Software as a Service September 14, 2020

B2C Pricing, tiers and free trials.

PayOffWizard

Hi, we are starting to launch the subscription version of our B2C SaaS, and since I've never done this before I would love to get some feedback on what we are doing.

So we are building www.BuyLocalized.com, it's a Made in USA filter for Amazon. It doesn't work great at the moment, the thing is we have to do a lot of manual research. To make it work great will be thousands of hours more and I don't know if we want to try and raise money to do that without serious validation. Due to that and my experience in digital marketing I figured we will make a big push to sell it now, while freezing building it, and then take stock afterwards.

Currently we have 3 tiers. A free unsigned in tier where people get access to just a few results. A signed in tier, where people who have given us their email address get access to a lot more results, and then a paid tier where they get all results, and access to a modal with more results and some filtering options.

We also have a free trial for the third tier. People who sign up for it give us their credit card right away, but their first billing period is 6 weeks, and they can cancel within the first 2 weeks no questions asked. Most people sign up for a year, and they get that same 2 week free trial.

We have those 3 tiers because of the value of an email address. The only way we have of contacting our tier one people is a little announcement bar we can put links into, getting the email address of a user allows us to much more easily convey our message and try and upsale them.

Our yearly plan is $24, and our monthly plan is $3. The pricing is a large issue. With the cost so low we will be dependent on either an absurdly cheap cost for acquiring free users, or a much higher premium tier adoption rate than I read is standard. We did a pricing survey and around this is what seemed to fit our beta users expectations.

So what I've written out, does it make sense? Is this a logical set up? For the moment we are concentrating on turning cold traffic into tier 3 sign ups, we are finding out if we will be able to do that. If not we will spend more time on upselling our free users to paid. Neither of us have ever built a SaaS before and I worry we might be messing this part up.

(We are launching our new website in a few hours, I know, our current one has issues!)

  1. 1

    Just go for it, you'd learn more after, things no one can guess up front.
    If your list is large you can do a staggered release seeing your conversions and such and thinking up a/b experiments you can do with the next groups... if the list is small, just don't sweat it.
    You can try raising prices after you have a baseline...
    Surveys aren't a good indication for pricing, until someone actually tries to pay behaviour isn't predictable, you'd usually get optimistic response vs actual buyers (many more say they would buy by far than would actually buy) and people voting for a lower price than they actual pay (from the ones that would actually buy)

    On the numbers side there is a high probability you can charge 29.95$ yearly and get the same amount of buyers as 24$
    For the 3$, 4-5$ would also probably get similar results
    (Context is psychology of pricing... people round numbers and such trickery...)

    1. 1

      Hi! We had some good discussions in the past, you had some great feedback.

      We tried paid ads a bit 2 weeks ago, they didn't quite work. The results weren't awful but they weren't profitable. One thing we've found is having that plans page as the 2nd step in the funnel instead of sending people to the Chrome store resulted in a huge drop in free sign-ups. We are starting round 2 today after the new website is ready. We don't have large enough lists to do a/b tests really.

      We did this survey: https://www.qualtrics.com/marketplace/vanwesterndorp-pricing-sensitivity-study/

      I was surprised how well it went. We had a pretty good response rate and the results were fairly consistent.

      I agree with you on the pricing at $29.95. We are really eager to get going, so we are launching ads round 2 today, and in 5 days we will make a series of changes to our pricing page. I will include changing the price then just so I can keep these tests more binary.

      1. 1

        There is a lot of details into running ads effectively. You can group 100 clicks as a test rather than wait a week if your trying still to get just initial information assuming your ad source is specific and consistent enough.. good example is a Google search. Bad example is Facebook. ...
        You can always bid less if there is enough supply... Like you don't have to see yourself as number one and try to take the pot...
        Split the traffic more granular to learn more .. depending on flatform more specific terms, locations, times...

        1. 1

          I can only use Facebook, can't target by browser in Google ads, and we've only got a solution for Chrome.

          I've got a lot of experience with FB ads and in the last test our results there had some hope, though not enough to keep it on without changes.

          1. 1

            I hate FB giving false hope with more quality users at the start, very bad for new advertisers...

            Are there bussiness development opportunities with pro USA bussiness initives?...
            Your targeting like group members of buy local USA groups? try bidding lower... Well I'll stop giving you fb ad advice .

            Surley there are different groups on different platform how would find this interesting and you'd be able to put a link for free... and some people who just promote the topic...

            1. 1

              Ya, it's annoying how FB does that, doesn't make much sense to me.

              I haven't really tried to get into business opportunities because I'm not American, I think it could come off a bit weird. I do however have experience in B2C, so the hope is we can make something so good that it doesn't matter where I am.

              The issue with organic is that we only have a Chrome extension. Less than half of internet browsers are on desktop, and only 58% of those are on Chrome. It means it's really hard to get traction in place like Reddit and FB, because you only have a solution for a quarter of the viewers. It also makes content creation less effective, unless I could find a way to narrow down not only by a desire to buy American/Not buy Chinese like we currently do, but also by browser, and I haven't figured out a way to do that.

              I've posted extensively in groups and stuff, and gotten hundreds of users from them. But the extension doesn't work well enough now to get recommended a lot, and I don't want to spend the money to get it there without more validation than it currently has. In a perfect world we can get new paid users through paid ads, and that would provide the validation to expand our project to get it to a great spot in a year or so from now.

              1. 1

                Paid users should be the same if they come from paid ad or from a free post ^^.
                You should probably consider setting up a US company if money wise it make sense... it is weird a non US person and company pushes us consumption for us users a bit..

                content wise you can configure the site at least to target us traffic only in google website owner tools... whatever the name was...

                is there no easy converter to publish the extention to the other browsers even if it's not to the same quality level?..

                1. 1

                  We've a US entity and have been waiting for a Tax number, COVID really slowed it down. We have received it though this week and I just need to do a bunch of phone calls.

                  Because of how our extension functions it's not quite as trivial (not allowed to be involved in 3rd party searches in Firefox). We can still do it, it's just not as trivial as usual.

                  Honestly I can't remember why we haven't done it by now. We've just been so swamped in clear improvements for the extension and then trying to market it, but still would have though we could fit it in.

Recommended Posts