Post-mortem of a Dead-on-Arrival SaaS Product

https://www.petekeen.net/marginalia-postmortem

by zrail • 13 years ago

141 84 13 years ago

bradhe• 13 years ago

You put it on HN and when it didn't move you gave up? What else did you try?

levosmetalo2• 13 years ago

Writing a Post Mortem to try to resurrect it from the dead. And submitting it again to HN, hoping that second life would be better.

Romoku• 13 years ago

Well he put the source on Github and it's very interesting to see how some projects come together.

zrail6• 13 years ago

Ha! No, no interest in resurrecting it at all, I just wanted to share what I learned.

bluedevil2k5• 13 years ago

He says later in the article he tried paid advertising, but it never led to sale.

zrail6• 13 years ago

I tried content marketing microsites, google ads, reddit ads, talking about it a lot, etc. Nothing really stuck, and then life got in the way and I lost interest.

tixocloud2• 13 years ago

I think you did a really great job in completing the project and launching. Disclaimer: I'm an MBA student so take my thoughts with a grain of salt :) I thought that your marketing was spot on in terms of trying to generate something but I'm just wondering if you tried targeting other types of customers or even zoning in on very specific types of programmers ( beginner level, expert level, etc)?

bluedevil2k5• 13 years ago

This was the biggest take-away from this article, one I've learned after wasting much time building my own SaaS "products". Launching without an audience means nobody shows up. It's a very true statement. One of my favorite sayings now, from Jason Cohen, is to get 30 people to fully commit to pay for your product before you even start coding it.

waterside81• 13 years ago

Cohen's advice is good for two reasons:

1) Obviously you see if there's any demand for your product but perhaps more importantly ...

2) It forces developers to put on their salesperson hat. Lots of us developers just like building products and just willfully ignore the other part - selling the darn things!

ronilan• 13 years ago

get 30 people to fully commit to pay for your product before you even start coding it.

I can't think of any valuable product that was created this way.

runako• 13 years ago

That's quite the sweeping dismissal of Kickstarter.

lifebeyondfife2• 13 years ago

Obviously it depends on your exit strategy. Do you want to (a) get a loss making but huge user base or (b) bootstrap a business that has cashflows from day 1?

Sounds like that advice applies more to the latter route than the former.

michaelbuckbee• 13 years ago

Well Jason Cohen wasn't talking in the abstract, he's talked about how he got checks up front before starting WP-Engine.

chrismorgan• 13 years ago

Note the careful wording: "before you even start coding it". Coding is a time-consuming thing, very fragile (if you get it wrong, you may almost need to start over again) and brittle (making changes is, connectedly, hard). Very significant validation can be done without ever writing code. (That's a lesson that I've been learning this year, as one who delights greatly in coding.)

lifebeyondfife2• 13 years ago

I likewise performed a detailed dissection of my previous failed website on my blog - very therapeutic. It still has one fairly regular user (me!) so I keep it going but I launched in much the same way: built it silently then managed to get about 10k+ visitors through HN. Only a few hundred gave it a go initially and within a fortnight or so, no tangible traffic at all.

Great learning experience though, and this post is a reminder to me to make sure I don't repeat the same mistake.

awwstn• 13 years ago

get 30 people to fully commit to pay for your product before you even start coding it.

This seems like sound advice, but I'd like to hear more on the execution of it. Do you actually get these people to sign contracts, or is it more of getting people to say "yeah, I'll pay you for that"? Do you let them know you haven't written any code, or just say it's "in development"? Etc., etc.

kareemm• 13 years ago

I wrote a post about how we got 8 commitments to pay before writing my current SaaS app. It has made well over 6 figures and counting:

http://blog.reemer.com/how-to-get-paying-customers-before-yo...

wikwocket2• 13 years ago

Brennan Dunn and Nathan Barry have both appeared on HN a number of times, and their approaches to selling eBooks and SAASes have included the option to pre-order the book/prep-pay for the SAAS subscription at a discount.

Something like this works great when you have an audience and/or have some decent previews to show. (btw, my takeaway here is not 'this only works for internet-famous people,' but rather, 'it's a great idea to build an audience for your idea and show them previews to keep the leads hot')

jenius4• 13 years ago

Just this weekend, I started coding up a tiny "saas product" that I thought about last week. Nobody knows about it, and that certainly has no launch page or audience. I got most of it done over the weekend, and really enjoyed putting it together in my favorite stack (sinatra, roots, marionette) and doing some design work, which I don't often get a chance to do.

I'll probably launch it late this week or next week. I have no audience, and don't expect massive profits either. I made it mostly because I want it. Since I plan on using it extensively, I honestly don't care if I have huge profits or a gigantic audience. I spent $5 on the domain name, am hosting it along with 10 other sites on a $5/month digital ocean box. I put in some great work this weekend, and am now going to end up with something that will fit nicely into my workflow.

It will ship with a decent free plan, and a $5/month pro plan with some expanded capabilities. If one person signs up for the pro plan, that covers the cost of hosting for this site and 9 other sites (bonus!) If 2 people sign up for the pro plan, it's profitable. If zero people sign up, I paid a very small cost for something that's going to be really useful to me.

Sometimes I wonder why so many developers gravitate towards huge ideas that they think other people might want rather than starting small, with things that you want. With this little thing I'm making, there's no risk. I've already won.

sejje• 13 years ago

Because so many developers want to build a business. You've aimed your sights at making $5/month.

bluedevil2k5• 13 years ago

What you've made is just a hobby then, not a business. (Not that there's anything wrong with programming being a hobby - it's more about managing expectations)

AznHisoka3• 13 years ago

Because we don't want to be stuck in a cubicle until we're 70 years old. That's why.

mcherm• 13 years ago

So..... what did you make?

(The fact that you didn't say shows that you REALLY aren't concerned about marketing! But I'm still curious.)

nashequilibrium• 13 years ago

One of the reasons i taught myself to program years ago, was because i wanted the power to make my life easy for myself. I am currently building a semantic supply chain for my ideas, i have being using a few notetaking apps but they do not satisfy my needs because they lack in their ability for me to query, visualize and interlink my ideas.

So i was reading a paper on supply chain management and thought maybe i can implement this process for taking my ideas from inception to prototype. This is not a startup and i will use the cheapest hosting i can find, but i am thinking of allowing people to use it in the web and app form. If it gets traction, i will adjust app accordingy, if not i have a cool process to get my ideas out and visualize their process.

The bottom line is: "Don't build technology unless you feel the pain of not having it." http://nathanmarz.com/blog/suffering-oriented-programming.ht...

chris-allnutt• 13 years ago

I disagree with this statement on the basis that this assumes all products are made with monetary profit in mind. I'm working on a product right now that is for myself and my spouse to solve a money management issue we have, it is something I will use even if nobody else does.

I will be adding the ability to purchase it as a subscription so that if others want to use the product they can too, but ultimately I'm building something I want to use.

Jason Cohen's statement is based on the idea that all software is written with profit in mind, and there are a good number of years in Open Source that prove that to not be true.

bluedevil2k5• 13 years ago

I think all SaaS products (which this article is about) have monetary profit in mind.

pjmlp• 13 years ago

Jason Cohen's statement is based on the idea that all software is written with profit in mind, and there are a good number of years in Open Source that prove that to not be true.

Usually in most cases the code is not the direct source of income for the developers.

mipapage• 13 years ago

get 30 people to fully commit to pay for your product before you even start coding it.

While I don't know about 30, we use some apps in house that clients began to use for their business. We're now looking at building those apps for ourselves and our clients are on board. We'll be saving the expenses, and have some folks on board before day one. Cool.

Along the road we have had the clients ask us if we can extend the app for certain use cases they have. Since we built their web-presence and integrated those with other tools.. Lets just say things are developing.

So maybe look to your existing clients (you know, the work you do during the tim you wish you were building your app). They are a great source for us, so far...

kpapke• 13 years ago

I have a question about the lesson here "don't write code without having committed customers first." Does this apply to a startup that is not a SaaS product? I am working on an app that has the chicken/egg problem where it can't make money straight away, it needs a strong user base to be effective.

I have been working on this app for the past year it seems like, with development really amping up in the last 6 months. We just brought on another developer to help us finish up and get it out the door. I submitted my market information and business proposal to a university marketing research class, and they gathered data from a 50 person survey for us, which validated our ideas & assumptions. When we do launch it will be a quiet soft launch with friends & family trying it out first. Then we will tweak it with their feedback before we really try to make it public and do serious marketing efforts.

All of that said, our monetization strategy won't be implemented until post-launch once we have this feedback from actual users and a good amount of them sticking around. The whole thing is bootstrapped and we haven't spent anything, aside from the weekend development hours. Does this seem like the right approach for launching our product?

wikwocket2• 13 years ago

It sounds like you have some validation, which is a huge step in the right direction, but more is always better. One reason is that, if you ask someone if they would like to use/have product XYZ, you will get very different answers and feedback than if you ask, "Would you like to pay $NNN for Product XYZ right now?"

Depending on your product, I might recommend starting to build an audience of people interested in your product. Try to interact with them as soon as possible. Get them interested. Then when you launch, you have a stable of warm leads, champing at the bit to go.

zachsnow• 13 years ago

I would rephrase as N customers, where N is enough to earn some significant percentage of your costs. For many kinds of businesses N might be only a handful, it completely depends on your market. For others 30 is nowhere near enough I'd imagine.

Of course, determining significance is left as an exercise for the reader.

jh32• 13 years ago

Reasons why I would not pay for this (I'm a developer):

- I already keep notes in markdown using nValt. It's fast. And it's free.

- I sync these notes, which are just .markdown files in a directory on my machine, with Dropbox. Now I can edit these notes on my iDevices.

- Your service costs me money to do what I do for free.

- If I need more bells and whistles, I use Evernote. Evernote is also free for me.

- Emailing myself notes with tags in the subject in is also free.

Why should I spent $5/mo to use a digital journaling service, and then more time to make this service work with the rest of my workflow, when I already have things in place that take care of my note taking problem?

I would need a great incentive to switch how I take notes. I'm thinking a bunch of other developers thought the same thing.

Finally, at least you shipped something. Nice retrospective. Keep at it.

Edit: It's interesting seeing this post and "Ask HN: How do you store and organize your startup ideas?"[1] on the front page at the same time :)

[1] https://news.ycombinator.com/item?id=6585375

drewlarsen• 13 years ago

I would need a great incentive to switch how I take notes. I'm thinking a bunch of other developers thought the same thing.

Just like the OP, you've generalized your personal experience to other developers. It is an easy trap to fall into. I agree with the takeaway called out in another comment: validate the idea as objectively as possible and as early as possible.

jh32• 13 years ago

Yes you're right. It's the only data point I have, so it's what I went with :)

What you're echoing is also correct.

tixocloud2• 13 years ago

It's funny but your list of reasons are kinda stuff learned in business school. Switching costs, substitute products, etc are common to all businesses in my opinion and a little bit of analysis might have helped the OP to identify his niche. However, that said, the experiences learned from failing will have a huge impact in the OP's success in the future.

TeMPOraL• 13 years ago

One more reason for me not to switch:

- why in hell would I want to use a browser-based solution for something a free, native, off-line program is much better?

yohann3052• 13 years ago

Thanks for posting the "behind the scenes" story.

You say you built a landing page with mailchimp signups for your new product. Did you get many signups? And I suppose these were just signups, no money involved, right? Did a good chunck turn into paying customers?

I would love to know more.

Thank You!

zrail6• 13 years ago

I had a list of about 300 people when I launched preorders and about 30% of them converted to paying customers. I was sending out weekly updates with progress toward preorders, and then after preorders started I sent out weekly updates with download links for people who paid and "buy now for X% off, won't last long" messages for people who hadn't.

The whole story is here: https://www.petekeen.net/adventures-in-self-publishing

yohann3052• 13 years ago

Wow! Nice conversion rate! Good job on not making the same mistake again!! And Good luck!

I'm reading the "Adventures" post as we speak.

AznHisoka3• 13 years ago

What was the initial reactions of people when they tried your product? Did they say they loved it and would want to use it everyday?

andyidsinga2• 13 years ago

its sort of silly, but one thing in the post that stood out for me was the $29/month for SSL and database.

It was a significant amount of money for this dev - but it could have been done even cheaper with the aws free tier and getting an ssl cert from somewhere like positivessl / commodo.

egomaksab• 13 years ago

It's actually $9/month for Heroku Basic DB plan and $20/month for SSL endpoint not cert (you can use your own domain and SSL cert). You can always piggyback on yourapp.herokuapp.com for free SSL cert.

falcolas2• 13 years ago

This is probably my background in sysops talking, but $29 a month for an nginx frontend, rails daemon and PostgreSQL instance sounds like highway robbery...

thebiglebrewski• 13 years ago

Great post Pete, I'm really enjoying your Stripe + Rails book. Keep it up!

zrail6• 13 years ago

Thanks!

heliodor• 13 years ago

I think the right pricing model for this one is freemium. People start using it, get addicted, then you charge them for extra storage once they go over a limit, like Dropbox and Workflowy.

marincounty2• 13 years ago

Agree!

alkagupta0309• 13 years ago

The Idea is nice but I wouldn't pay for it mainly because of the reasons stated by Pete Keen and another reason: When I have free tools to be doing the exact same things, why would I want to pay? If I had to launch a SaaS product, I would definitely create a buzz about it amongst my peers and give free subscription for a limited period enough to get the user used to the new idea.

nonchalance• 13 years ago

I apologize in advance if I sound snarky, but I (and I suspect many others here) hadn't heard of Marginalia until today. How much did you spend on ads and where did you advertise?

ma2rten• 13 years ago

Don't be silly. When he said he spend a lot on advertising he didn't mean that put up a superbowl ad. He probably spend it on Google Adwords for people who where searching for related terms.

VLM2• 13 years ago

"people who where searching for related terms"

I think that would be a strange thing to search for, and I'm not sure it would be a profitable demographic.

Sometimes the problem is not just "do people search for something" but who are those people doing those searches?

marincounty2• 13 years ago

Being on this page was his best advertising--ironic?

WhitneyLand• 13 years ago

Great post for you because the honest introspection will lead you to more success.

Great post for us because even though these lessons have been written about before, it seems to take lots of repetition for them to sink in all the way.

VLM2• 13 years ago

I looked at the original announcement and looked specifically at the technical details list. If I didn't have the stuff on the tech details list available, and had to accomplish roughly the same non-technical task, there are a lot of ways to take notes more or less for free. I can't operate without gmail and I guess google drive so I need not worry about an app built on them.

Dropbox is not a negation of my argument, in that they handle the annoyance of keeping a storage service up.

In my cartoon mind if gmail is down, then storing notes in gmail doesn't work, but the bigger headache is gmail is down not that I can't access some minor notes. So just use gmail for free. Or github. Or a browser extension to "save email to dropbox or google drive"

The other problem is you have a very narrow window of opportunity for a programmers note taker because of habits. The concept of writing README and TODO files and inserting source comments or placeholders, perhaps with tags to search for and ASCII art for markup, has been a part of programming education and tradition for a long time. Could you sell a note taker app to noobs who don't already have a github / text editor habit? Well, maybe, but it'll be harder.

So in my cartoon mind I have an idea it gets entered into TODO or perhaps as a TODO flagged comment in the appropriate spot and its done. You need a wedge to fit your process into it. Like if PUSHes to github regularly failed or github filtered out source code comments to "save disk space" or something. But if it works great, something new is going to be a hard sell.

Now if you did something unusual once you got the data that might sell. Here's a nice web front end that uses GIT and/or github and/or dropbox and/or google drive and/or gmail as a backend.

JabavuAdams• 13 years ago

Interesting. I've been throwing around the idea of roughly the same thing for myself. I was keeping a TiddlyWiki, but that's stopped working. Developer solution ... write my own!

In general, developers are a terrible market because they think they can do whatever you did better / cheaper / faster, and they won't shut up about it.

zrail6• 13 years ago

I would say that developer tools are a terrible market for that reason. Information products (screencasts, books, etc) can be extremely successful in the same market. Examples abound, including patio11's marketing training[1], egghead.io[2] for AngularJS training, Peepcode[3], Railscasts[4], and various other ebooks [5][6][7].

Most developers that I know have two common traits that make info products an easier sell. First, they're innately curious about their field. Second, they have access to a company credit card that will happily pay for training materials.

[1] https://training.kalzumeus.com

[2] http://egghead.io

[3] https://peepcode.com

[4] http://railscasts.com/pro

[5] http://www.nathanbarry.com/authority

[6] https://www.petekeen.net/mastering-modern-payments

[7] http://www.jstorimer.com/products/working-with-ruby-threads

rajbala• 13 years ago

"The next big thing will start out looking like a toy" -- Chris Dixon

http://cdixon.org/2010/01/03/the-next-big-thing-will-start-o...

cookiecaper• 13 years ago

Any service that is dependent on network effects needs a plan to cultivate activity before it launches. It's crucial to offer engagement to the Real People that are early adopters on your product, and the reality is that if you go hands-off and attempt to wait for the audience to self-materialize, you'll get a lot of people who drop by, make one post, and check for interaction a couple of times, and then never come back. It's a bad deal for everyone.

Get together a group of people -- pay them if you have to -- who are committed to using your product before you do a real public launch, and then new users will immediately be engaged and it will grow from there. Launching without this spells disaster.

If you are launching SaaS that doesn't depend on other parties' involvement, you should still gauge interest, but it's OK to start small.

ggjain• 13 years ago

Rightly said, SaaS easy to build hard to sell. Products hard to build are easy to sell, as it has value and willing high value customers. But in all, a product or business without a clear customers acquisition strategy is a miracle.

johnrob• 13 years ago

Are you still going to use your product after the SaaS version is shut down?

zrail6• 13 years ago

No, I switched to greener pastures[1] a long time ago.

[1] https://www.petekeen.net/git-backed-personal-markdown-wiki

njr123• 13 years ago

Just as an aside, if you have a project that you're not sure is going to take off, there is no reason to shell out $29/month for it. Shared hosting will do the job for $5 - $10 a month. No its not super reliable, but if you have no users, who cares? And if you do get popular, my experience has been that they can handle bursts of traffic surprisingly well, until you can move to proper hosting.

Post-mortem of a Dead-on-Arrival SaaS Product

Related Stories