Taking a look at iNaturalist's latest form 990 (2024)

General
1

One way to better understand a US-based nonprofit organization is to look at its form 990 filing to the IRS. iNaturalist posts theirs on their Financials page. As of early 2026, the latest filing is for calendar year 2024. So I thought I’d take a look, and share my thoughts of a few of the things I see in there.

First, I think it’s useful to understand how much it costs for the organization to fulfill its mission (which it does primarily by developing and maintaining the platform, along with a bit of outreach). It looks like they spent $3.1MM in 2024 to run the organization but also used $2.2MM worth of additional services received in-kind (probably mostly mapping services, cloud storage services, and legal services):

item $ % source
Expenses from Financials 5,289,769 100.0 Sched D, Part XII, item 1
Donated services / facilities 2,238,192 42.3 Sched D, Part XII, item 2a
“Functional” Expenses 3,051,577 57.7 Part IX item 25

Of the functional expenses, most of this – $2.6MM (84%) – goes to paying employees. It looks like they record this as standard payroll for US-based employees and as contract services for employees outside the US:

Total Compensation $ count source
US staff 2,129,224 15 Part I, section B, item 15 / Part I, item 5
Non-US staff 428,072 3 Part IX, item 11g (more detail in Schedule O) / Part VII, section B, item 1
All staff 2,557,296 18 (sum)

(Note that they were still growing staff in 2024, and I think there may have been some turnover, too. So the $ amounts may reflect only part of a year’s work for some employees, and the employee counts reflect any employees paid during any part of the year.)

Here’s the breakdown for US employee compensation:

item $ % source
Total Compensation 2,129,224 100.0 Part 1, section B, item 15 / Part 1, item 5
Base Pay for officers & key employees 505,424 23.7 Part IX, item 5
Base Pay for other employees 1,350,755 63.4 Part IX, item 7
Total Base Pay 1,856,179 87.2 (sum)
Pension 57,723 2.7 Part IX, item 8
Other Benefits 86,035 4.0 Part IX, item 9
Payroll Taxes 129,287 6.1 Part IX, item 10

The next largest chunk of the functional expenses is classified as Information Technology – $295K (10%) – which I assume is the cost of the infrastructure and other hardware and tech services needed to run the platform that was not donated. The remaining 6% is just admin and other miscellaneous stuff.

…

Ok, now let’s look at iNaturalist’s revenue. It looks like they got $4.7MM in revenues, plus an additional $2.2MM worth of in-kind donated services.

item $ % source
Revenues from Financials 6,949,552 100.0 Sched D, Part XII, item 1
Donated services / facilities 2,238,192 32.2 Sched D, Part XI, item 2b
Revenues 4,711,360 67.8 Part I, item 12

Most of the $4.7MM comes from gifts to the organization of $4.4MM (94%), based on Part I, item 8 (current year). Of that, more than half comes from a single contributor (which I assume is the Moore Foundation). The top 5 contributors accounted for almost three quarters (73%) of the contributions, and the top 20 accounted for almost four fifths (78.5%). I was a little surprised that there were only 20 contributors who gave $5000 or more in 2024. The remaining contributors’ contributions accounted for a little more than one fifth (21.5%) of that total.

Here’s a summary of contributions (from Schedule B, Part I):

Contributor(s) $ % cum %
#1 2,501,038 56.4 56.4
#2 250,000 5.6 62.0
#3 165,708 3.7 65.7
#4 161,500 3.6 69.4
#5 150,000 3.4 72.8
#6 through #20 256,138 5.8 78.5
contributors <$5000 951,992 21.5 100.0

If you think about these contributions in comparison with expenses, the Top contributor basically covered staff expenses in 2024. The remaining contributions are less than the value of the in-kind donations though. So if iNaturalist ever lost those in-kind donations, they might have to pay for those services from reserves, if they couldn’t raise more money than they did in 2024. That said, in 2024, revenue less expenses was $1,659,783. So they are doing fine for now.

…

iNaturalist had very minimal liabilities – just $105,184 of accounts payable and accrued expenses (from Part X, item 17) at the end of 2024 – and it had significant assets totalling $5.9MM, including $317K in cash and $5.4MM in savings and cash investments (Part X, items 16, 1, and 2). Most of the assets were unrestricted (Part X, items 27 and 28).

This means that they likely would be able to run things exactly as they did for another year, even if they didn’t get any donations of cash or in-kind services.

…

All in all, it seems like iNaturalist’s operations are relatively straightforward, and they were in a strong financial position at the end of 2024. We outsiders won’t have insight into what happened in 2025 until they release the 2025 990 in April or so.

I’m expecting contributions to be up due to the Google.org grant(?), but staff costs might also be up in 2025 for various reasons. Still, I expect them to have a large surplus again for 2025.

In my mind, they seem to be running things relatively conservatively. It seems like they have room to spend more to develop the system faster or work on more things, but maybe that’s constrained more by organizational concerns than by finances for now? Alternatively or in addition, it seems like they could probably spend more on outreach or (fundraising) development.

I guess one thing that’s probably not captured in the form 990 is the value of the work of volunteer site curators, translators, and code contributors. These all directly affect the platform, and It might be interesting for the organization to try to quantify that, if they haven’t already.

26 Likes
What is the monetary value of your volunteer work for iNat?
3

“Total number of volunteers (estimate if necessary) 1,300,000”

Very interesting. Is everyone that uses the site considered a user or a volunteer?

2 Likes
4

It would be interesting to put a nominal value on An ID. Altho all IDs are ‘equal’ whether it is hopeful Kingdom or the taxon specialist IDing the first obs of a new sp - but that could also be assessed by taxon level ?

More difficult to put a hard value on curator’s work - but perhaps iNat could count online at iNat time (which would not cover research away from iNat) ? Taxonomy and moderating social media need to be assessed separately.

4 Likes
5

to me, difficult or not, if you’re going to quantify the value of volunteer work, this is where you start. i think a lot of curator work, especially community moderation, is stuff that needs to happen and otherwise would fall upon staff. or if you think about it in terms of operations, if all the curators went away, and you couldn’t find any new volunteer curators, how much would you have to pay someone to take care of that work?

8 Likes
6

Can someone with a better mind for numbers please let me know if I am correct, or how far off I am? I am better in the field than in accountancy.

If math and understanding are correct, iNaturalist spent $5,289,769 on total expenses.

All observations for the year are 56,525,801 (including casual).

That equates to $0.0936 per observation uploaded.

Casual observations (some 2,471,695) cost $231,565. Which is 4.38%.

While some casuals may play an important part in understanding biodiversity (a bug on your house plant, where you need the one to identify the other), what is the cost of “rubbish” observations? Pictures of friends, Pictures to out of focus to identify anything, Duplications and others? In other words, observations that hold no value to biodiversity monitoring.

These sums exclude curators and identifiers who provide an extremely valuable free service.

PS: It is amazing how fast the numbers change. In writing this, an additional 35 observations were uploaded for 2024.

It will be interesting to see these sums again in 2025, and already an increase of total observations exceeding 11%.

2 Likes
7

When people say they stay on iNat because, altho social media, it is not toxic. They forget the quiet and hidden work which is WHY it is not toxic. There is, junk and nastiness - altho we mostly don’t see it before it is whacked away. Sometimes a complaint on the forum, before people learn to flag it away to moderators.

8 Likes
8

Many of the costs will be fixed so it doesn’t literally cost 9 cents every time someone uploads an observation. If there were fewer observations uploaded, the total cost wouldn’t go down accordingly. The unit cost of each observation would rise.

13 Likes
9

I do agree.
More looking at the cost of observations that are not carrying value, and what they cost.
Got to start somewhere.

11

Thanks to @pisum for delving into this topic. I’m a nonprofit professional who is used to reading 990s. This is an excellent analysis and I agree that these two years look pretty solid. No red flags from my point of view. But really for me to assess a nonprofit I like to see at least three years of financials, but we will have to wait as they have not been their own nonprofit that long.

As to those who have mentioned the value of in-kind/volunteer services, only in-kind services that are documented with an invoice showing the value of the services and that $0 was paid by the org are allowed on the 990.

But, if I was writing grants for them I would absolutely use some creative math to assign a value to the volunteer efforts of identifiers. To estimate the value of volunteer hours, nonprofits generally use the website indepdendentsector.org (https://independentsector.org/research/value-of-volunteer-time/) which currently shows for 2025 the value of volunteer time as $34.79. So technically, each hour you spend doing IDs on iNat is worth that much in in-kind volunteer services. Many grants allow you to use volunteer hours as a match. I am not saying they are doing this, just that I would if I was fundraising for them.

16 Likes
12

Wow. Surely at some orgs that’s more than many actual staff get paid!

3 Likes
13

Yes, this is true. I once wrote a federal grant that included my own hourly rate to manage the grant which was sadly around $25/hour, and the volunteer hours were valued at more than that. It’s the unfortunate nature of nonprofit. I have tried to advocate throughout my career to bring nonprofit worker salaries closer to private sector and dispel the myth that nonprofits spend too much on “overhead.” But that’s another topic for another time.

10 Likes
14

I did see while I was perusing the 990s that iNat had someone benchmark the salaries, so hopefully with iNat staff at least there isn’t that problem….

1 Like
15

This made me think (warning might be off topic) and I had some fun with calculations:

As of 1/11/26 I have 61,983 ids and have reviewed 68,767 observations.

Side note

I suspect this number “observations reviewed” to be higher, this is from selecting all “qaulity grades” in identify mode and clicking the “double left pointing horizontal chevron” on the the right side to get a percentage reviewed of total.
If I include every observation I have perused in the observation tab and observations where I deleted an id/comment and it was marked as unreviewed(known bug)
I get a value closer to 78 thousand (78,459 based on estimates of observations percent id collected during timing of 30 observation reviewing)

I did some sample identifying of some common types/filters I would do to get some estimates for time to review 30 observations(the default layout).
My very conservative average was 137 seconds to review 30 observations (of which I left an id on 79%).

Data
Category Time(in sec) Identified
Broad id of identified=false 238 25/30
Confirming Vitis rotundifolia in NA 80 30/30
Correcting Phaeolus schweinitzii id’s 99 30/30
Sight id of Florida plants 140 10/30

With certain identification types my limitation is internet speed for loading pictures, these values are for my quickest work, I also sometimes individually key out species and take minutes per observation, but I am trying to make the most conservative estimate of my time

So if I have reviewed a minimum of 68,767 observations taking a minimum of 137 seconds to review 30 observations that is 314,035.96 seconds spent reviewing observations or 87 hr 13min and 56 seconds.
I value my time at 25/hr currently, that would mean my id’s are worth $2,180.75.

Upper end estimate

My upper end average estimate for my observations reviewed (~78k) and time per observation (~5min per 30 obs) puts this value to ~$5,448

We can also take into account Flags I have created (79…maybe I should be a curator…if only there weren’t a bunch of “curators” who have never used their powers) and estimating time spent per each flag is worth $101.87.

Breakdown of flagging activity
Flag type quantity Assumed time in minutes per flag
Taxon flags 38 5
Photo flags 15 3
Identification Flags 10 .5
Observation Flags 9 .5

From this assumed time the total is 4.075 hours

One last metric I can track is bug reports (14) and implemented feature requests (1) assuming a minimum of 3 minutes per big report and 10 minutes of work for the feature request, that would value at $25.75.

This brings the total to $2,308.37
Final value does not account for taxa I have added display pictures to or modified their display picture as I cannot figure out a way to obtain this metric but I suspect it is in the hundreds.

I will state that I do not except any compensation for my volunteered time and this is just for gits and shiggles (other than maybe having my option and feedback listened to).

I would be interested to see other peoples evaluations of their work (maybe in a separate thread as to not be off topic).

2 Likes
16

i think it’s odd that iNaturalist considers every observer a volunteer. to me, observing and even identifying are more just using the system. to the extent that someone wants to claim that effort as volunteer work, i would think it’s more appropriate to count that as volunteer work in the context of observing or identifying for a particular project or place. for example, if there’s a group that is monitoring pollinators in a particular park, then it seems to me like volunteering to observe and identify pollinators in that park would be more correctly counted as volunteer hours for that monitoring group, not for iNaturalist.

maybe an exception would be something like when folks identify for one of iNaturalist’s tests, such as computer vision / accuracy verification, since that goes directly towards a project led by iNaturalist.

to me, site curation, translation, system code development, and ambassador stuff are things that are clearly closer to volunteer work for iNaturalist, since these are things that would otherwise fall on staff to do.

that said, it’s still interesting to try to quantify the amount and value of identification, observing and other work that people undertake for the benefit of everyone (not just iNaturalist) which otherwise would not occur without the system. if someone had to pay folks to create this sort of dataset, i’m sure it would have cost in the many millions – if not billions – of dollars to do that as a commercial effort.

6 Likes
17

@anneclewis I am about to start a separate thread, so you might want to hold your comment for a minute

1 Like
18

Agree with @pisum’s thoughts.

In September of 2023, shortly after iNat became an independent 501(c)3 nonprofit, they received $10,000,000 from the Gordon and Betty Moore foundation. This donation was intended to launch iNat as a stand alone organization. Judging by the 2023 and 2024 990’s, they are getting this funding piecemeal, probably 20-25% of the donation each year over 4-5 years.

My guess is that they are banking the money until they get a better idea of what their actual expenses are (per @lily-of-the-valley 3 year trends) and see what their fundraising can bring in. It would be very hard to project sustainable expenses and revenues without this information. ETA: I noticed they hired a fundraiser this year. Honestly, I think they need at least two: major gifts and recurring/annual gifts.

Per my other post on a different thread, when you look at the global reach and impact they have, they are doing a lot with relatively little. I certainly wouldn’t put them on the same level, financial wise, as the “Big Green” organizations like the Nature Conservancy, Audubon or the Sierra Club.

ETA Part 2: You can find the Form 990 for any US nonprofit at candid.org.

8 Likes
19

May I offer a different perspective from someone who is not engaged in the administration of any non-profits?

TMI.

I won’t be following this thread any further but I’m glad there are some iNatters who have the time/energy/interest in doing so.

Please alert me when/if iNat is broke or broken. Until then, …

5 Likes
22

(this is a reply to another user’s post that is now deleted, stating something to the effect of: “volunteer hours must be defined somewhere”.)

the form instructions state:

Line 6.

Enter the number of volunteers, full-time and part-time, including volunteer members of the organization’s governing body, who provided volunteer services to the organization during the reporting year. Organizations that don’t keep track of this information in their books and records or report this information elsewhere (such as in annual reports or grant proposals) can provide a reasonable estimate, and can use any reasonable basis for determining this estimate. Organizations can, but aren’t required to, provide an explanation on Schedule O (Form 990) of how this number was determined, the number of hours those volunteers served during the tax year, and the types of services or benefits provided by the organization’s volunteers.

iNaturalist does not provide any details in Schedule O.

4 Likes
23

That’s totally fine - all of us are different. I assume you’ve been participating or reviewing this thread also: https://forum.inaturalist.org/t/is-the-inat-directorship-taking-inat-in-the-right-direction-and-paying-enough-attention-to-the-user-base/74434

I only respond bc by the time something is broken in a nonprofit, it can be really hard to course-correct. I say this as someone who has served on a board and has witnessed orgs literally fold overnight due to internal problems (usually financial errors that were not caught). By then, there is both an internal rebuilding that has to occur, AND an external trust rebuilding that can take a long time.

2 Likes