I was part of a discussion on UtterAccess in mid-January which got hijacked a bit about the decline of Access. As George stated it:
The discussion made me look further into a couple of my previous posts, such as:


and made me want to examine whether there was/is a quantifiable decline in Access interest.
Stacked Exchange (SE) provides access to their data and they are one of the biggest forum providers in the world. So, I set off to see what I could find out from their data.
Using the Stacked Exchange Data Explorer and the documentation
it didn’t take long to get a clear picture of Access’ rise and fall.
The Stacked Exchange Access Statistics
I built a simple query to determine the number of questions posted per year and the results were
| Year | Count | % Change |
| 2008 | 335 | — |
| 2009 | 1599 | 377% |
| 2010 | 2138 | 34% |
| 2011 | 3015 | 41% |
| 2012 | 4187 | 39% |
| 2013 | 6423 | 53% |
| 2014 | 6576 | 2% |
| 2015 | 6678 | 2% |
| 2016 | 6854 | 3% |
| 2017 | 6104 | -11% |
| 2018 | 5241 | -14% |
| 2019 | 4481 | -15% |
| 2020 | 3499 | -22% |
| 2021 | 2726 | -22% |
| 2022 | 565 | — |

So we can clearly see a yearly increase through 2013 and things started to stagnate for a few years, then start to decline around 2016 and consistently every year thereafter.
The Bigger Picture
This however got me wondering about the more general picture of databases, not just Access. So I set out to review more well known databases to see how Access fared in the grander scheme of things. So I started running the same query for other common database and I got the following picture.
| Year | ms access | sql server | mysql | postgresql | oracle | azure sql database | db2 | nosql |
| 2008 | 335 | 2444 | 1159 | 177 | 674 | 0 | 56 | 4 |
| 2009 | 1599 | 11506 | 9437 | 1012 | 2888 | 22 | 236 | 49 |
| 2010 | 2138 | 16709 | 21705 | 2248 | 4559 | 67 | 374 | 531 |
| 2011 | 3015 | 22634 | 41218 | 3946 | 7411 | 228 | 664 | 882 |
| 2012 | 4187 | 28809 | 57938 | 6287 | 10603 | 454 | 804 | 1208 |
| 2013 | 6423 | 37485 | 74687 | 9104 | 14264 | 521 | 1201 | 1097 |
| 2014 | 6576 | 38461 | 80302 | 11461 | 16216 | 442 | 1225 | 1105 |
| 2015 | 6678 | 38905 | 76878 | 13330 | 16586 | 763 | 1226 | 1216 |
| 2016 | 6854 | 39027 | 72032 | 15036 | 16318 | 838 | 1199 | 1378 |
| 2017 | 6104 | 37681 | 65215 | 16041 | 16192 | 967 | 1209 | 1046 |
| 2018 | 5241 | 31518 | 49502 | 15310 | 15117 | 1096 | 1332 | 1036 |
| 2019 | 4481 | 27564 | 40080 | 15922 | 13868 | 1012 | 1210 | 797 |
| 2020 | 3499 | 23456 | 37006 | 19785 | 14200 | 1046 | 1074 | 992 |
| 2021 | 2726 | 19047 | 31994 | 21341 | 12482 | 952 | 914 | 1102 |
| 2022 | 565 | 391 | 691 | 388 | 213 | 22 | 22 | 24 |
Looking at the bigger picture, we can see a common decline amongst many of the big databases with the major exception of PostgreSQL!
I wasn’t surprised to see MS Access & SQL Server decline, but thought that Azure SQL Database questions would have significantly increased over the same period, which simply is not the case.
Isladogs’ (Colin’s) Data
If you review the original thread I mentioned at the beginning of this article, Colin posted the statistics (taken from https://www.utteraccess.com/topics/2062462/posts/2792955) from another forum.
| Year | Count | % Change |
| 2003 | 8551 | — |
| 2004 | 10778 | 26% |
| 2005 | 10203 | -5% |
| 2006 | 12361 | 21% |
| 2007 | 11101 | -10% |
| 2008 | 12479 | 12% |
| 2009 | 9632 | -23% |
| 2010 | 9015 | -6% |
| 2011 | 8315 | -8% |
| 2012 | 6631 | -20% |
| 2013 | 3755 | -43% |
| 2014 | 3606 | -4% |
| 2015 | 3357 | -7% |
| 2016 | 2019 | -40% |
| 2017 | 1991 | -1% |
| 2018 | 723 | -64% |
| 2019 | 687 | -5% |
| 2020 | 423 | -38% |
| 2021 | 328 | -22% |
Once again, we see the same downwards trend, but in this instance it appeared much earlier. As of 2008, we start to see a consistent decline.
Other MS Office Apps?
I was interested in also comparing Access’ trend against the other Microsoft Office Applications and so I pulled the data for each of them, giving:
| Year | ms-access | excel | word | outlook | powerpoint |
| 2008 | 335 | 435 | 144 | 436 | 50 |
| 2009 | 1599 | 2188 | 555 | 1465 | 357 |
| 2010 | 2138 | 6527 | 8333 | 2149 | 803 |
| 2011 | 3015 | 6117 | 1123 | 3317 | 915 |
| 2012 | 4187 | 10605 | 1434 | 4363 | 1403 |
| 2013 | 6423 | 18224 | 1776 | 5826 | 1965 |
| 2014 | 6576 | 23444 | 1672 | 6768 | 1768 |
| 2015 | 6678 | 29485 | 1870 | 8939 | 1984 |
| 2016 | 6854 | 32088 | 1966 | 9081 | 1822 |
| 2017 | 6104 | 35001 | 1886 | 9120 | 2091 |
| 2018 | 5241 | 30987 | 1948 | 8874 | 1831 |
| 2019 | 4481 | 28705 | 1836 | 8093 | 1797 |
| 2020 | 3499 | 25176 | 1712 | 7574 | 1976 |
| 2021 | 2726 | 23512 | 1551 | 7331 | 1853 |
Graphically we can see a noticeable decline in both Access and Excel.
Empirically, we see a decline across the board for all application starting around 2017-2018. The real question is why? What does this mean, if anything?
What’s Your Opinion?
So, are these trends telling in your opinion? Are they indicative of a shift away from certain technologies? Away from Microsoft Office?
OR
Is it normal to see questions decline since so many online resources now exists and people simply don’t need to ask as many questions?
OR
Are people simply using different forums and these forums are simply loosing their popularity?
More To Come, Maybe
I’ve asked UtterAccess if they’d share their yearly post counts with me as I think it would offer an interesting perspective, see if the same tendency is true there, or not. I know that I’ve noticed a decline, but is it of the same magnitude as what is observed on SE?
I’m still waiting to hear back, but will update this post should they be willing to share such information.
I used the SE data explorer to compare Access related questions in relation to all questions asked in a given year. It seems Stackoverflow has seen less questions overall after approx. 2016. I speculate that many questions that are relevant, now simply already have a good answer there, so maybe less Qs need to be asked overall. For technologies like VBA or Access, that see no updates, this might be true even faster than other ones.
Thank you for that added piece of the puzzle!
That said, I just crunched the numbers, and you are correct there is a decline starting around 2016 and it even had a gain in 2020, but it is still far less pronounced that the Access decline. This is all very interesting.
Nice info Dan!
It is also a good idea if you can include the age bracket (demographic) of those who are still developing serious application in MSAccess.
We can make a long list of complains that MSAccess is outdated, buggy, not suitable for serious db development so on and on, etc..Despite its blisters there is NONE that can take its place yet for desktop db front end development tools.
Jam.py Application Builder is quite similar to what MSAccess can do but it’s not yet there. MSAccess forms/reports/queries/vba are so RAD and powerful, using it’s UI as front end to sql server or any relational db server MSAccess becomes a commando in desktop front end data processing war.
(WebAssembly: Native desktop apps are dead – long live native desktop apps!)
I’m hoping in the near future somebody can can up a tool to compile MSAccess UI into a WASM or something like a virtual MSAccess runtime bundled with a browser.
Coding since Access beta and I’ve never had to ask any new questions, I always find a useful answer there. The fasted programming method is to re-use other people’s code
Maybe the lack of new questions is a sign of long term stability and the exhaustion of questions to ask.
I’d hazard a guess part of the overall decline is the fact of the increased history of accessable answered questions on places like stack overflow (for all databases).
Most people only post a question after having done a fairly extensive search to see if a solution or discussion exists already.
Over the years the likelihood of an unsuccessful search is reducing which adds to the reduced requirement to make a post. No stats for that. I certainly ask less questions than before – possibly because I’m getting just better / possibly because in addition to the fact that I am more familiar with the places I can get ready made answers.
Point taken and I do believe that is without a doubt one factor.
For some, yes, this is most certainly the case, but a great many still post the same questions, over, and over, …