Eight Months In: What Changed for a Custom Home Builder Since the Seven-Month Case Study

In September we published the seven-month case study of a second-generation custom home builder and remodeler in a small Wisconsin lake town, and said we would pull the same numbers again every 30 days. This is the first re-pull. Every figure below was pulled on October 7, 2026, the same way, for date ranges of the same length moved forward one month. Google’s search data ran through October 4, so “the latest four weeks” means September 7 to October 4.

One thing first. This is a visibility update. It proves what Google shows, not what the phone did. The owners have not yet told us what the calls from their listing turned into.

The Numbers, Then and Now

Three months against three months (Search Console, US). February through April, the site appeared in Google search 12,510 times and got 359 clicks. In the case study, June through August was 19,315 appearances and 318 clicks. July through September is 18,344 appearances and 344 clicks. The season caveat still applies: February to April is a Wisconsin builder’s slow season and summer is the busy one, so some of the gap is the calendar.

Month one against the latest full month. February: 112 clicks on 2,954 appearances. August: 109 on 6,814. September: 129 on 5,150.

How many different searches. The site appeared for 205 distinct searches in February, 381 in August and 371 in September. On page one: 70 in February, 130 in August, 164 in September.

Page one, latest four weeks, cleaned up. We take every search where the site averaged a top-ten position, then cut other businesses’ names, job-hunting searches and anything the builder does not sell. The case study had 66 on page one, 31 in the top three and 4 in first place. This time 170 searches came back and 105 survived the cut: 44 in the top three and 11 in first place, four of which are the business’s own name.

The searches that matter most.

• “custom home builders near me”, checked from inside their town on a rank tracker: 19th on August 21, first on September 4, still first today.
• “bathroom remodeling + their town”: 5th on 13 appearances in August, 4th on 11 in September. The tracker has it 3rd, unchanged.
• “general contractor + their town”: 2nd in February, 6th in the case study, still 6th in the latest four weeks, on 4 appearances. The tracker has it 4th, unchanged too.

The rank tracker (58 terms, checked from inside their town). On September 4: 19 in first place (16 of them not the business name), 28 in the top three, 36 in the top ten, 18 not ranking. Today: 26 in first place (23 not the business name), 36 in the top three, 41 in the top ten, 15 not ranking.

The map grid. Nine terms, 49 checkpoints each, across an eight-mile radius. The grid kept the same shape between the September 1–2 and October 1–2 readings, so for the first time the two can be compared. Average position, September then October: home addition contractor 14.1 → 15.8, new home construction 15.2 → 15.5, home remodeling contractor 16.6 → 17.7 (the September reading shows as 16.5 today), custom home builder 17.1 → 17.7, bathroom remodeling 17.3 → 17.8, kitchen remodeling 17.7 → 17.7, basement finishing 18.1 → 18.7, general contractor 18.8 → 18.0, deck builder 19.2 → 19.4. That is one month of movement. There is still no reading from before the work started.

The Google listing (from a synced copy of Google’s listing data). August first, as it reads today, then September. Views 516 → 675 (nearly all of that rise is views on desktop Maps, 86 → 280, which we cannot explain yet and are not counting as a win). Search appearances 384 → 354. Interactions 172 → 159. Calls from the Google listing 39 → 23. Direction requests 50 → 57. Website clicks 83 → 79. Google back-fills this data, so August reads a little higher than the case study printed (views 506, interactions 167, calls 38, website clicks 79). January, the month before work began: 330 views, 271 search appearances, 154 interactions, 16 calls, 63 direction requests, 75 website clicks.

Reviews. Still 17, still 5.0, still 100% answered. No new review since May 19.

What Went Up

Clicks to the website moved for the first time: 109 in August, 129 in September. Seventy of the 129 were people typing the business’s name, against 61 of 109 in August. Search Console could tie 9 September clicks to searches that were not the name; the case study counted 4 for August, and today’s re-pull of August shows 3.

On the tracker, four terms have two dated readings that moved up between September 4 and today: “light commercial construction + their town” 2nd → 1st, “home addition contractors + their town” 3rd → 1st, “basement remodeling + their town” 3rd → 1st, and “commercial general contractor + their town” 8th → 4th. Six more that were not first on September 4 are first today.

What Went Down

Appearances in search fell. 6,814 in August, 5,150 in September. Over three months, 19,315 became 18,344. The likely read, not a finding: August is peak building season and September is the start of the slide toward winter, and we have no 2025 data to check that against. We are watching October before calling it anything else.

Three-month clicks are still below where they started: 344 against 359.

Calls from the Google listing fell from 39 to 23. That is still above January’s 16. The likely read, not a finding: fewer people searched in September, so fewer tapped the call button. The listing’s search appearances fell too.

Listing interactions and website clicks from the listing slipped, 172 → 159 and 83 → 79.

The map grid got slightly worse on seven of nine terms, stayed level on one, and improved on one. The builder still averages roughly 15th to 19th across the market. The likely read, not a finding: half-point moves on one monthly reading can be noise, but the direction is wrong, and the listing work has not moved the map yet.

Three tracked terms slipped. “Commercial builder + their town” 1st → 2nd, “home builders + their town” 1st → 3rd, and “local general contractor + their county” 2nd → 3rd.

Fifteen tracked terms still do not rank at all, including every roofing term and every term for the next county over.

“General contractor + their town” has not recovered. Second in February, 6th now.

No new reviews. Reviews are earned on the job. Our part is the replies, the display and the review request tool. As of the case study the request had not gone out, and no new review has arrived since.

What We Are Doing Next

The tracker is empty for roofing and the next county over, so dedicated pages for those are the next build. The map is where this builder is weakest, so the local SEO work stays pointed at the listing. And we still need the one thing only the owners can give us: what the listing calls became. If they would rather have it counted for them, the option is a tracked phone line on the website, the same kind our own AI receptionist runs on. That choice is theirs.

We hold ourselves to the same monthly re-pull in the 60-day case study we ran on ourselves.

Common Questions

Why did clicks go up when the site was seen less often?

Clicks rose from 109 to 129 while appearances fell from 6,814 to 5,150. The likely reason is position. More of the searches the site shows up for are now near the top of page one, 44 in the top three against 31 a month earlier, and people click the top few results far more than the rest. Fewer appearances in better spots can produce more clicks.

Is a drop in calls from the Google listing a bad sign?

It is a number to watch, not a verdict. Calls from the Google listing went from 39 in August to 23 in September, which is still above the 16 in the month before work started. September is past peak building season, and the listing was shown in search less often too. One month does not make a trend. Two would.

Why does the map grid say 18th when the tracker says first?

They measure different things. The rank tracker checks the regular results from one spot inside their town. The map grid checks the map results from 49 spots spread across an eight-mile radius, most of them far from the builder’s address outside town. Being first in one place and about 18th on average across a wide area are both true at once.

What does “page one” count, and why did it jump from 66 to 105?

It counts searches where the site averaged a top-ten spot over four weeks, after we remove other businesses’ names, job searches and services the builder does not offer. This time 170 searches came back and 105 passed. Most have only a handful of appearances each, so a bigger count means wider coverage, not a flood of visitors.

Can you compare the map this month to last month?

Yes, for the first time. The grid kept the same 49 checkpoints and eight-mile radius between early September and early October, so the two readings line up. Seven of nine terms were slightly worse, one level, one better. We still have no reading from before the work began, so we cannot say anything about movement since the start.

Why are there still no new reviews?

Most likely because customers are not being asked. The builder has 17 reviews, all five stars, all answered, and the newest is from May 19. Reviews come from customers at the end of a job. We can reply to them, show them on the website and set up the request, but the request has to go out after each job.

When will you know if this turned into work?

When the owners tell us what the calls became, or when calls to the website run through a line that counts them. Neither has happened yet. Until one does, every update in this series reports what Google shows and how many times people tapped the call button on the listing, and stops there.

If you want the same honest read on your own business, book a free audit or call (920) 383-3387. We will show you what Google shows for you today, including the parts that are going the wrong way.