Although retention is widely considered to be the most important metric to get right when building (or investing in) a business, it’s also one of the least understood. Why? Because unless you’re a growth expert or an experienced investor, you’re often relying on anecdotes, dated blog posts, and misguided benchmarks.
Hi Lenny, you say "But just know, startups rarely increase retention significantly." Curious to know why you say that? Does this mean a startup with a retention problem should pivot the idea?
Hey Lenny, if you have any resources on split of mobile vs main enterprise software retention rate i.e JIRA on desktop vs JIRA on mobile retention rates/ Outlook on desktop and Outlook on mobile retention rates. Every SaaS ends up having product on multiple platform, but standards of the retention rates on different platform is scarce. Please share /add on this article - will be super helpful.
Hi Lenny - thanks a lot for the very thoughtful research!
How do we tie in market penetration rates into dollar revenue retention? Businesses operating in low penetration categories can demonstrate higher dollar retention than more mature/penetrated categories.
How can we get a sense of net dollar/customer retention in ’steady state’?
Interesting insights, but, i think since each business within the same category is trying to solve very different need of the customer, not sure if such standardized benchmark will ever make sense.
I think the retention metrics will continue to make sense, given a long enough period. What will differ, IMHO, are engagement metrics, which are, as I see them, how often and for how long you use the product given that you are retained. A retained Airbnb user (after 1 year), for example, will probably use the service once a year, whereas a retained Instagram user (also after 1 year) will have used the service almost every week during the year. WDYT?
Only meaningful engagement leads to retention and retention then adds up to revenue. If defining meaningful engagement changes from company to company and category to category, retention changes too and so does revenue and hence standardization is a ??
Hi Lenny, thank you for sharing this article it's been on my reading list for a long time. Have you shared any materials that discuss how to capture these metrics using more basic analytics tools like Google Analytics or SQL (when the raw data exists)?
Hi Lenny, When looking at the good User Retention data for Consumer SaaS category, under the expert recommendations there are some numbers. Are those for monthly, yearly, or monthly/yearly mix plans?
Ps: You specified the Netfix/Hulu numbers are monthly/yearly mix but the export recommendations aren't specified.
Hi Lenny, for consumer transactional, what is the assumption on frequency of use within the 6-month horizon?
If a user only makes two transactions, in month 1 and month 6, I presume they're classified as retained under the numbers above. That said, this might be enough frequency for Airbnb but not for lower transaction marketplaces like Uber.
Across companies with different transaction sizes, do you have guidance on how frequent users need to be making transactions within the 6-month horizon to be classified as a "good retained user"?
For instance, for Airbnb transactions are in $100s so maybe once every 3 months across the 6-month horizon, whereas for Uber transactions are in the $10s so maybe once every two weeks across the 6-month horizon?
This is one of the best resources out there on retention. Thank you.
Clarification: Is the 6 month retention reference for consumer saas considered to be "N-Day" or "unbounded" retention over a 6 month period? It sounds like N-day given the comment that it is measuring users that came back in the 6th month.
Lenny wonderful wonderful research! Thanks for sharing.
In the consumer SaaS section, I noticed all the examples are entertainment related, did you find anything in the more 'aspirational/self help' side of consumer SaaS - examples being Audible, Blinkist, Calm or Headspace, Coursera/Udemy - essentially subscriptions that users have for products that require a bit more effort?
Lenny - Longtime reader and love your newsletter, but I think there's a problem here.
The user retention numbers for mid-market and enterprise SaaS here are misleading.
If you dig into the sources cited, none of them appears to be user retention, or even account user retention (ie, the % of accounts still using the product). Here's what's on offer.
- Atlassian: customer retention, or how many accounts are still subscribing (not USING) the product
- Salesforce: customer retention
- Slack: customer retention
- ADP: gross revenue retention (GRR)
- Workday: GRR
- Quickbook: Dead link
So were your expert recommendations using customer retention, GRR, or some mix? Was anyone describing user retention?
Frankly this is a disappointing because I had been using these "user retention" benchmarks with leadership at my co and will now have to explain why this isn't accurate .
I call out most of these points, don't I? I define mid-market and enterprise SaaS retention around paid customers (read the intro paragraph to that section), and I note for Atlassian, Slack, and QuickBooks the measure is customer retention. For ADP, Workday, and Salesforce I used the best number I was able to find on their customer retention.
If you're looking for which metric to focus on, I'd look at both customer retention and net revenue retention. I wouldn't focus as much on user retention in these cases, but obviously that's a leading indicator of the others.
Wondering if you know of any benchmarks on *user* retention for mid-market and/or enterprise SaaS companies? ie for Atlassian or Slack, even if the *customer* is retained, what % of *users* are still active after x months?
Thank you - this was a great resource. (I found the "user retention" section that included some "customer retention" data a bit confusing for a second as well, but I was able to clearly see which metric was used per company type.)
Really late here, but I think some of the consumer social numbers use different definitions. Snapchat and Twitter are using D90 numbers (number of users that came back on D90) whereas the other ones are using Month 6 numbers (number of users that came back between D181 and D210).
Hey Lenny, could you quickly tell me whether or not this is "unbounded" or "bounded" retention? Thanks so much!
Hi Lenny, you say "But just know, startups rarely increase retention significantly." Curious to know why you say that? Does this mean a startup with a retention problem should pivot the idea?
Check out this post: https://www.lennysnewsletter.com/p/how-to-increase-your-retention-issue
Hey Lenny, if you have any resources on split of mobile vs main enterprise software retention rate i.e JIRA on desktop vs JIRA on mobile retention rates/ Outlook on desktop and Outlook on mobile retention rates. Every SaaS ends up having product on multiple platform, but standards of the retention rates on different platform is scarce. Please share /add on this article - will be super helpful.
Unfortunately not :(
Exactly what I am looking for too! did you find it?
Hi Lenny - thanks a lot for the very thoughtful research!
How do we tie in market penetration rates into dollar revenue retention? Businesses operating in low penetration categories can demonstrate higher dollar retention than more mature/penetrated categories.
How can we get a sense of net dollar/customer retention in ’steady state’?
Great question, didn't look into that
Interesting insights, but, i think since each business within the same category is trying to solve very different need of the customer, not sure if such standardized benchmark will ever make sense.
I think the retention metrics will continue to make sense, given a long enough period. What will differ, IMHO, are engagement metrics, which are, as I see them, how often and for how long you use the product given that you are retained. A retained Airbnb user (after 1 year), for example, will probably use the service once a year, whereas a retained Instagram user (also after 1 year) will have used the service almost every week during the year. WDYT?
Only meaningful engagement leads to retention and retention then adds up to revenue. If defining meaningful engagement changes from company to company and category to category, retention changes too and so does revenue and hence standardization is a ??
Hey Lenny, could we do something similar for the health and wellness space? I assume retention will be way lower than any of the above categories
What companies would you love to see included in that benchmark?
Hi Lenny, thank you for sharing this article it's been on my reading list for a long time. Have you shared any materials that discuss how to capture these metrics using more basic analytics tools like Google Analytics or SQL (when the raw data exists)?
Thanks in advance!
Check this post out https://www.lennysnewsletter.com/p/measuring-cohort-retention
Hi Lenny, When looking at the good User Retention data for Consumer SaaS category, under the expert recommendations there are some numbers. Are those for monthly, yearly, or monthly/yearly mix plans?
Ps: You specified the Netfix/Hulu numbers are monthly/yearly mix but the export recommendations aren't specified.
Great question. Mixed.
Hi Lenny, for consumer transactional, what is the assumption on frequency of use within the 6-month horizon?
If a user only makes two transactions, in month 1 and month 6, I presume they're classified as retained under the numbers above. That said, this might be enough frequency for Airbnb but not for lower transaction marketplaces like Uber.
Across companies with different transaction sizes, do you have guidance on how frequent users need to be making transactions within the 6-month horizon to be classified as a "good retained user"?
For instance, for Airbnb transactions are in $100s so maybe once every 3 months across the 6-month horizon, whereas for Uber transactions are in the $10s so maybe once every two weeks across the 6-month horizon?
Hi Lenny, do you have any knowledge on how those retention numbers apply to the casual mobile gaming industry?
I don't :(
This is one of the best resources out there on retention. Thank you.
Clarification: Is the 6 month retention reference for consumer saas considered to be "N-Day" or "unbounded" retention over a 6 month period? It sounds like N-day given the comment that it is measuring users that came back in the 6th month.
Thank you Manish!
This is unbounded retention—what % of the cohort is still coming back at 6 months or later.
How does this apply to freemium models?
The numbers are primarily based on paid accounts
Hey Lenny, thanks for such a great article!
Apologize in advance if this is a dumb question.
When you refer to Atlassian, Slack & QuickBook's "Customer Retention" are you referring to users or accounts?
e.g. Google has 1,000 employees. Google is 1 account with 1,000 users.
Great question. Accounts.
Lenny wonderful wonderful research! Thanks for sharing.
In the consumer SaaS section, I noticed all the examples are entertainment related, did you find anything in the more 'aspirational/self help' side of consumer SaaS - examples being Audible, Blinkist, Calm or Headspace, Coursera/Udemy - essentially subscriptions that users have for products that require a bit more effort?
Thanks!
I unfortunately shared all the companies I have stats on. But if you ever find these, do share!
Lenny - Longtime reader and love your newsletter, but I think there's a problem here.
The user retention numbers for mid-market and enterprise SaaS here are misleading.
If you dig into the sources cited, none of them appears to be user retention, or even account user retention (ie, the % of accounts still using the product). Here's what's on offer.
- Atlassian: customer retention, or how many accounts are still subscribing (not USING) the product
- Salesforce: customer retention
- Slack: customer retention
- ADP: gross revenue retention (GRR)
- Workday: GRR
- Quickbook: Dead link
So were your expert recommendations using customer retention, GRR, or some mix? Was anyone describing user retention?
Frankly this is a disappointing because I had been using these "user retention" benchmarks with leadership at my co and will now have to explain why this isn't accurate .
I call out most of these points, don't I? I define mid-market and enterprise SaaS retention around paid customers (read the intro paragraph to that section), and I note for Atlassian, Slack, and QuickBooks the measure is customer retention. For ADP, Workday, and Salesforce I used the best number I was able to find on their customer retention.
If you're looking for which metric to focus on, I'd look at both customer retention and net revenue retention. I wouldn't focus as much on user retention in these cases, but obviously that's a leading indicator of the others.
Wondering if you know of any benchmarks on *user* retention for mid-market and/or enterprise SaaS companies? ie for Atlassian or Slack, even if the *customer* is retained, what % of *users* are still active after x months?
Thank you - this was a great resource. (I found the "user retention" section that included some "customer retention" data a bit confusing for a second as well, but I was able to clearly see which metric was used per company type.)
Really late here, but I think some of the consumer social numbers use different definitions. Snapchat and Twitter are using D90 numbers (number of users that came back on D90) whereas the other ones are using Month 6 numbers (number of users that came back between D181 and D210).
Yep! I call that out right where I share those numbers.