Original research · San Diego County · 2,068 Google Business Profiles
Contractor Google review statistics 2026: 2,068 San Diego profiles measured
By Joshua Lyman, founder, Roscoe Site Pro
- Data collected September 7, 2026
- Definitions frozen September 14, 2026
- Published October 30, 2026
- Updated October 31, 2026 (method and comparisons expanded)
- Next read January 2027
We read 2,068 contractor Google Business Profiles in San Diego County on one day, newest review first. Two out of three had no new review in the last 30 days. The median profile is rated 5.0. The work is rated well. The proof has stopped coming in.
Key findings
- 66.2% had no new review in the previous 30 days; 48.9% had none in 90 days; 27.8% had none in a year. n = 2,068
- The median profile is rated 5.0 and has 16 reviews. Profiles rated exactly 5.0 have a median of 10 reviews; profiles rated 4.8 to 4.9 have a median of 53. n = 2,065 rated profiles; 1,219 and 378
- 48.9% were rated 4.8 or better and quiet for 30 days. n = 2,068
- 41.5% have never replied to a review. The median owner reply rate is 19.5%. Profiles with 100 or more reviews reply to a median 91.0% of them; profiles with 1 to 9 reviews reply to a median 0.0%. n = 2,068; 245 and 790
- The top 10% of profiles hold 68.4% of all reviews in the sample; the top 20% hold 81.2%. 2,065 profiles with at least one review
- In North County, 67.4% were quiet for 30 days and 45.9% had never replied. Fallbrook was the quietest city (85.2%), Vista the most active (59.9%). n = 1,299; cities with 50+ profiles
County-level shares carry about ±2.0 points at 95% confidence (n = 2,068); city rows state their own.
Out of every 100 profiles, this is how many got a Google review in the last 30 days. n = 2,068.
This is not a survey. We opened each profile, sorted the reviews newest first, and recorded what was there. The method, its limits, and how this compares with other studies are at the bottom. The aggregate data is free under CC BY 4.0.
Most profiles have gone quiet
66.2%
had no new review in the last 30 days. 48.9% had none in three months. 27.8% had none in a year.
When each profile last got a review. n = 2,068 measured profiles.
The median profile here is rated 5.0, so this is not a rating problem. 33.8% of profiles got a review in the last 30 days, and recency matters to the people reading: in BrightLocal's 2026 consumer survey, 74% said they look for reviews written in the last three months.1
A perfect 5.0 is usually a small, quiet profile
Here is what the typical 5.0 profile looks like in this market, next to the typical 4.8 to 4.9.
The typical profile in each group. Medians, so half are above and half below. n = 1,219 profiles rated 5.0 and 378 rated 4.8 to 4.9.
Perfect scores are concentrated among small profiles that stopped getting reviews: a median of 10 reviews, the newest one 117 days old, 68.9% quiet. The 4.8 to 4.9 band is the busiest group in the sample: a median of 53 reviews, a fresh one about every month, and the owner replies to 71% of them. A 5.0 built from a handful of hand-picked customers gives a homeowner nothing recent to read, and picking who to ask is against Google's content policy in any case.2
Profiles that reply are the profiles with recent reviews
41.5%
have never replied to a single review. The median profile replies to 19.5% of its reviews.
Smallest and largest profiles side by side. Black: share quiet for 30 days. Orange: median owner reply rate. n = 790 and 245.
This is one day's snapshot, so it does not show that replying causes reviews. What it shows is that in every size group we cut, the two move together: 25.1% of profiles with 1 to 9 reviews have ever replied, against 93.5% of profiles with 100 or more. For comparison, UENI's 2026 study of 229,882 reviews found owners replied to 29.3% of five-star reviews and 9.8% of one-star reviews.3
A few profiles have most of the reviews
All 130,843 reviews on the 2,065 profiles that have at least one, split by who holds them. Top 10% = 206 profiles.
The median profile has 16 reviews; the mean is 63.4, pulled up by a few large profiles. The top 10% of profiles hold 68.4% of the reviews and the top 20% hold 81.2%. Those large profiles are also the ones that reply the most (91.0% median reply rate) and go quiet the least (20.4%).
North County, city by city
1,299 of the measured profiles carry a North County address. 67.4% were quiet for 30 days, 45.9% have never replied to a review, and 45.9% were rated 4.8 or better and quiet (the last two match by coincidence; they are different sets of profiles, see Limits).
Share of profiles in each city with no review in the last 30 days. Cities with 50 or more profiles.
| City | Profiles | Quiet 30 days (±95% CI) | Median days since last review | Never replied | Median rating |
|---|---|---|---|---|---|
| Escondido | 255 | 67.5% ±5.7 | 101 | 46.7% | 4.95 |
| Vista | 187 | 59.9% ±7.0 | 60 | 43.9% | 5.00 |
| San Marcos | 165 | 66.7% ±7.2 | 75 | 40.6% | 4.90 |
| Oceanside | 150 | 70.0% ±7.3 | 106 | 47.3% | 5.00 |
| Carlsbad | 148 | 61.5% ±7.8 | 74 | 37.8% | 5.00 |
| Poway | 99 | 64.6% ±9.4 | 55 | 38.4% | 4.85 |
| Encinitas | 97 | 71.1% ±9.0 | 173 | 49.5% | 5.00 |
| Fallbrook | 81 | 85.2% ±7.7 | 232 | 60.5% | 5.00 |
The ± figure is the 95% confidence half-width for that city's share, in points. City rows under 100 profiles carry roughly ±8 to ±10 points. 611 of the 2,307 attempted profiles (26.5%) had no readable street address and are not in any city row; see the limits below for what that does to the table.
What to do about it
Five things the numbers support. None cost money and none need an agency.
Open your Google profile. Sort reviews by newest. Count the days since the top one. Over 30 puts you with the 66%. Over 90 puts you with the 49%.
Profiles with 100 or more reviews had a median of 5 days since their newest review. Send the review link the day the job is done, and send it to everyone.
41.5% of profiles in this sample have never replied once. Replies are public and stay on the profile.
A perfect score on a profile with 10 reviews and nothing new in 117 days gives a homeowner nothing recent to read. Google's content policy prohibits selectively soliciting positive reviews.2
They hold 68.4% of the reviews. Search your trade and your town, open the profiles with the most reviews, and look at the dates on the newest ones. That is the pace in your market.
How this compares with other studies
Three published studies cover nearby ground. The numbers differ for reasons worth knowing before you quote any of them side by side.
- Smarfle, Google Review Benchmarks 2026 (7,481 businesses, 15 service categories, 25 US cities, June 2026) reports a median of 165 reviews per business.4 Smarfle sampled the top 20 Google Maps results for each search, which is the winners' bracket. Our list includes the long tail of profiles that never rank, which is why our median is 16. Both are right about what they measured.
- UENI, How Small Businesses Respond to Google Reviews (229,882 reviews, 18 months to August 2026) measures replies per review and finds a 29.3% reply rate on five-star reviews.3 We measure per profile, over the newest 100 reviews, and report the median profile's rate (19.5%) and the share of profiles that have never replied (41.5%).
- BrightLocal, Local Consumer Review Survey 2026 (1,002 US consumers) is the demand side: 74% say they look for reviews written in the last three months.1 Our study is the supply side of the same market: 48.9% of contractor profiles had nothing from the last three months to show them.
We know of no other published study of contractor profiles at the county level with a per-city cut. If you know of one, tell us and we will link it here.
Quote this
- Of the 2,068 San Diego County contractor Google Business Profiles measured on September 7, 2026, 66.2% had no new review in the previous 30 days, while the median profile was rated 5.0.
- 48.9% of measured profiles were rated 4.8 or better and had no new review in 30 days.
- 41.5% of measured profiles had never replied to a review; profiles with 100 or more reviews replied to a median 91.0% of them.
- Of the 1,299 North County San Diego contractor profiles measured, 67.4% had no new review in 30 days and 45.9% had never replied to a review.
- The top 10% of contractor profiles hold 68.4% of all Google reviews in the sample.
Method
Where the list of profiles came from
The 2,307 profiles are our own prospecting list of contractor and home-service businesses in San Diego County, built before this study and for a different purpose. 371 rows were matched from California CSLB contractor license records to a Google Business Profile (307 of them returned review data); the rest were found through Google Maps searches for home-service trades in North County cities, run over the months before the study. The final search, on September 7, 2026, covered eight trades (roofing contractor, HVAC contractor, plumber, electrician, tree service, concrete contractor, remodeling contractor, garage door company) in eight cities (Oceanside, Vista, Carlsbad, San Marcos, Escondido, Encinitas, Poway, Fallbrook); the trade table below shows the wider mix the earlier searches produced. The list is weighted toward North County and is not a random sample of all contractors in the county. It is also not the top of a Maps search: it includes profiles that never rank, which is the main reason our medians are lower than studies that sample page-one results.
| Trade, judged from the business name (approximate) | Profiles | Share |
|---|---|---|
| Remodeling, construction, general contracting | 373 | 18.0% |
| Plumbing | 242 | 11.7% |
| Not clear from the name | 223 | 10.8% |
| Electrical | 145 | 7.0% |
| Tree service | 142 | 6.9% |
| Roofing | 125 | 6.0% |
| Landscaping | 107 | 5.2% |
| Heating and air | 98 | 4.7% |
| Pools and spas | 96 | 4.6% |
| Concrete, masonry, paving | 92 | 4.4% |
| Painting | 89 | 4.3% |
| Windows, doors, glass | 75 | 3.6% |
| Garage doors | 67 | 3.2% |
| Handyman | 55 | 2.7% |
| Drywall and stucco | 55 | 2.7% |
| Fencing | 46 | 2.2% |
| Cleaning and restoration | 31 | 1.5% |
| Solar | 7 | 0.3% |
Trade is read from the business name with a keyword list, so these shares are approximate. Profiles whose name does not say the trade are listed as not clear. n = 2,068.
How each profile was read
On September 7, 2026 we requested every profile's reviews through the DataForSEO Google Reviews API, sorted newest first, with a cap of 100 reviews per profile. 245 profiles (11.8% of measured profiles) hit the cap. For each profile we recorded the star rating, the total review count Google shows, the date of the newest review, the number of reviews in the previous 30 and 90 days, and how many of the reviews we read carry an owner reply. Every row keeps the exact Google search URL it was read from, including the locality parameter, so any single row can be checked by hand.
2,068 profiles returned review data. 239 (10.4%) returned nothing: no rating, no count, no reviews. Those are failed reads, not quiet businesses, and they are excluded from every rate on this page. 3 measured profiles have zero reviews and therefore no rating, which is why rating cuts total 2,065 rather than 2,068.
Definitions
- Quiet for 30 days
- The newest review on the profile is more than 30 days old on the read date. Same rule for 90 and 365 days. Because every profile was read newest first, this is exact even for capped profiles.
- Owner reply rate
- Of the reviews we read on a profile, the share that carry a reply from the business. For capped profiles that is the newest 100. "Median owner reply rate" for a group is the median of the per-profile rates. "Never replied" means zero replies among the reviews read; the 3 profiles with no reviews are counted here, which moves the share by under 0.2 points.
- Median
- The middle value: half of the profiles are above it and half below. We report medians because a few very large profiles would distort an average.
- Top 10% of profiles
- The 206 profiles with the most reviews, out of 2,065 profiles with at least one review. Share held is the sum of their review counts over the sum for all 2,065.
- North County
- The city in the profile's address is Carlsbad, Encinitas, Escondido, Oceanside, San Marcos, Vista, Poway, Solana Beach, Del Mar, Fallbrook, Ramona, Rancho Santa Fe, Bonsall, Valley Center, or one of the smaller communities around them. City comes from the postal address embedded in each row's search URL.
Limits
- One day, one county. This is a snapshot from September 7, 2026. It shows what profiles looked like that day; it does not show trends or causes. We plan to read the same list again in January 2027 and publish the change.
- Not a random sample. The list is our prospecting list, weighted toward North County, with a mix of license-record and Maps-search origins. Profiles found through license records were quiet at 62.9% (n = 307); profiles found through Maps searches were quiet at 66.8% (n = 1761). The headline does not depend on which half you look at.
- Hidden addresses. 611 attempted profiles had no readable street address, most likely service-area businesses that hide it. They are counted in every county-level figure but in no city row, so the city table leans toward contractors with a public address. The 533 measured profiles without an address were quiet at 68.1% and had never replied at 36.2%, against 65.5% and 43.3% for profiles with one.
- The 100-review cap and reply rates. For the 245 capped profiles, the reply rate covers only the newest 100 reviews. Owners who reply are likelier to reply to recent reviews than to old ones, so the 91.0% median for 100+ profiles is probably a little high as a lifetime figure. As a control that does not depend on the cap: 93.5% of profiles with 100 or more reviews have replied at least once, against 25.1% of profiles with 1 to 9.
- Definitions were fixed after collection. The data was collected on September 7 and the definitions were frozen on September 14, before analysis for publication. A pre-registered study would be stronger. Freezing the definitions means every figure here was computed one way and will not move.
- Correction. Our first internal pass counted the 239 empty reads as quiet businesses and put the 30-day figure at 70.2%. That was the wrong denominator. The corrected figure is 66.2%, and 70.2% should not be quoted. Corrected September 14, 2026, before publication.
- Coincidences. 48.9% is both the 90-day quiet share and the share rated 4.8+ and quiet 30 days; 45.9% is both the North County never-replied share and the North County 4.8+-and-quiet share. We checked set identity: they are different groups of profiles.
- Out of scope. Nothing here speaks to licensing, complaints, revenue, job volume, or any named business. The data does not contain those fields and we do not infer them.
- Small cities. City rows with under 100 profiles carry roughly ±8 to ±10 points of sampling uncertainty; the ± column states it per city.
Data, integrity, and access
Download the aggregate CSV (CC BY 4.0). SHA-256 of the file as published: e467cb09c9d7cc755739b48c273f629d2e8a88ec3cedc2929f6f67a28fd5b225. The two frozen per-row files it was computed from are hashed 9ab126c574e26b11c6e337b263a9383aa669f7b5af5e69e0f1e2ef1130d19890 and 660cd2a2255926aa6b71bb0823089b129fd741b5fab31a6efff71d84cb771f23; any copy we release will match.
- CSV columns
tablewhich table the row belongs to (headline, days_since_last_review, by_review_count, by_rating, owner_reply_rate, north_county_city, concentration) ·groupthe population or bucket ·profilesthe count of profiles in that group (n) ·valuethe statistic ·metricwhat the value is (for example silent_30d_pct, median_rating) ·notesecondary figures for the same group, in plain text.
The analysis script that generates every table, chart and sentence on this page from the frozen files is published as-is; nothing on the page is typed in by hand. The per-row file (no business names in the public release; names and place IDs under a data agreement) is available to journalists and researchers: [email protected].
Questions people ask
- How many profiles did you measure?
- 2,307 were attempted and 2,068 returned review data. The 239 empty reads are excluded from every rate.
- Why San Diego only?
- It is the market we work in, so we had the list and could check rows by hand. The method runs anywhere, and the January 2027 read uses the same list so the change is measurable.
- Is "quiet" the same as "no customers"?
- No. It means no new Google review in the window. A busy contractor who never asks for reviews is quiet by this definition.
- Why medians instead of averages?
- Review counts are heavily skewed: the top 10% of profiles hold 68.4% of reviews. An average would be pulled up by those few large profiles.
- Can I get the row-level data?
- Aggregates are free under CC BY 4.0. Row-level data is available to journalists and researchers under a data agreement; email us.
- When is the next read?
- January 2027, same list, same definitions, so the change is measurable.
About the author
How we handle reviews for clients
Sources
- BrightLocal, Local Consumer Review Survey 2026, 1,002 US consumers, published February 11, 2026. Accessed October 31, 2026.
- Google, Maps User Generated Content Policy: Prohibited and restricted content, "Rating manipulation": businesses may not "discourage or prohibit negative reviews, or selectively solicit positive reviews from customers." Accessed October 31, 2026.
- Pedro Fonseca, UENI, How Do Small Businesses Respond to Google Reviews? We Analyzed 229,882, updated August 19, 2026. Accessed October 31, 2026.
- Ihor Lavrenenko, Smarfle, Google Review Benchmarks for Local Service Businesses (2026), 7,481 businesses, data current June 2026. Accessed October 31, 2026.
Version history: v1 published October 30, 2026. v2 October 31, 2026: added key findings, sampling frame and trade mix, address and cap controls, confidence intervals, outside comparisons, data dictionary, file hashes and the analysis script; removed pricing from the article body. No figure changed.