George, this is a goldmine of a post, thank you!
I’ve been thinking along very similar lines as to how to manage new projects going forward, even landing on the same price point.
Good for you making it work and sharing your learnings!
6 Likes
Keeping applications open until Friday 
Please follow the instructions in the post, most don’t!
1 Like
This has been a super interesting thread to read! I came across Designjoy months ago and wondered if the same model would work with Bubble. I’m curious to see how far this model can be scaled with Bubble.
Also, I really appreciated the inputs from both @georgecollier and @boston85719 and I’d love to see more discussions related to the business side of building apps on Bubble.
3 Likes
Do you have a way to handle clients who have built their app/MVP and now want someone else to take over the further expansion of features so they can focus on the business side of things? I suspect that might fall within your subscription model, but I’m also mindful of that fact that you may not be familiar with the app has you hadn’t built it first iteration, so that may take some getting used to.
1 Like
If the app is well designed and well built, put them straight onto a subscription plan.
If it’s not well built, suggest a $10K rebuild whilst preserving all of their data. Alternatively, if they do not want to rebuild after explaining the benefits of that and fully understand they’ll be building up technical debt, put them on the subscription plan with that acknowledgement.
like the way that you think to build solutions for clients efficiently, would like to work with you already applied in the form waiting for you response.
OK that makes sense.
If you’re charging $999 per month for the starter plan where you work on additional features and app expansion one at a time with up to a 72 hour delivery, at the top end (assuming all app features require a 3 day turn around time) - that’s about 10 new features back to back each month. Another way to see it is $100 per (complex) feature, or less for those less complex and require less time.
You have mentioned as an example of one way to the subscription based model:
hey, can you add a new filter X on the search page’, they can subscribe to the $999/mo plan, wait 48 hours for me to turn it around, and then pause it, saving the remaining amount for later. This avoids me getting tied up negotiating (and working on) dozens of $50 requests!
Isn’t it easier to just charge (and categorise) new features within 2 or 3 levels of complexity and bill ($50 or $100) for the one off feature rather than ‘pausing’ and having this accumulation of ‘time’ rolling over from one month to the next? If clients can ‘pause’ the clock on their subscription, it seems like they’re simply crediting their account with you and over time this is debited with the features they request. You will also start to the build a database of what is categorised at what level of complexity and what was charged for it for new agency staff to follow as a guide. In addition, it takes the pressure off delivering it in a certain time frame, but you could agree a threshold for turn around of new features.
2 Likes
Because MRR is king 
Assuming clients always have a backlog of requests (which they don’t).
So, why subscription over pre-paid (or even just 3 pricing tiers for feature complexity)?Subscription has a couple of benefits:
-
MRR. No need to actively ask clients to topup. It auto-renews until they cancel so it’s way more predictable. No invoicing admin for me as it’s handled automatically, and the client doesn’t have to pay. Limit the barriers to getting paid by just have their payment collect itself.
-
Psychologically, clients like that they can have one outgoing payment that they can predict. They can plan ahead for 6 months and know they’ll spend $6,000 to have virtually any request fulfilled. Anecdotally, clients prefer that to metered/pre-paid billing with different preset pricing for feature complexity - even when it might cost more. Predictable and plannable pricing reassures clients.
-
On that psychological note, when creating requests with a prepaid but fixed price model, there is some degree of restraint when making requests. Each request does cost $100/$200 etc. Meanwhile, with a subscription, the client doesn’t think about that so much. They pay for the subscription first, and then make as many requests as they want to make the most of the subscription. With fixed one-off pricing, clients have to think about cost every time they make a request. With subscriptions, the only time they think about cost is when they subscribe for the first time.
Let me know what you think! Appreciate your thoughts on this @js7319 
2 Likes
@georgecollier great post. I’m just adding my perspective. I agree that a Bubble MVP should never be 6 figures (or even approach it).
With that ceiling in mind, the real discussion is what constitutes a true MVP?
If it’s a landing page + minimal features, that obviously < 10k. BUT let’s keep in mind that the point of an MVP is to test assumptions and hypotheses related to the product’s value prop.
To truly test assumptions, one has to build a product that includes enough of the proposed value so as to truly test an assumption. What that looks like and what it costs are in an extremely wide range (let’s ballpark that at ~3k to ~50k).
For example, one of my recent projects was an MVP for a home health aide job listing app.
The value prop was that patients (often older and not tech-oriented) would be more likely to find and commit to an aide if they could more easily find appropriate matches (without having to reach out to determine if the aide’s schedule worked for their needs), and that caregivers would be more responsive if not bombarded by irrelevant requests.
To test this value prop, I had to:
- store the daily schedule for aide and patient and upon the patient’s search, determine which aides were available for 75%= of the patient’s “schedule”
- exclude aides that were unwilling to travel the distance to the patient (based on their preferred means of travel and max travel time)
- create a LinkedIn connection mechanism so that aides contact info wouldn’t be shared with too many patients (patients only received aide’s contact info upon connection and could only have 2 outstanding connection requests at a time);
- sort listings by aide’s avg response time so as to reward prompt responders (which would make the patient’s UX immeasurably better)
- notify patients when there are new results matching their criteria.
My point: Building an MVP without these features wouldn’t be a MVP for this value prop while building an MVP with these features (overlaid on top of a job listing app) would likely cost more than 10k…
1 Like
Completely agree.
The example you’ve given sounds like something I’d be very reasonably able to include in the $10k. For reference, I shoot for 80 hours of hard building in that MVP (leaving aside time for admin/scoping etc). Obviously with this model you can have multiple MVP packages (a mini MVP and a… less mini MVP?) although I decided against it as it’s a really nice branding point to be able to say we just have one price.
On an unrelated note, my mind still gets blown that companies pay $50K for a relatively small Wordpress site while I’m charging $10K for a marketplace MVP 
I need to learn how to connect with enterprise/traditional services businesses that could improve productivity with some tailor made internal tools… anyone’s advice on that is appreciated…
I certainly wouldn’t provide 80 hours of my time for that amount (although a typical MVP would only need ~18). Also those features are examples of requirements to test any assumption but are on top of the other features for a job listing site (e.g., API integration with their custom DB) and since it was built in Bubble, HIPPA compliance takes a substantial amount of work.
True but that’s apples and oranges; not for a MVP and more paying for content and layout etc
To me, that’s a recipe for super unhappy clients. I agree with you on the issues with wireframing and scoping (went thru it a few times with airdevs and the like, and the back and forth is torturous).
@georgecollier I’ve played around with building a scoping intake which would guide the user thru all different types of options, screens and features and then spit out a quote / timetable / wireframes instantly (or pretty close to it) but that would require a robust freelancer marketplace to make it worthwhile.
and using that definition of a MVP, for the p2p marketplace specs you provided, that MVP can be done for <10k but its also seems too vanilla to actually test any assumptions or value props.
Note: one of Bubble’s biggest weaknesses is module compatibility from app to app (how many folder names have to be individually updated upon pasting a reu?! 
) and a no code solution that handled that better can pave the way for a 5k flat MVP rate…
1 Like
Bubble apps can’t be HIPAA compliant full stop so if you’ve promised anyone that then you might need to walk that back 
Because the example was generic This was pretty much project scope for a site I did for a P2P chef marketplace - they were testing if customers wanted to buy from local personal chefs and it allowed them to do exactly that!
The client still contributes to the scope. We just meet with them and help put their words into writing. It’s not like we hear the client’s ideas and twist it into something vaguely similar that can be done for $10K. The client’s involved as much as they want to be, but if they can tell us their vision, we can get a pretty solid draft scope for their review which tends to only need a couple of changes!
Already delivered a fully HIPPA compliant solution on bubble (entire MVP was delivered in under 3 weeks).
for such a value prop a simpler MVP can test their assumptions.
What kind of workarounds did you have to do? External database? Any 3rd parties inspect the app from end-to-end for data compromises/noncompliance?
This thread comes to mind on why I am doubtful: For Hire: HIPAA Compliance Specialist for Bubble.io Mobile App - #8 by J805
2 Likes
Stop 
It’s been stated, by the owners themselves, you shouldn’t use this platform for an app that requires HIPPa compliance
4 Likes
Stop? it’s already been done and isn’t that difficult, realtively.
We have a new forum master but only if you can prove your point
Not to be a naysayer or whatever, I call bs unless you only used bubble as a front end solution and passed absolutely 0 data thru their workflows or api connector. If this is the case, I would most likely never use bubble in the first place. Weweb or ftw
I mean, if you did dude you might as well sell a Udemy course and double your cash. Cause nobody else knows how to do it with bubble and if they do they have, up to this point, been tight lipped
1 Like