Baas · Enterprise software

Should I use a BaaS for my enterprise SaaS app?

Anonymous

Last updated on September 19th, 2017

I am a non-technical founder that has hired a dev shop to develop my MVP for a cloud-based enterprise SaaS responsive web application. I have landed my first client and am finalizing the build details. The dev shop is recommending we use a BaaS provider rather than building out the back-end ourselves. I have tried to research the question and have come up against my limits as a non-technical founder.


Any advice would be appreciated. My concerns relate to:


1) how my scalable my product will be beyond the first client if I build it using a BaaS to start?

2) whether angel investors will frown on my use of a BaaS provider

3) the long-term costs of BaaS providers versus building the back-end myself

4) any latency and security issues from using a BaaS provider.


Any other issues I should consider?


Thanks.

Aji Abraham

September 19th, 2017

If you are using a decent backend provider availability shouldnt be an issue. Main concern would be the cost as the some backend providers can get expensive with large userbase. If you are funding the development from your pocket your primary concern should be to reduce the development cost, and BaaS is a good to do that. If you are developing an Enterprise product, the number uses would be in 1000s instead of millions. right? So it should be okay, cost wise. For large userbase your own backend is normally cheaper.


There are apps like Shazm which 50 million plus users using firebase. so angel investors should be concerned. You can always move to public cloud like AWS down the road post funding. You normally get better speed latency with your backend assumng your tech team knows what they are doing. Good luck

Niresh Agarwal Growth Strategist

Last updated on September 19th, 2017

My suggestion would be to use BaaS provider(s) wherever possible and as long as it is not the differentiator part of your offering during initial stages as it will be more cost effective and faster time to market. The things that I would suggest chalked out right from the beginning:


- An architecture that allows easy replacement of one provider to another to in-house solution as business scales

- An understanding of what would it take in terms of resources, budget, and time if you were to develop it in-house

- Understanding of the breakeven cost as users and usage increase over time

- A plan for cut-over if and when the switch is needed


If you have the above as part of back-up slides of your pitch to investors, I think you should be good.


#1, 3, and 4 in your question are more related to evaluating one solution over another and/or the feasibility of any solution at all (e.g. if a BaaS solution does not provide the required latency and security, you should drop that from your short-list right away.)


Please feel free to im me if you need any help reviewing and/or discussing the architecture or a BaaS solution/provider. Happy to help...


Good luck,

Niresh

Pankaj NAURIYAL Looking for a full stack "rock star" developer

September 21st, 2017

Hi Anonymous,


BaaS can actually be a life saver for you. I bet this question which you have posted on the CoFounderlab.com is running through some sort of cloud backend service.


1) Depends on what backend compute tier you start with. Is generally pretty easy to scale with growth in subscriptions.


2) No. They will actually give you credit for keeping your initial capital expenses low and instead paying for services from subscription revenue.


3) You will have to review your financials ofcourse. Depends on your subscription revenue versus your cost to support those subscriptions (spread over all customers) through the BaaS service. Point is, you can start with lower tier BaaS to support the initial customers and increase this as you grow your subscriptions


4) Choose your BaaS provider wisely


I am an angel investor and the founder of a SaaS startup myself and we are using GCP. We are just in our product alpha and don't have any paying customers. But a Google backend (Compute engine, Firebase) has allowed us to focus on our product solution and grow momentum for feedback from perspective customers.


I am sort of technical so understand some of this conceptually but by no means a cloud services guru. So I am at the point where I am actively looking for a full time developer to join our team. Someone who has hands on experience to configure, optimize and scale Google Compute Engine (Our BaaS). If anyone reading is interested, please DM me.


As you grow, you will need such in-house help as well. Good luck.

HANSON, Barry Ever Ready To Learn, Grow and Add Value!

September 24th, 2017

Hi Anonymous, two things if you don't mind.


1. When building a business and "see, know, trust" is so important, is there a reason why you are "Anonymous"?


2. BaaS? Pretty difficult to provide any valuable feedback when there is virtually no details.


I don't expect a reply. I just hope that my two questions may help you crack the code and get more support for whatever it is you are doing,


Best

Hanson

Anil Sharma Founder at Trillo.io

September 19th, 2017

You can use APaaS. There are quite a few low-code platforms. Mendix, Outsystems, Trillo.io, Salesforce Lightning. BaaS may be one of the backend piece. APaaS may be a right platform if you have UI, Security, Data Access needs.

Danny Duong Founder at Titleize.com

Last updated on September 19th, 2017

It's not easy to answer without knowing the detail of your SaaS app.


1) how my scalable my product will be beyond the first client if I build it using a BaaS to start?


It really depends on what your app does. If you don't need sophisticated search, heavy reports, intensive computation, BaaS would be fine. There are a lot of startups, and SaaS apps have been using Google Firebase to scale successfully.



2) whether angel investors will frown on my use of a BaaS provider

I'm not an investor, but as an engineer, whatever gets the job done with the business requirements would be fine.


3) the long-term costs of BaaS providers versus building the back-end myself

Starting with BaaS would help you launch faster (a little, not much), but when you need to scale, it will be more costly than your own backend. Also, you can do less with BaaS than with your backend (in term of functionalities)


4) any latency and security issues from using a BaaS provider.

I think it's easier to deal with security with BaaS than with your own servers. But there're limitations. For example, if you need higher level of security (processing payment, ACH, customer EIN, SSN,...) then you need to look at PCI Compliant. And you have choice but go with your own backend servers.


Edit:


To clarify my answer, when I say your own backend server, it can be either Heroku, AWS, Azure or other cloud based platforms.

George Feil Director of Systems Architecture at Welltok, Inc.

Last updated on September 19th, 2017

I highly recommend going with a BaaS over building your own backend. Most startups go this route these days, and there are many larger enterprise scale companies that continue to rely on cloud services to run their apps. I'll refer to your questions in turn:


1. Regarding scalability, it is much easier to scale up quickly using a BaaS or PaaS. Adding a new virtual machine instance, and bringing it online, can be done in a matter of seconds. Even if you own spare hardware (an expensive proposition), you won't be able to beat this metric.


2. Unless you are in the hardware platform industry yourself, investors will more likely frown on your investing in hardware and DevOps resources instead of focusing on developing your app. Building your infrastructure from scratch is a heavy investment that likely would require a long time to return a profit.


3. See #2 above. So long as your application runs efficiently, and you choose your platform provider wisely, you will benefit more choosing PaaS over building out your own infrastructure.


4. From a security perspective, there's no inherent disadvantage in choosing a BaaS over building your own. So long as you configure admin access appropriately (MFA is a must), and configure cross-service communications correctly (SSL also is a must), and encrypt your storage systems, you can safely run most applications in the public cloud.


As for latency issues, so long as the cloud services you depend on are all located within the same geographic region (AWS makes this easy to do), you can avoid most latency issues.

Jimmy RD

Last updated on September 19th, 2017

question 1 is for your tech team, and dependant on the product you are building.

2) angels should love BaaS. building hooks to back-end service connections (e.g.; Twilio) is a waste of time your developers could use for the next killer feature!

3) you should measure this against the internal cost to maintain your custom-written code. Its the same argument vs. Salesforce, beanstalk, etc.

4) there shouldn't be, if you're using a reputable product. Try MongoDB's stitch!


Curt Sahakian

Last updated on September 19th, 2017

" The dev shop is recommending we use a BaaS provider rather than building out the back-end ourselves. "


I am not a techie. But I have presided over my fair share of project debacles.


I know why the Dev shop wants to do it. But you do not appear to know why you want to do it. And the Dev shop has been unable to explain why in a manner that you understand.


I would turn my back and run. They are not the Dev shop you are looking for and most likely BaaS is not the architecture you are looking for.


Just from your questions it seems evident that your gut is telling you this.


How exactly is BaaS going to deliver a better and more predictable MVP.... FROM THE PERSPECTIVE OF YOUR CUSTOMERS AS OPPOSED TO THE PERSPECTIVE OF YOUR DEV SHOP?



Anonymous

September 19th, 2017

A little off topic, but I'd be interested to know how your experience went with outsourcing the MVP? I've heard good and bad, but if will to share, would love to know your journey!