Ask HN: Why surveys are so expensive?

3 pointsmichaeloblak2y ago5 comments

All SaaS survey products are insanely expensive. It’s always $0.04-$0.15 per response.

SurveyMonkey - $0.035, TypeForm - $0.045, HotJar - $0.08, Survicate - $0.179

It’s saving *a single record in the database* after a request to the API. Plus wizard for creating the form. Then download the CSV with the results. Also, all companies are charging on a monthly basis. Most surveys goes from time to time. I’m talking from my experience, but there are months where I need to survey 100 users and months where I need to survey 30k users. Why it’s not billed on a ‘per response’ basis?

Do most companies write their own survey software, is there anything cheaper or are you using some open source?

5 comments

LinuxBender2y ago

Why surveys are so expensive?

Not sure. Maybe they have a monopoly on the hosted aspect of it and feel their fancy graphs and analytics make it worth more. The more challenging aspect of surveys is handling email and not being blocked. I would just dole that out to an email marketing company assuming the email addresses are not tied up in a contract that forbids sharing them with a 3rd party and assuming that 3rd party is trustworthy. Many are not.

Do most companies write their own survey software

I've seen it done. Less bells and whistles but it gets the job done. Some employees don't like answering internally hosted surveys for the risk of not really being anonymous, if that was the goal. I suppose it depends on who your target audience is.

is there anything cheaper or are you using some open source?

There are some old perl CGI survey scripts out there but I do not have a link handy. If you don't need all the bells and whistles fancy graphs, analytics then those old scripts may work just fine if you just want structured feedback on something. I'm sure some of the developers here could write a survey form in {favorite language of choice} in their sleep.

Honestly I think just sending a well formed email to everyone and get free-form answers may work for some use cases. Even a dedicated chat channel can work if people do not need to be anonymous. If the survey is for customers I think email or web forms could work but probably takes more time to consume but then you may get more honest answers and more details that the survey may have missed. Customers appreciate human interaction especially if someone responds to them and lets them know if/when their feedback resulted in changes.

neximo642y ago

Because they are valuable.

The prices aren't typically what it costs to make the survey, its what people are willing to pay for them.

Another example might be document signing.

herci842y ago

Hi, my name is Kamil and I run Survicate. It's not as simple as you say and pricing for a survey software might be actually one of the most challenging ones. First of all there is always a minimal cost to serve a customer. In our case we offer 24/5 livechat support for all, and even customers who collect 50 responses might use our support extensively in some cases. Then a response (submission) in one survey doesn't cost the same as in others. To really flatten the cost you'd need to create a pricing based on 'an answer to a question', as a response is a row and an answer is a cell. So a survey with 50 responses and 50 question cost more to store and process than a 3 question survey with 500 answers. But it's not about the number of questions only. Most of customers identify their respondents with merge tags or pass additional user attributes via js they can use to filter results - every single attribute counts as you'd count questions, that's yet additional cell. But that's not all - in case of Survicate we offer targeted in-app (saas, mobile) and website feedback surveys that require a tracking code to run. That generates cost (CDN) even if you don't collect any responses, and if you put it on a high traffic website to run short surveys, the cost of a response must take into account the traffic cost as well. But we're not even half-way through yet :) In case of one-time off surveys the number of responses you collect in a month really doesn't influence the cost that much. But you analyze the survey - the more responses, the more questions, the more to compute. So an NPS survey collecting 1k responses a month is a 12k responses survey after a year - so everytime you click 'Analyze' the cost is 12x+ higher than in the first month. The more often you analyze, the higher the cost. We offer unlimited users - curiosity multipled ;) You want to be up-to-dated with results, so you enable reports to get emails with new replies - that's again grinding through all the responses daily. You want to act on feedback, so you connect Slack, Hubspot, Google Sheets - not only that generates additional cost. Having said that - we're cooking a pricing change that will address the issue of the irregularities in responses volume you mentioned.

And yeah, surveys are super valuable, and if you weight the value of feedback you get against the cost, it's peanuts. Stay curious and dare to ask ;)

Kamil

PaulHoule2y ago

I am LMFAO as the "probability based" surveys like the Gallup Poll that you will find archived at

https://ropercenter.cornell.edu/

are truly stupendous in cost.

michaeloblakOP2y ago

I haven't clarified the root post - I'm talking about the small pop-up on a website or inside an application.

j / k navigate · click thread line to collapse