I Deleted 1137 Photos—And Lost 3.4 Million Views: Here's What I Learned

[Trail_blazer]

Over the last 2–3 days, I deleted 1137 photos from my contributions—and something unexpected happened.

Why did I do this?

I regularly remove low-performing or irrelevant photos. It’s a routine I follow to maintain quality and relevance in my contributions. Sometimes duplicate or off-topic photos get uploaded, and I make it a point to keep my content lean and valuable.

But I also have a long-term goal—which I’ll share when the time is right.

The Deletion Exercise:

I began by sorting my 12,035 photos by view count. I was happy to see that none had fewer than 750 views, which was encouraging.

Then I changed the sorting to “by date”, and that’s where I noticed something interesting.

There were plenty of old photos (ranging from months to even 2 years old) still sitting with just 30 to 250 views. So, I decided to clean up. For example, if a POI (Place of Interest) had 4–5 photos, I kept only the top 1–2 performing ones and removed the rest.

After two days of this cleanup, I had deleted 1137 photos.

I also lost about 5685 points, which honestly didn’t bother me.

The Shocking Discovery: View Count Drop

What did shock me was this:
I lost 3.4 million views.

Most of the photos I deleted had visible view counts of only 50–200. A few had 500 or so. On the surface, it didn’t seem like I was losing much.

Here’s what the math should have looked like:

But the actual loss was 3,400,000 views, which means on average each deleted photo was carrying ~2990 views—not just 50–200 as displayed.

My Theories:

1. The visible view count was inaccurate or outdated.
   The actual number of views was much higher, but not updated on the front end.

2. A few high-performing photos were deleted by mistake.
   I do recall removing 5–10 photos with 10k–30k views, mainly because the place was permanently closed or removed from Maps. But they alone can’t explain a 3.4-million drop.

Final Thoughts:

I also suspect that my top-performing Star photo, which shows 4 million views since early 2024, might not be updating anymore. It’s still the main photo of a popular school, which likely gets 3k–4k views daily.

This experience has made me rethink my strategy.
While quality control is important, I’ll now be more cautious about mass deletions, especially when the actual impact is so unpredictable.

---

Have you ever experienced something similar?

Would love to hear your thoughts.

[Steve_UK]

@Trail_blazer this gives me anxiety (I don’t have so many photos) haha but is very commendable to tidy up and basically only giving google maps what it wants.

The view drop is such an interesting point, and thanks for the sacrifice to find this out
First thought would be extremely time consuming but it would be interesting to add up all the view counts and see how it compares to the total count. it seems likely it would be 10’s of millions different! to tie in with other discussions, it would be less than 1%, but your test suggest otherwise.

[Trail_blazer]

Yes, it does feel like surgery—and I actually perform this cleanup exercise almost every year. But this year was different.

First, the sorting didn’t reveal those low-view (50s) photos like it used to, which is strange. And second, despite similar purges in the past, I’ve never seen such a massive view drop. This time it genuinely caught me off guard.

It also made me rethink how view counts are shown. The face value we see might not reflect the actual count. I strongly believe it’s either underreported or updated very slowly.

A good example: If you search Delhi Public School, Sector 19 Faridabad, the first photo is mine, and it’s been the top photo for over a year. It used to get 3k–4k views daily, but it’s been stuck at 4 million for ages. So where are those daily views going?

Some fellow LGs suggest only unique visitors are counted—but even then, it’s unlikely to be just 10–20 people a day for a place that sees heavy traffic. Earlier, the same photo racked up thousands of views every single day.

There’s definitely more happening behind the scenes, and your idea of adding up all photo views vs the profile total is really tempting—though yes, it would take forever!

Still, all of this makes our contributions and discussions even more interesting, doesn’t it?

[WilfriedB]

Hello @Steve_UK and @Trail_blazer.
Regarding

@Steve_UK

it would be interesting to add up all the view counts and see how it compares to the total count.

Since May 1, 2024 I track the view numbers of all my photos and can ensure you the total number of views is always at least a couple of hours, sometimes several days late. In other words, the sum of all individual views was always a little higher than the total views, I see under My Contributions. Therefore, I assume @Trail_blazer underestimated, how many he did delete.

I never do this kind of cleanup. I just keep my eyes on what becomes hidden and try to understand why that happened. Even in the case of unknown place, I never delete them - I just make sure, there are not too many (currently 43 out of 7407).

Interesting is also to follow up the “semi hidden”, but this is a bit cumbersome.

[Trail_blazer]

@WilfriedB I removed a few photos that were favorited by people but still had zero views and they were months old. How’s that possible?
Being favorited means they were pretty much publicly visible yet nothing on count

[WilfriedB]

@Trail_blazer

that were favorited by people but still had zero views and they were months old. How’s that possible?

@Trail_blazer, clearly, the “one percent rule”: The visible count only changes if the difference to the previous sum is more than 1% (or more than 5 views for the low end.)

For example, if the current number is zero, it will never show 1,2 ..4 views, but only when it is 5 or more; If the current views is 100,000, you see a change only after exceeding 101,000.

[Trail_blazer]

Thanks @WilfriedB went back to your old mega thread the 1 % rule is at comment number 1463 by StephenFink. Quite complex though most can’t make meaningful out of it i guess. Yes I do remember reading all these somewhere on connect here and there. How new algo has changed things. I would repeat my exercise with some more concrete backing next time. But I still have the question if the change doesn’t reach 1% where does that (less than 1% views) stay ? It reflects when it crosses the threshold but before that where does it located ? On the servers? Or its somewhere in our contributions ?

[WilfriedB]

@Trail_blazer

if the change doesn’t reach 1% where does that (less than 1% views) stay ?

@Trail_blazer For sure, it doesn’t get lost, but is further cumulating in the background in some database(s) on some server(s).

[Trail_blazer]

@WilfriedB the 1% percent of my star photo is just 40 thousand. I used to get 4000 everyday I wonder where it’s failing even to get 1300 views a day in order to grow it’s stuck at 40,71,121 total views. It will only update when it breaks the threshold of 40711? Otherwise the count remains as is.

[AdamGT]

@Trail_blazer

I began by sorting my 12,035 photos by view count. I was happy to see that none had fewer than 750 views, which was encouraging.

Then I changed the sorting to “by date”, and that’s where I noticed something interesting.

There were plenty of old photos (ranging from months to even 2 years old) still sitting with just 30 to 250 views.

@Trail_blazer having discovered this discrepancy I would have thought that you would have stopped to investigate why this was so! In any event, without having full knowledge of Google’s data updating cycle, this kind of activity requires careful monitoring and recording of actions vs outcome.

@Trail_blazer

My Theories:

1. The visible view count was inaccurate or outdated.
   The actual number of views was much higher, but not updated on the front end.
2. A few high-performing photos were deleted by mistake.
   I do recall removing 5–10 photos with 10k–30k views, mainly because the place was permanently closed or removed from Maps. But they alone can’t explain a 3.4-million drop.

Yes I agree with these possibilities.

@Trail_blazer

I also suspect that my top-performing Star photo, which shows 4 million views since early 2024, might not be updating anymore. It’s still the main photo of a popular school, which likely gets 3k–4k views daily.

I don’t see that this is relevant to your post.

@Trail_blazer

This experience has made me rethink my strategy.
While quality control is important, I’ll now be more cautious about mass deletions, especially when the actual impact is so unpredictable.

I agree so let me propose what might be a better strategy:

Firstly, let me say that deleting “irrelevant photos”, to use your words, “duplicate or off-topic photos” I think is a good idea as no doubt in time the Maps AI would find and mark these as hidden. However, even though I have done this myself, I wouldn’t prescribe to deleting a large volume of photos at the one time and then checking the impact in terms of total views as you appear to have done. From experience, I believe that doing things in smaller chunks and monitoring is a far better and safer approach.

It sounds like you deleted somewhere between 380 to 570 photos daily over the 2 to 3 days and obviously when you delete photos the overall views will drop but I don’t think you monitored the delta views each day. If you did I would have thought that you would have mentioned so. Anyway, here’s what I think would be a better approach:

1. Instead of doing this once a year, I would do it monthly.

2. Do your T100 System end of month checking and where necessary updating your Maps data.

3. Immediately after step #2, export the T100 System’s Stats Summary and thereby very easily and quickly (within seconds), generate your base data for the month (recording your starting position regarding many important stats including Number of Photos, Total Photo Views, Avg Views/Photo and number of Hidden photos etc.).

4. Delete a much smaller chunk (say 10 to 100) “irrelevant photos” and if you must, note their individual and total views before deletion (a check on overcoming the mistake in your second theory).

5. Any number of times you like through the new month, repeat steps 2 and 3 which should only take a few minutes of your time. The important step here is to check the latest T100 System’s Stats Summary and analyse the results and make sure that everything checks out as you expect. Keep in mind here that factors beyond your control (views and number of hidden photos etc) may change through the month.

6. Repeat steps #2 to #6 each month!

From the facts you presented, you expected to drop between 1m to 1.2m or even 1.3m views but dropped close to 3.5m views. Depending on the timing issue I mentioned above, I believe that the refined approach would have helped avoid this mistake and give you a much more controlled way of performing your cleanup exercise and hopefully you will not be caught off guard.

HTH

[WilfriedB]

@AdamGT

Delete a much smaller chunk (say 10 to 100) “irrelevant photos” and if you must, note their individual and total views before deletion

I agree, @AdamGT and would even go a step further saying “as few as possible, but not more than 20 and only, if you really must do so …”.

[TerryPG]

Great post @Trail_blazer This is something I do on a monthly basis now. Much easier than trying to do it all at once. Coming back tomorrow to ask and add a bit more.

[tony_b]

@WilfriedB

@Trail_blazer, clearly, the “one percent rule”: The visible count only changes if the difference to the previous sum is more than 1% (or more than 5 views for the low end.)

@Trail_blazer @WilfriedB this is what I was going to say, but having read the whole thread first, I see the point is covered.

My take on it is that the views are accumulating in the background long before the display changes. You may have deleted some photos just before their threshold to change display numbers, so what you see was less than what’s on Google’s record for that moment.

BUT it doesn’t seem logical if the photos you deleted all had small counts under 200. There must be a minimum of 5 new views, and 1% of 200 is only 2. So worse case scenario is that you delete a 200 view photo while it was at 204 - before it reached 205. Even a photo showing 1000 views which should change at 1010 - if you deleted it when it was 1009 in the background, the difference is only 9. Not enough to account for your difference of millions.

I did a major purge once, but have been lazy about making it a regular routine. When I next do it again I’ll have to improve my tracking of the relevant numbers impacted.

[Trail_blazer]

@AdamGT

However, even though I have done this myself, I wouldn’t prescribe to deleting a large volume of photos at the one time and then checking the impact in terms of total views as you appear to have done

This isn’t the first time I’ve deleted thousands of photos — it’s probably the third time I’ve done it — but I never experienced a drop in views like this before.

• My photo-per-place ratio was 5.57, which had been bothering me for a long time. Having 13,000 photos spread across just 2,200 places felt like a mini spam issue. At some POIs, I had uploaded 5–6 photos. So, I went back, reviewed those locations, and deleted the ones ranked lower (the ones where views weren’t even visible). I may have lost views in that process. Now, my photo-per-place ratio is down to 3.83. Considering I had 13,000 photos, achieving this ratio was quite a challenge.
  

  

• Why did I upload so many photos in the first place? To get a good position at a particular POI, I try different angles and timing. Some photos perform well, some don’t — so I go back and delete the underperformers.

• My average views per photo shot up from 13k to 15k after doing this clean-up, as shown in the mini stats — but unfortunately, that was temporary. Soon after, I lost a significant number of views — equivalent to what I would gain in about 34 days.

• Last but not least, I’ve crossed Level 10 (100,000 points) almost twice, only to drop back to Level 9 due to deleting underperforming photos. Once you reach Level 10, there’s nothing beyond — whether you’re at 100k or 2 million points, it’s still just Level 10. Hunting for points becomes meaningless unless there’s a greater purpose. I’m still competing in views with people who have 4–5 times more photos than I do. The tools at Top100LG give great insights, and I use them often — but I’m always striving to improve, to match or even surpass those with stronger stats.

[Trail_blazer]

Yes @tony_b that’s most probably is they key answer. However I also went to the locations i contributed tens of photos and chopped down the photos lying at bottom i didn’t know how many views they had but most probably they had a lot then I guessed and this 1% rule seems cumbersome to me. It seems delaying the already delayed updation system. In fact If one has to delete lots of media there is no right time that 1% rule is always holding back some views in at back-end you likely to lose more views than expected irrespective of the time unless you closing monitor all photos and their views change.

[WilfriedB]

@Trail_blazer

I used to get 4000 everyday I wonder where it’s failing

@Trail_blazer some thoughts on this:

• After February 2024, it not only changed when new view numbers are published, but several users, who tracked to total numbers already in 2023 noticed a massive drop in deltas per day. So, if you were used to get 4,000 more daily, you might get 2,000 (as some reported 50% deduction) or even less now (others said -95% which sounds a bit exaggerated).
• As we know, the number of views depend on the position in the thumbs bar. Watching my photos more closely, I realized many movements there: A photo among the first 10 today, can be at the very end tomorrow or even “semi hidden”, i.e. not visible in “All” but still public. The latter, I noticed just yesterday, but for a new place with few views.
• We always tend to think, the views are all gained on the listing where we posted the photo. But this is not true! Many of our photos ore (without our knowledge!) also shown at the pins for the area, city, county, region or even the country. Did you ever check the “By Me”! tab of the pin for your home town or region? If not, have a look and you will be surprised. I believe the majority of high number views come from a prominent position att one of those place and as stated above, the position can and will change eventually.

@Trail_blazer

My Theories:

1. The visible view count was inaccurate or outdated.
   The actual number of views was much higher, but not updated on the front end.
2. A few high-performing photos were deleted by mistake.

Theory 2 seems to me the most likely plus the fact of really loosing track while scrolling down and deleting occasionally. Like @AdamGT, I would also suggest to record the total views before deleting anything and than every single delete, as well as the total views after deleting, but also a day later since the total views get updated not more than twice a day and sometimes even less frequently.

Just my thoughts.

[Trail_blazer]

@WilfriedB

As we know, the number of views depend on the position in the thumbs bar. Watching my photos more closely, I realized many movements there: A photo among the first 10 today, can be at the very end tomorrow or even “semi hidden”, i.e. not visible in “All” but still public. The latter, I noticed just yesterday, but for a new place with few views

Thanks @WilfriedB I get it that there are tens of views how views are affected. Just letting you know that my star photo garnered 3.6M views since 2019 to 2024.
Then I noticed it was dropped from 1st position but for almost a year it’s back to number 1 position while I have many other good goings but nothing beats that.
DPS, Faridabad
Take a look for yourself.
And as far experimenting is considering that’s not likely to stop though I did learn few things so I will change my approach now.
I’m basically concentrating increase my average view per photo and per day. Plus also trying to cover more places.

[TerryPG]

I am glad you did this @Trail_blazer and glad the loss of points did not bother you.
I did my first major purge in the summer of 2023, when I was temporarily , removed from local guides. I like you deleted all my similar photos and especially non performing photos. I lost enough points to move down form level 10 to level 9.

I have for a long time removed low producing photos , but like you I made a bit of an unwise deletion last year, when I removed 12 photos with a million views + each of POIs that were now closed. I lost over 14 million photo views and it took months to regain that.
I agree with what @AdamGT states here

@AdamGT

Firstly, let me say that deleting “irrelevant photos”, to use your words, “duplicate or off-topic photos” I think is a good idea as no doubt in time the Maps AI would find and mark these as hidden. However, even though I have done this myself, I wouldn’t prescribe to deleting a large volume of photos at the one time and then checking the impact in terms of total views as you appear to have done. From experience, I believe that doing things in smaller chunks and monitoring is a far better and safer approach.

It’s much less of a hit to your numbers and I also agree with this statement.

@AdamGT

• Instead of doing this once a year, I would do it monthly.

Every month I check which photos and now videos have crossed the 3 month threshold and I remove everything that doesn’t have a 100 views. I scroll down to 6 months, 9 months and do the same with a different number and scroll down to 1 year. After 1 year my photos need to achieve 3000 views. These numbers are my numbers and different guides can set their own levels .I know @tony_b have had discussions about this over the years. I am a little more lenient since the crash.

Today I updated my photos with over a million views and concur the 1% rule in effect. The old big producers are not moving very fast. Many just don’t or haven’t moved in months. But I am happy to say some of my newer million views are actually climbing over 2% a month, Exciting.
I still go thru all my photos once a year in August, but because I am now doing this on a monthly basics it doesn’t affect the older contributions, and doesn’t hurt as much.

I am still undecided on what to do with closed POI’s photos and on the other hand I still have over 250 photos that are still “uploading” since 2019, but that’s another problem.

Either way Congrats on cleaning up your photos. It’s worth it.

[Trail_blazer]

Thanks @TerryPG I see where I lost the views. Looks like my case is no different than yours. Probelm is once you see those poorly performing photos it becomes sort of OCD to remove them. I’m still confused about such removals in future I will see what can I do. In the hindsight I don’t want thousands of photos just to gain views by sheer numbers of photos I want good quality photos and good average. Yes I have many photos with good views but places closed. I’m keeping them and will take cautious move in future

[tony_b]

@Trail_blazer

I’m basically concentrating increase my average view per photo and per day. Plus also trying to cover more places.

Bear in mind that after February 7th, 2024, the views accrue at a much slower pace. I have a chart with two simple formulas in it to track Views per Photo, and Views per day. From June 2023 to January 2024 my views per photo climbed from 20,263 to 27,829, but from February 2024 until now, there has been a steady decline to the current 22,804. This is because all the new photos I add are getting significantly less views per photo and diluting my overall average. The number of photos doubled in that time period.

At the same time, overall views per day continued to trend upward, because with more photos to view, there will be more views occurring, even if at a very modest rate.

@TerryPG

I know @tony_b have had discussions about this over the years. I am a little more lenient since the crash.

Yes Terry, I did a major purge around the same time as you, but I haven’t been as consistent as you with the follow up. I wanted to do an experiment tonight to cross-check what @Trail_blazer was experiencing, but it is midnight now and my eyes won’t focus on what I wanted to delete.