General Business Forum Brought to you by Salesforce

Creating a web-based app / platform yourself

Discussion in 'General Business Forum' started by Lucky8, Oct 10, 2020.

  1. gpietersz

    gpietersz UKBF Ace Full Member

    1,580 397
    Later, or as a basis for quotes, yes. For now something a lot more bare bones would help. One thing I have often found useful is wireframe sketches of what you envisage it looking like to be a good first step.

    For most web apps I prefer developing an MVP and improving incrementally - going down the "release early, release often" route.

    Very true, people do use a mix of things successfully. For example, one of my clients has:

    1. A Wordpress public website
    2. A static HTML page with a fairly complex JS front end app. This is their subscription service.
    3. A Django back-end for 2
    4. Various other bits to help run it all

    This sort of setup is not at all uncommon.
     
    Posted: Oct 15, 2020 By: gpietersz Member since: Sep 10, 2019
    #41
  2. Lucky8

    Lucky8 UKBF Regular Free Member

    215 26
    @gpietersz, let me ask you now that I've probably tired poor fisicx with my questions - what do you consider the differences between web apps and the websites with CMS which fisicx recommends?

    And to your client examples, when do they use a website with CMS, when a web app, and with regards to the set up I have described, the push-pull interaction, the desktop/phone integration, which route do established similar-to desktop/mobile benefits/rewards systems go down?

    I have wireframe sketches and I'm doing journey mapping. I don't think all this will be the problem (I have experience of this and UX), which is perhaps why I am skipping it a bit for the purposes of this thread.

    What is, I'm afraid, still clear as mud as where this will lead me and what I need to be prepping for now. If it's some training I need for example, I want to get that in now in parallel to the mapping. What skills do I need to start looking for elsewhere. How much will this cost and what do I need to budget for, etc etc. I'm trying to answer these questions, and just mapping it doesn't do any of that.
     
    Posted: Oct 15, 2020 By: Lucky8 Member since: Jan 17, 2019
    #42
  3. fisicx

    fisicx It's Major Clanger! Staff Member

    33,491 9,930
    There is no single answer to this. There is usually combination of systems in use. It all depends on what you want to achieve. And even moreso, your marketing plan.

    For example, if you are going to target FB users through an advertising campaign they will probably only need a mobile interface. You might convince some of them to install an app on their phone. But if you are targeting middle aged dads via an email campaign then you may only need a desktop system.

    So to answer you question...

    It depends.
     
    Posted: Oct 15, 2020 By: fisicx Member since: Sep 12, 2006
    #43
  4. Lucky8

    Lucky8 UKBF Regular Free Member

    215 26
    No, it won't be like that. The reason I gave benefits/rewards as my analogy, is because the set-up is similar too. So companies are the client, company/users will be licensed. All the research into which devices are needed has been done: it's desktop and mobile.

    Doesn't all I've shared give you insight into what I want to achieve?
     
    Posted: Oct 15, 2020 By: Lucky8 Member since: Jan 17, 2019
    #44
  5. mattk

    mattk UKBF Newcomer Free Member

    2,337 855
    You're really getting bogged down in semantics. It doesn't matter if you call it a website, web app, web-based application or whatever. Nowadays almost every website has a database backend, user profiles and serves dynamic content. That's what you want to do.

    I agree with @fisicx, I would be looking at a CMS to sit behind all of this. The key requirements are the ability to have user profiles, serve configurable surveys and allow logic to determine the questions which are asked. Focussing on these three things should allow you to identify some suitable systems.

    I implement HR technology for a living and have seen most major employee benefit solutions. They are generally very complex as employee benefits tend to be. I do wonder if you're biting off than you can chew. Have you actually researched the market to ensure you're not entering a very crowded area and spoken to companies to ensure you have a product that might actually have some demand?
     
    Posted: Oct 15, 2020 By: mattk Member since: Dec 5, 2005
    #45
  6. fisicx

    fisicx It's Major Clanger! Staff Member

    33,491 9,930
    Not at all. It’s still very woolly.

    It would help if you posted a link to an example of something similar.
     
    Posted: Oct 15, 2020 By: fisicx Member since: Sep 12, 2006
    #46
  7. billybob99

    billybob99 UKBF Regular Free Member

    1,795 608
    As with anything you need to have a decent pot of bunts (£££) you are willing to risk, invest, spend - however you want to word it.

    Better use of your time is to actually get to the point where you know this idea will fly and people will pay for it, then worry about building it, not the other way around.

    Then you can go to bed happy knowing you have a list of people who are ready to open their wallet as soon as you say open sesame.

    Chances are when you market it, real world results won't match your anticipated results.
     
    Last edited: Oct 15, 2020
    Posted: Oct 15, 2020 By: billybob99 Member since: Apr 23, 2013
    #47
  8. Lucky8

    Lucky8 UKBF Regular Free Member

    215 26
    OK, thanks for the replies. I skipped the whole question of "is there a market?" and "have you done your research?" because it wasn't a query I was posing in this thread or is an area that I needed any help with. Perhaps I should have made that clearer. Yes, I've done all that, it wouldn't enter my mind to look into this or pose a question on here if I hadn't! By the way, this is not an employee benefits model, I said that the platform would be like one in terms of functionality.

    So, here's an example of a much bigger but similar platform in terms of functionality: https://www.rewardgateway.com/uk/solution/employee-benefits
     
    Posted: Oct 16, 2020 By: Lucky8 Member since: Jan 17, 2019
    #48
  9. gpietersz

    gpietersz UKBF Ace Full Member

    1,580 397
    One thing you could do is start with a website that just lets people enter the information and process it manually to start with and then add more automations as you go. I have read that this is how Groupon started. Its a bit "fake it till you make it" but as long as you actually provide the service promised I do not feel its unethical.

    In terms of the functionality it looks to me as though most of your functionality is custom rather than what you can expect to work out of the box with a CMS. Its precisely because I develop things like this that I switched from Wordpress to Django many years ago (after trying several other things, including some classic NIH developing my own CMS and web app server).

    A CMS will give you content management out of the box - its does what it says on the tin.

    A web framework will give you the useful tools a CMS would (use profiles and authentication, and admin UI, etc.), may have third party code you can reuse to do some of what you want (pick something with a good ecosystem) and will be far better suited to developing custom functionality.

    In your case I would suggest looking at something you can learn to code yourself - take a look at the major frameworks. Even if you just get started and hand it over to to someone else to develop long term, the familiarity will be useful in managing it (hiring the competent people and checking code quality). One of the biggest problems I see is non-IT people managing developers end up with no idea that they have an unmaintainable code base.
     
    Posted: Oct 16, 2020 By: gpietersz Member since: Sep 10, 2019
    #49
  10. fisicx

    fisicx It's Major Clanger! Staff Member

    33,491 9,930
    That uses a range of different platforms to deliver the service. They have a website to promote the tools, another website/web app to administer the service, mobile app for users and a whole load of server side stuff to make it all work.

    As has already been said. There is no single solution. Whatever it is you want to deliver will almost certainly need more than one answer.

    Reading the rewardgateway case studies it seems each one was bespoke. There wasn't a one application they sold to everybody. And they spent a lot of time researching each client before writing a single line of code.

    Which brings us back to you. We still don't know what it is you want to do.
     
    Posted: Oct 16, 2020 By: fisicx Member since: Sep 12, 2006
    #50
  11. Lucky8

    Lucky8 UKBF Regular Free Member

    215 26
    @gpietersz - you're spot on with the "fake it 'til you make it' approach. In some ways behind the scenes that's what I thought we might have to do, and I'm comfortable with that.

    To help me understand - and a big thank you to you, @billybob99, @fisicx and others for taking the time to go through this with me - could you elaborate on some points you've made and what you mean. If I quote each bit I'm not clear on it might be easier and it would be great if you could refer to the rewardsgateway.com example I gave as we're both looking at the same thing then.

    In tangible terms by looking at rewardsgateway.com, what do you class as custom and what would be out of the box CMS? Note, they sell a package that is 100% off the shelf, and a package that is totally customisable and by that I mean they revamp the UX, which pages load when etc with a client for a few months. The engine I presume stays much the same. They provide a main website and an app can also be used for some bits.

    I'm not sure if you're referring to all this tailoring for clients when you refer to custom functionality or if you are saying the functionality *I* want is custom? I'm considering offering a lightly tailorable off the shelf product to start with.

    Back to the example, if a CMS was used, what would be a real-life consequence of that? ie so that would mean a user could do abc on rewardsgateway.com but not xyz? Or information can only be served this way, but not that way to a user. Or is it more about our ability to do things with the platform, designing and maintaining it? I'm having trouble translating the characteristics of the various technologies into their impact and implications for the purpose of the platform.

    In terms of "web frameworks" is this the same as website themes? Or the no-code platforms like Bubble which was mentioned previously? Any other recommended web frameworks?

    I definitely know this is not something I want to maintain, this will be handed over to others to refine and grow while I concentrate on the business. I just want to a) get a prototype up and running and b) be able to move swiftly to get a full (if early) product developed to be used by our first few clients. No code or little code routes to get us up and running would be great.
     
    Posted: Oct 16, 2020 By: Lucky8 Member since: Jan 17, 2019
    #51
  12. fisicx

    fisicx It's Major Clanger! Staff Member

    33,491 9,930
    This is really difficult to do without a lot of resources. I've been developing wordpress plugins (that function as web apps) for years and nobody wants to tailor anything themselves. They give you a spec and pay you to make it so.

    There isn't ever going to be a no-code way to do this. Everything you want to do will need coding. Even a billy basic version is going to need someone to develop the functionality.

    To give you an idea of an off the shelf product you could customise. Years ago I was part of a team that built a coupon platform. The idea was small shopkeepers could sign up and join in a national coupon scheme. They spent just over £100,000 on the application and mobile apps and over £50,000 on marketing. It failed in less than a year. Everyone thought it was a great idea but it just didn't get any traction. That's the sort of costs you will be looking at.
     
    Posted: Oct 16, 2020 By: fisicx Member since: Sep 12, 2006
    #52
  13. Lucky8

    Lucky8 UKBF Regular Free Member

    215 26
    All the information I've given doesn't seem to be explaining this to you. Would you like to try asking me the questions that you'd like answered?

    No, not tailored themselves, tailored by us, only at the UX level.

    What is it about the no/little-code development platforms that wouldn't achieve this?

    OK, so that example with a cost is helpful as an illustration. So talk me through in a summary about the elements to that coupon platform, and why it cost £100k. Thanks.
     
    Posted: Oct 16, 2020 By: Lucky8 Member since: Jan 17, 2019
    #53
  14. Lucky8

    Lucky8 UKBF Regular Free Member

    215 26
    P.S.

    They have a 100% off the shelf product and that's closer as an analogy.
     
    Posted: Oct 16, 2020 By: Lucky8 Member since: Jan 17, 2019
    #54
  15. fisicx

    fisicx It's Major Clanger! Staff Member

    33,491 9,930
    You say it's not employee benefits, what is it you want to build? Who are you targeting? How do you plan to market this thing?

    You say users set up profiles and answer questions. Are they quizzes, surveys, conditional forms?

    What happens on completion? Do they get a prize?
    Because they are limited on functionality. They give it away for free knowing you will need their paid help to do anything useful
    8 developers on £400/day plus graphic designers, content managers and others (I was employed for my UX skills). That 100K was spent in a month.
     
    Posted: Oct 16, 2020 By: fisicx Member since: Sep 12, 2006
    #55
  16. Lucky8

    Lucky8 UKBF Regular Free Member

    215 26
    If the functionality is so similar to the example I gave, and all the ways I've described how it works, how I market it etc shouldn't make any difference surely.

    This is a small function which we might not do, although the user profiles part is mandatory anyway. If we did, they'd be multiple choice quizzes. No prizes.

    Can you give some concrete examples of these limitations, referring to the set-up I have described?

    Ah sorry, I meant the elements of the work needed, rather than the resources. ie. What needed to be done by the resources you listed.
     
    Posted: Oct 16, 2020 By: Lucky8 Member since: Jan 17, 2019
    #56
  17. gpietersz

    gpietersz UKBF Ace Full Member

    1,580 397
    I agree with this. The functionality is too custom and specific to do with a no-code solution - and I do mean the functionality you want.

    Maybe, but depending on the functionality required and the details if could be a lot more or a lot less. I think picking the right platform to start with and starting with an MVP would make a big difference.

    No, to clarify, I am talking about backend web frameworks. They are essentially a web development libraries plus some useful code to make it easier to setup and manage. A web frame work is a base on which you might write a CMS or any other kind of web app. They can also provide a backend API for mobile apps and similar. Here is the tutorial for the one I use: https://docs.djangoproject.com/en/3.0/intro/tutorial01/ I love it and its been what I have used for most of my web development work.

    Other alternatives I think are worth looking at are https://guides.rubyonrails.org/ and https://laravel.com/docs/8.x

    Its hard to be sure without a demo, but, if you look at the "employee communications" features page, its stuff like "customised blogs" and "unlimited pages". That is functionality you would expect a CMS to provide.

    On the other hand things like surveys and vouchers are not standard CMS functionality. You might find CMS plugins to do those, but I think you would need to be lucky to find ones that do what you want, which means writing your own, which then means you would be better off with a framework.

    You could also pick a framework that has a CMS written in it, use the CMS to get started quickly and then write your custom code in the same framework. Its easier to integrate that way.

    You could also use an entirely separate CMS and integrate where necessary.
     
    Posted: Oct 16, 2020 By: gpietersz Member since: Sep 10, 2019
    #57
  18. fisicx

    fisicx It's Major Clanger! Staff Member

    33,491 9,930
    Yes it does! How you market and to whom you market drives the whole UX/UI
    This is all going to be bespoke code
    A lot of meetings to develop the framework, library integration, UI development, workflow development, thousands of lines of code and everything in between.
     
    Posted: Oct 16, 2020 By: fisicx Member since: Sep 12, 2006
    #58
  19. mattk

    mattk UKBF Newcomer Free Member

    2,337 855
    It doesn't really work like that. What a CMS will give you is a lot of the backend code, such as a database, business logic and a lot of the front end for standard activities, such as a user creating a profile, updating it and so on.

    Think about the standard CRUD operations for a user. In order to create a user profile you need database tables to hold the values, you need a web page to collect the data and you need the business logic to take the user's input, validate it and pass it to the database.

    A CMS gives you all this for free. Without the need to code all that stuff yourself.

    You will then find some that have very specific features, such as the ability to create surveys, which seems along the lines of what you're trying to achieve.

    The other benefit of a CMS is that it allows you to produce content, in a consistent format, without having to hand code pages.

    For a very simple example, have a look at surveymonkey.com. They allow you to create a survey, without the need to write any code. It includes the ability to add logic, so you can skip a question based upon an answer and so on.

    The flip side is once you have configured your survey, surveymonkey.com will create a web page you can display to your users and it has all the standard functionality you'd expect, tickboxes, free text and so on. Again, you didn't have to code any of this.

    Now, I wouldn't call surveymonkey.com a CMS. As I said, you seem to be getting bogged down in semantics, rather than focussing on the features and functionality you're trying to achieve.

    My advice would be to focus on the features you want. Then do some research to find off the shelf tools which can do them. For example, this link shows 10 free survey tools. From there, you need to drill into the each to find if they offer the ability to have user profiles and so on until you have a shortlist of tools which meet your requirements.

    Don't get me wrong, I'm not saying it is easy and you may well have to compromise (which is why most large web-based applications are custom built), but for an MVP this is the quickest approach.
     
    Posted: Oct 16, 2020 By: mattk Member since: Dec 5, 2005
    #59
  20. Lucky8

    Lucky8 UKBF Regular Free Member

    215 26
    I guess we just disagree on this point. The UX is based on our research and client requirements, user behaviours and journeys etc, but isn't driven by our marketing stategy. Maybe we're talking about different things?

    I had a quiz on a simple WP website once, it was just a plugin. Hardly any coding required. Is this so very different?

    Translated into the customer's perspective, rather than the developer's project plan, where did the 100k get spent, on what components? eg. 10k to create a coupons database which looked like ABC. 10k to create the shopkeepers database which meant DEF. That kind of thing.
     
    Posted: Oct 16, 2020 By: Lucky8 Member since: Jan 17, 2019
    #60