How we got our first 100 paying customers | Hacker News Reader

How we got our first 100 paying customers

stackfield.com

110 pointsmattiemass35 comments

Interesting PPC numbers. If they don't like the figures they were seeing

> Whether it is “project management” (€ 6.00 per click), “social collaboration” (€ 4.40 per click) or other similar keywords, the costs are in high ranges.

They would absolutely hate my industry. HR software, the number one keyword for people looking for our app http://www.staffsquared.com, is currently about £30 per click! It's symptomatic of this particular niche being occupied by incumbents who can afford not to optimise their Adwords campaign.

I should probably write a blog post at some point about how we got around using Adwords to grow.

I would be interested in that, I'm sure many others would too.

Interesting. Okay I'll give that some actual thought and see if I can get something down to submit to the HN crowd.

At least this post on HN drove your average CPC down ;)

Indeed :)

wow, £30 for a click?!

Yep, was about £10 three years ago before shit got crazy.

Some pharmaceutical keywords see bids in the hundreds.

I'm glad you found a strategy that got you your first 100 paying clients. But now you're unable to determine your CAC or CLV and have no idea how to scale acquisition. Or at the very least, your CAC is through the roof because your employees are the ones hustling (social, support, visibility, etc). That's difficult to scale and/or measure properly.

Have you completely abandoned the traditional methods?

I don't know why you've been downvoted, you're absolutely right: there's still a cost of acquisition here, and it's likely one that's pretty difficult to track down.

That is true - do you have a suggestion? Apart from Facebook and AdWords, what kind of marketing should OP try?

It must be a lot more satisfying to have new customers that were recommended by people that trust your product rather than some dumb/probabilistic marketing channel.

In a VC-backed world, that is hard to come by because everybody is so focused on hypergrowth, but if the company can afford to organically grow the business based on timeless good features (your product delivers value and you care about your users) than it should feel much better.

I was curious about their "encryption" since they seem to emphasize it a lot.

https://www.stackfield.com/security

This page indicates they're using RSA-2048 and AES-256. Wow, that's so vague. So I signed up.

It's using Javascript Cryptography, which is never a good sign: https://www.nccgroup.trust/us/about-us/newsroom-and-events/b...

Their RSA implementation is vulnerable to a padding oracle attack (Bleichenbacher's 1998 attack):

https://www.stackfield.com/Scripts/Plugins/rsa.2016012205415...

https://www.stackfield.com/Scripts/Plugins/rsa2.201601220541...

They're using AES-CTR, but they're not authenticating the ciphertext.

https://www.stackfield.com/Scripts/sf.wssecurity.20160122054...

(JSBeautifier comes to the rescue:)

    function DecryptOrganisationByMaster(e, a) {
        if (e === undefined || e.MyOrgRoleId === 3) {
            return false
        }
        var b = e.EncryptionCode,
            c = e.EncryptedCode;
        if (b === undefined || c === undefined || b === "" || c === "") {
            return false
        }
        if (OrganisationPasswords[e.OrgId] !== undefined) {
            return true
        }
        var d = Aes.Ctr.decrypt(c, a, 256);
        if (d === b) {
            OrganisationPasswords[e.OrgId] = a;
            return true
        }
        return false
    }

This really should have been reviewed by a cryptographer before being branded so heavily as an "encryption" solution.

EDIT: Also, their Aes.Ctr.encrypt() function doesn't accept a nonce:

https://www.stackfield.com/Scripts/Plugins/jquery.aes.201601...

See https://gist.github.com/paragonie-scott/53428f0947337d66a786

Not being snarky, but you really need a copywriter to review the wording on your site.

dayly (should be daily)

the data are encrypted

Over 10.000 companies joined Stackfield

Keep your information, that are not intended for the public,

> Not being snarky, but you really need a copywriter to review the wording on your site.

It looks like they're a German company, so it would definitely be valuable to get a native English speaker to review the site.

> Over 10.000 companies joined Stackfield

Yeah. Took me a while to see the error in that sentence, it's a good example. In german we write that with a . and not with a , which explain this one.

the data are encrypted is correct - "data" is plural.

Language Log has a post about this topic

http://languagelog.ldc.upenn.edu/nll/?p=4396

I quote:

"My own view is that there are contexts where it’s okay to treat data as a plural, but none in which you can’t treat it as a singular—and that contrary to what many “reasonable” usage writers counsel, this isn't simply a matter of “style and personal preference.” "

Other, shorter, humorous, posts about data:

http://languagelog.ldc.upenn.edu/nll/?p=4401

http://languagelog.ldc.upenn.edu/nll/?p=14970

http://languagelog.ldc.upenn.edu/nll/?p=20635

Language is a convention. It can change.

The convention seems to be the data is these days. Like agenda, which took the same road in the past.

FYI "dayly" should be "daily"

But very nice site

I think what's useful here are the numbers around PPC. I think many people have totally unrealistic expectations about CAC through these channels.

I'd be interested how they would now perform if they returned to PPC...

I operate a SaaS that is based around Google Analytics Dashboarding (http://www.embeddedanalytics.com). I have never had success with Google Adwords, very low conversion rate to signups (<2%). Traffic coming from referral sites (mostly stackoverflow.com) is much better.

There's a typo on your home page:

"Stackfield offers you a selection of tools that every team needs to get the dayly work done."

Dayly should be daily.

Also cool post!

They mention going through an accounting department in the case of big companies, I wonder if anybody had more luck with this?

"More luck" in what sense? All enterprise companies will have plenty of obstacles thrown in the way of new suppliers. The most classic: if you are not on the suppliers' list, i cannot buy from you; and you cannot get on the suppliers' list unless I buy from you. Catch-22.

Joel Spolsky recommends to figure out the average limit for single-item purchases on corporate credit-cards, and stay under that. This may or may not work in your particular segment.

Summary: Company did not find a marketing channel. Company relies on word of mouth.

    Orig wordcount: 2152
    TlDr wordcount: 13
    Saved: 99.40%

You forgot some useful data:

"While our first customers came from the above mentioned sources, including press articles, presentations, etc., almost 80% of our first 100 customers came through referrals. Another interesting point at this stage: hardly any user or customer recommended us due to cash benefits or discounts. We offer a 10% discount to our users when they invite new customers. However, when we looked at the number of actual referrals, we realized this program had little response."

How about: company discovered that having a good product your customers really use and recommend helps to sell the product.

I enjoyed the post. Nothing ground breaking, but a nice reminder that no amount of marketing will help you if your customers are not satisfied with what they get. Good luck to OP!

If this really is a bot, I'm just astonished. Well done to the crew behind this.

I'm skeptical. The phrase "marketing channel" does not appear verbatim on the article. I'd say it either has really good NLP or there's a human somewhere.

There's probably some 'selection bias'. I think they might run the bot on every article, but post the summary only if it matches well.

If this is a bot, then why isn't it running on at least all of the front-page HN stories?

But anyway, please keep posting :)

I expect they manually write the summary, then let the "bot" calculate wordcount and post, i.e. like a command: "tldr Company did not find a marketing channel. Company relies on word of mouth."

This post was way more interesting than the blog article.

What was the company? I didn't catch this in the article.

Reply on news.ycombinator.com