all this work, and then after all that time and energy, I have to start all over again. I needed something running in the background. Professional, clear, easily defined, and speaking exactly to who my client is. And that’s what Emily has been able to create for me.
What happened with your first lead funnel?
Emily put together my first lead funnel. I’ve never done any of this on my own. And it wasn’t a paid or boosted one at all. It was just a lead funnel that I put out on my business Facebook page. Within hours, I had two people asking for information. They gave me their name, their email address, and their mobile number, and I was notified immediately so I could follow up with them directly. With that kind of instant effect, we said, okay, let’s boost this.
The reason I think it works so well is that Emily looked out there and saw what my clients are asking for that agents in the area aren’t addressing on their websites or in the content they’re posting. So I’m speaking directly to my clients, and that’s what Emily has created for me.
What would you tell an agent who is thinking about working with Emily?
If you’re on the fence about whether or not to work with Emily, just make the phone call. Get the information. I think what you will find is that there are layers upon layers of technology, marketing tactics, and strategies that you’re just not tapping into, and you are missing the boat. If you don’t get in now, you’re going to be behind, because others will use this, and you want to stay ahead of that. You want to be the go-to person. So just have the conversation. Talk to her about your pain points, your strengths, and what you’re trying to create, and let the two of you collaborate on how to get ahead of your competition.
Transcript lightly edited for readability.
This is one funnel, one post, and one agent’s account of it. Two inquiries are inquiries, not appointments, listings, or closings, and none of those are claimed here. The count and the timing are client-reported, and no form log or analytics record is published behind them. The post was boosted after the first two inquiries, and nothing on this page describes what the boosted run produced. Individual results vary.
Oak & Algorithm · June 20 to September 19, 2026
Does the company selling visibility actually have any?
Fair question, and most agencies never answer it. Here is the full Search Console record for my own business across three months, including the number that does not flatter it, followed by what changed after a full rebrand: ChatGPT now names this company in 98 percent of its answers for a search where it named it in 25 percent sixteen days earlier.
Starting condition
A business that opened in August 2025
About a year old when this window closed. No domain history, no backlink profile worth the name, no legacy rankings, and a category already crowded with national agencies spending real money on ads.
Work delivered
The same stack I sell to clients
- Google Business Profile optimization and local SEO
- An authority website built around questions people actually search
- Service, market, and comparison pages structured for AI answer engines
- Weekly content and repurposing, published without gaps
Evidence source
Google Search Console, exported September 21, 2026
My own Performance and Generative AI features reports for the ninety-day window ending September 19, 2026, compared against the prior ninety days.
The numbers below end September 19. Two days later this company changed its name and its domain, which is supposed to reset every trust signal a business has. Here is what the record shows since:
- ChatGPT went from naming Oak & Algorithm in 25 percent of its answers to 98 percent, ranked first, for “ai training for real estate teams raleigh.” Same search, same grid, October 10 to October 26. See Case 08.
- ChatGPT named this company at all 49 of 49 grid points for “who helps realtors with google business profile in raleigh” on October 19, and Google Maps had it in the top three at 46 of 49 points for the matching search on October 26.
- 23 of the 27 people who reached out to hire me since late July found me through an AI platform, and 19 of the 27 became clients. See Case 04.
- Google’s AI and ChatGPT both named the business first in cold tests two weeks after the rebrand. See Case 06.
- From Raleigh, 10 tracked searches rank in the top three and 16 in the top ten on the new domain. Semrush Position Tracking, desktop, 58 keywords tracked since the rename, October 28, 2026. It is a different keyword list from the 181 below, so the two visibility percentages are not compared.
| Signal | Figure | The honest reading |
|---|---|---|
| Average national position | 27.4 | Page three for most searchers in the country. Page one from Raleigh, where Semrush puts me second behind Reddit. Both are real, and the national one is the one that limits traffic. |
| Average click-through rate | 0.26% | Two things caused this, and only one of them is about rank. See the note below. |
| Weekly AI-feature impressions by September | 180–240 | Growth from roughly zero in late June. AI answer engines began surfacing the site inside a single quarter. |
| Queries with impressions | 1,000+ | Across 240 pages. Breadth arrived before depth, which is the normal order and the reason the click number is still small. |
Measured nationally, month by month
Zero ranking keywords in October 2025. 229 by August 2026.
| Month | Keywords |
|---|---|
| Oct | 0 |
| Nov | 3 |
| Dec | 3 |
| Jan | 6 |
| Feb | 13 |
| Mar | 49 |
| Apr | 66 |
| May | 73 |
| Jun | 130 |
| Jul | 233 |
| Aug | 229 |
Source: Semrush Domain Overview history, Google US database, monthly snapshots, October 2025 to August 2026, pulled September 22, 2026, for my previous domain before the rename. Unlike the Raleigh report below, this counts every keyword Semrush found the site ranking for nationally, not a list I chose. The AI Overview figure counts keywords whose results include an AI Overview; it does not mean my site was cited in each one.
Measured from Raleigh
Source: Semrush Position Tracking, Landscape report, organic, desktop, Raleigh, North Carolina, local volume, August 31 to September 22, 2026. Tracked on my previous domain, before the rename. This is a Raleigh measurement, not a national one, and Semrush visibility is calculated only across the keywords I chose to track.
I read that number as a ranking problem for months. Page three nationally, so nobody clicks. That is part of it and it is real. It is not the bigger part.
The bigger part is the two numbers directly above this note. Second only to Reddit, nearly three times Luxury Presence, forty-two keywords in the top three, and an estimated traffic of 0.05. Both are real. They are measuring different things.
I had been tracking the same keyword list since the month I formed the business. I ranked first for “AI visibility for loan officers” and “real estate marketing services Lake Norman.” Those are good terms and I would choose them again. Almost nobody types them. And much of the rest of the list was informational: terms people search when they want to learn something, not when they want to hire someone. So the visibility was real, the traffic was real, and it was made of readers. A dashboard that looked healthy while the pipeline it was supposed to feed came from somewhere else entirely.
That is the failure mode I now look for first in every audit, because it is invisible from the inside. A business can rank well, measure diligently, and still be measuring the wrong thing for a year. Mine did.
Since late July 2026, 19 people became clients. Seventeen found me through an AI platform and two through Google. Not one came from a referral. That is first-party and self-reported, it is about my own business, there is no third-party verification behind it, and it belongs in the inquiry column rather than the revenue column. It is here because it is the reason the visibility work is worth buying.
Impressions are exposure. Clicks are visits. Neither is revenue, and none of these figures are presented as a closed transaction. A 0.26 percent click-through rate is a weak number and I printed it on purpose, along with what caused it. This is my business in one category, and it does not predict what yours will do.
Oak & Algorithm · Late July to October 28, 2026
What happens when AI tools start naming you instead of someone else?
Since late July, 27 people have reached out to hire me. Twenty-three said an AI platform gave them my name. Nineteen of the 27 are now clients. One of them, in September, was a high-production Realtor in another state, roughly 900 miles from here, replacing the national marketing company he had used for years. He described his problem to Google Gemini, asked who he should talk to, and it gave him my name alongside a handful of others.
How they found me
They asked an AI, not a search engine
No referrals. No paid ads, because I do not run any. Each of the 23 described a problem to Gemini, ChatGPT, or another AI tool and contacted the names it returned. Not one of them had heard of me first.
What made that possible
The same work I sell to clients
- Identical business information everywhere my business appears
- Pages structured as a question, a direct answer, then the explanation
- Content written to answer what the market searches, not to fill a calendar
- Everything published on an owned domain, not a rented feed
Attribution method
They said so, in their own words
Each person told me how they found me, in their own words, during the first conversation. There is no tracking parameter behind any of it, because AI referrals rarely leave one. That is a real limitation and it is why this is labeled self-reported rather than measured.
Different states, different price points, different problems. Loan officers, agents, builders, a lodge. Some named the platform, usually ChatGPT or Gemini. Others just said “AI.” The only thing they share is that a machine said my name before a human ever did. The other four inquiries came through Google, two, and social media, two.
Because of the calendar, not the work. This business opened in August 2025 and engagements run six months and longer, so almost every client is somewhere in the middle of their agreement rather than at the end of it. Referrals tend to arrive after someone has watched a full cycle finish and can point at a result, not during month three. I expect that number to move over the next six months as current engagements complete and clients offboard. Until then, zero is a fact about how young the company is and how long the work takes.
Traditional search rewarded being findable. Ten blue links, and the searcher decides. AI search rewards being nameable. The platform returns two or three names and there is no page two. A business that is not in that answer was never in the running and will never know it happened. Those are different problems, they need different work, and the second one is the one almost nobody in real estate is doing yet.
These are inquiries and signed clients, not revenue. The people behind the count are not named here, because the count is about where they came from, not about their results. It comes from my own records and every attribution is self-reported, so I cannot verify any of it in an analytics tool. Not all 27 became clients: 19 signed, two chose not to, and six are still in conversation, with second meetings scheduled or a decision pending. Nothing here says the same prompt returns the same names tomorrow. AI platforms change their answers constantly.
Jeff Peterson · Raleigh, NC
How do you make a Realtor’s relocation outreach useful before asking for referrals?
The strategy shifted from generic “send me Raleigh referrals” outreach to a specific Triangle golf-community resource that could support a real conversation with agents and relocating buyers.
Starting condition
Good agent, weak differentiation
The existing message did not give referral partners a memorable reason to respond, save the resource, or connect Jeff with a relocating client.
Work delivered
A niche referral asset and supporting system
- Triangle golf-community website and market resource
- Relocation-focused messaging and positioning
- Domain, email, profile, and outreach preparation
- Ongoing-content framework around the niche
Documented change
The outreach had something useful to lead with
The completed strategy replaced a generic referral request with a concrete market resource. That is a positioning and delivery result. No inquiry or revenue figure is asserted.
This campaign was already described publicly with permission. It is presented here as a historical project snapshot, not as evidence of a current engagement. The reviewed material did not establish attributable inquiries, closed referrals, or revenue.
Oak & Algorithm · October 6–7, 2026
Can a rebrand that is two weeks old survive AI search?
A business name change is supposed to be the most visibility-destroying thing a company can do. New domain, new entity, every trust signal reset. Barely two weeks after this company's rebrand went live, I ran cold tests in incognito windows: no login, no history, no brand names in the first query. Google's AI returned a table of recommended AI visibility partners for Raleigh Realtors with this company in the first row, and when asked a two-word follow-up, "Oak & Algorithm?", it replied "Good catch, you hit the nail on the head" and explained the rebrand unprompted.
The test
An unbranded question, asked the way a buyer would ask it
The first query named no companies: "I am a Raleigh, NC real estate agent looking for a company to help improve my visibility in AI." A separate branded comparison query was run the same night, and the equivalent question was put to ChatGPT in a fresh session. Every result was captured with its timestamp and query visible.
What came back
First row, then a confirmed identity
Google's AI listed this business first among its recommended partners, built comparison tables against national platforms with this company's real published pricing in its row while competitors showed "quote-based," and correctly connected the new name to the old one. ChatGPT independently produced a due-diligence analysis using the same pricing, positioning, and review details.
Why it happened
The engines cited this website
In the fullest result, six of eight citations were pages on this domain: services, pricing, about, and a category comparison page written here. Published prices filled the table because they exist to be read. The "formerly" entity work connected old name to new. Machines quoted the structure they were given.
ChatGPT's analysis found the disclaimer on this site stating that no provider, this one included, can guarantee AI rankings or placement, and cited it as "a point in its favor." The honesty was retrievable, and the machine scored it as trust. That is the clearest evidence I have that publishing plainly beats promising loudly.
Two days later I ran it again, this time in logged-out ChatGPT, fresh incognito windows, no history. Asked three different ways for marketing help for Raleigh real estate agents, and once for a fractional marketing director for a real estate business, ChatGPT named this company first in all four answers, against a different set of competitors every time, with current published pricing quoted correctly. Asked what happened to Real Estate Concierge Services Co, it answered that the company had rebranded to Oak & Algorithm and had not shut down. The same session found the gaps too: no mention for team AI training, custom home builder marketing, or general small business marketing. Those misses are as useful as the wins, and they are already being worked on.
These are dated samples from one night, not a standing rank. AI engines resample their sources constantly; published research puts repeat-query consistency for local business answers near 7 percent, so the same question on another day can return different names. Claims inside AI answers, such as references to "independent analyses," are machine synthesis, not citations of real third-party studies, and are not repeated here as fact. The honest claim is narrower and better: on dated, replicable cold tests two weeks after a rebrand, multiple engines retrieved, recommended, and correctly explained this business from its own published pages. The same tests re-run in January 2027, and the results will be added to this page either way.
Ken Jansen · Mooresville, NC
What happens when a referral-only agent builds a search presence from zero?
A licensed Lake Norman agent with three years in the business, 8 to 10 closings a year, and every one of them from referrals. No Google Business Profile worth the name, no blog, no relocation content, and no presence in AI answers. Twelve weeks later, four closings traced back to the new presence, including one from a relocating buyer who found him through ChatGPT.
Starting condition
Good agent, invisible online
Three years licensed in Mooresville. Closing 8 to 10 deals a year, entirely referral-dependent. When a buyer or relocator searched the market, he was not in the answer, on Google or in AI tools.
Work delivered
Three pillars over twelve weeks
- Google Business Profile built out, with a review request system he used every week
- A seven-page relocator hub for people moving to Mooresville and Lake Norman
- A weekly blog, each post answering one question buyers actually ask
- Structured, question-first pages built to be cited by AI tools
Documented change
Four closings from the new presence
By week twelve, 6 of the 8 AI queries tracked from day one cited Ken or his content. Four closings in the twelve weeks came from the new presence: two from Google Business Profile inquiries, one from a blog post, and one from a relocating buyer who found the hub through a ChatGPT recommendation.
Ken approved publishing these figures. The commission total is an estimate built from his average commission, not from settlement statements. Each closing's source was reported by Ken from what his clients told him, and the ChatGPT-referred buyer is self-reported, because AI referrals rarely leave a tracking trail. The two AI queries that did not cite him were dominated by large national publishers. This work was delivered under the company's former name, Real Estate Concierge Services Co, LLC. One agent, one market, one twelve-week window: it shows what the system did here, not a promise of the same result elsewhere.
Oak & Algorithm · October 10 to 26, 2026
What can sixteen days of visibility work change for one search?
On October 10, ChatGPT named Oak & Algorithm in 25 percent of its answers for “ai training for real estate teams raleigh,” and Google Maps did not put me in the top three anywhere on the grid. On October 26, with the work below running, ChatGPT named me in 98 percent of its answers and ranked me first. No ads. Same search, scanned the same way, sixteen days apart.
Starting condition
A real service, invisible for its own search
The AI training page existed. Gemini, Google AI Mode, and Google AI Overviews did not name this company for it, and on Maps I sat outside the top three at every grid point.
Work delivered
The same system I run for clients
- A service page built around the questions teams actually ask, each answered directly
- A Google Business Profile kept current with services, posts, and reviews
- The same business information everywhere the company appears
- Dated proof published on the page, so engines have something specific to quote
Evidence source
Two dated Local Falcon scans
A 5-mile grid centered on North Raleigh. October 10 used 25 grid points and October 26 used 49, so the percentages compare and the raw counts do not.
For “who helps realtors with google business profile in raleigh,” ChatGPT named Oak & Algorithm at all 49 grid points on October 19. On October 26, Google Maps had me in the top three at 46 of 49 points for “google business profile optimization for realtors raleigh.”
These are dated scans, not a standing rank. AI answers resample their sources every time they run, and nobody controls whether a platform names a business tomorrow. Maps still has four businesses ahead of me on this search, and Google AI Mode and AI Overviews are not naming me for it yet. Visibility is not an inquiry, and none of these figures are presented as one.
Open Measurement
What is being measured right now?
These engagements have baselines captured and review windows running. None of them appear above yet, because the numbers are not finished or the client has not approved a public version.
A Virginia visibility foundation
Google Business Profile audit, weekly blog publishing, and repurposing. Baseline captured. First review compares the same definitions and the same date length.
A Raleigh listings and builder brand
Content running across three audiences at once: consumers, agents being recruited, and builder partners. Measured on publishing consistency and profile activity before anything else.
AI Operating System implementations
Measured on hours recovered and publishing consistency, not on impressions. An operating system that saves six hours a week is doing its job even if search never moves.
A coastal North Carolina foundation
A Visibility + Authority Foundation that started August 11 and finishes in November. The end-of-service report compares the profile, the AI answers, and the inquiry sources against the starting point.
Proof Standards
What will not be dressed up as a result?
Good marketing evidence is specific enough to inspect. When the record stops, the claim stops too.
A deliverable is not a lead.
A completed website, profile, workflow, or campaign proves that work was delivered. It does not prove a buyer contacted the business.
An impression is not revenue.
Search impressions show exposure. Clicks show visits. Revenue requires a separate client and transaction record.
A snapshot is not an active-client roster.
Historical work remains useful evidence without implying that every named person is a current client.
A private claim is not public proof.
Named recurring-engagement stories are published only after permission, dates, scope, outcomes, and limitations are approved.
Questions Before You Hire
How should you evaluate these case studies?
Do these results guarantee the same outcome for another Realtor, lender, or builder?
No. Market competition, starting condition, access, approvals, website history, budget, and continued execution all change what happens. These examples show the work Oak & Algorithm delivered and the documented observations for those specific situations, nothing more.
Why does Ashley’s first-month case report visibility instead of leads?
Because the Search Console report documents impressions, clicks, average position, and visible pages. It does not document calls, forms, appointments, closings, or revenue. The listing she won later from a seller who found her on Google is reported separately, as client-reported, and is not mixed into those numbers. The label follows the evidence, not the other way around.
Why publish your own numbers when they include a 0.26 percent click-through rate?
Because a marketing company that only shows the flattering half of its own report should not be trusted with yours. The 57,767 impressions and the 0.26 percent click rate are the same report, and the gap between them turned out to be the most useful thing in it. Part of it is rank: page three nationally does not get clicked. The larger part was that I had been measuring the wrong keywords since the month I started, so the traffic I was earning was people reading rather than people hiring. Finding that is worth more than the flattering number would have been.
People find you through AI tools. Is that repeatable, or is it luck?
Once is luck. Twenty-three times in three months, across different platforms and different states, is a pattern. What produced it is not mysterious: consistent business information everywhere my business appears, pages structured so a machine can read the question and the answer directly beneath it, and content published on a domain I own instead of a feed that disappears in a day. That work is repeatable. A specific result on a specific day is not, and AI platforms change their answers constantly.
Zero referrals. Should that worry me?
It is a fair thing to notice, and the honest answer is that it is a calendar problem rather than a quality problem. The business opened in August 2025 and engagements run six months or longer, so most clients are mid-agreement rather than finished. People refer someone after they have watched a full cycle play out and can say what changed, which is not something you do in month three. I expect the number to move over the next six months as current work completes. What zero referrals does prove right now is that none of this pipeline came from a network. It came from being findable and nameable.
Why are the people behind the inquiry count not named?
Because some are prospects rather than clients, and the count is about how they found me, not about their results. The clients named on this page approved it. The same standard applies everywhere: nobody is named until they approve the scope, the dates, and the limits.
Why is the Jeff Peterson example called a historical snapshot?
The work and the strategy were already described publicly with permission, but the reviewed evidence does not confirm current engagement status on this page. Calling it a snapshot keeps the useful delivery example without presenting an old status as a current one.
Where are the recurring fractional-marketing results?
Waiting on approval. A named recurring-engagement story is published only after the client signs off on the public scope, the dates, the evidence, the outcomes, and the limitations. Until then, the fractional partner page explains the operating model without turning private client records into public claims.
Did the company name change affect any of this data?
No. Oak & Algorithm was formerly Real Estate Concierge Services Co, LLC, renamed in September 2026. Same company, same founder, same clients, same records. The Search Console history moved with the domain, and every figure on this page was recorded by the same business under one name or the other.
What should be measured before a new engagement begins?
The baseline has to match the work: profile condition, website visibility, indexed pages, publishing cadence, inquiry sources, CRM status, and any revenue measure the client can actually attribute. The first review then compares the same definitions across the same length of time. A baseline captured after the work starts is not a baseline.
Start With the Evidence
Find out what needs to change before choosing the scope.
The Visibility Audit is the diagnostic starting point. A fit call is better when the need is already clear and you want to discuss a website, a foundation project, or ongoing fractional support.
No ranking, inquiry, closing, revenue, or AI-recommendation outcome is guaranteed.