Ask HN: SaaS vs. build it yourself when launching a startup

In general, you are never going to want to reinvent the entire wheel. I would say anything that is &quot;core&quot; to your business is probably best to build in house while you can use existing saas to help with the rest. Lets break this down a bit for your use case :<p>- Local delivery: There is quite a bit of offline work involved in partnering with the companies etc to deliver for them (i m guessing, not an expert in local delivery). So just wanted to throw this out first.<p>- Delivery logistics using software: This is probably the core of your business in my opinion. So you may want to build this in house. For example, the entire UI&#x2F;UX that a user will face to get local delivery via your app&#x2F;website.<p>- Other things like payments etc, use existing tools of course like stripe etc.<p>Think of it this way. If the &quot;SAAS&quot; that you use instead of building in house goes out of business, will your business completely collapse or will you be able to move on to something else (yes it will still be painful i m sure) ?

What do you believe your core competency is? Focus on that.<p>If you don&#x27;t know, then use as many pre-built parts as possible; that will allow you to get out and interact with customers ASAP. Then, use their feedback to guide what you need to do better.<p>Once you know what your customers aren&#x27;t getting, but they want, from your service or your competitors, solve those problems.

You can build out many pieces in time as needed. The question I ask is &quot;what will grow my startup the fastest&quot;?<p>It often makes sense to use a &quot;good enough for now&quot; SASS solution while you focus on more critical pieces of the company. As you grow, you can migrate off of it.<p>For example, you might want to consider using Trakin [1] to jumpstart your delivery logistics codebase while you focus on the human side of deliveries and acquiring customers to deliver from &#x2F; to.<p>Then, as you grow, if Trakin doesn&#x27;t meet your needs you can replace it with an in house solution. Or you can pay for enterprise support and functionality. Or you can switch to another system. Whatever solves the current problem and helps you grow fastest!<p>[1]: <a href="http:&#x2F;&#x2F;trackin.co" rel="nofollow">http:&#x2F;&#x2F;trackin.co</a>

Local delivery - seems worth doing yourself to better understand all customer nuances<p>Payment processing&#x2F;other existing website tech - buy instead of build

I&#x27;d be happy to share my experience as I&#x27;ve been talking with a lot of delivery companies as the founder of <a href="http:&#x2F;&#x2F;trackin.co" rel="nofollow">http:&#x2F;&#x2F;trackin.co</a><p>What I can say is that 70% of the basics regarding E-commerce&#x2F;local delivery is the same whatever your market is (tour optimization, tracking, driver performances, dispatch, send orders to vendors...). Then there&#x27;s often specs, based on what your business is exactly doing. That&#x27;s why we decided to have a focus on food delivery at first, because we can answer every need you may have, from online ordering (adding pickup and catering) to customer relationship and promotion(adding yelp or G+ reviews) to what the drivers need to know regarding their delivery for that specific.<p>Meanwhile we&#x27;ve built our solution in a flexible way, and with an API, because a lot of things that we&#x27;ve built take a lot of time and expertise and we know other delivery businesses need it too. (And it&#x27;s often hard to realize it before you actually start building it in house...)<p>So objectively if I had to start a food delivery business, I would use Trackin (obviously), and if it was a different service, I&#x27;d still use a (good) E-commerce solution + an API from a trustable 3rd party to help me manage the deliveries. That could be Trackin on something different based on your needs. A company with a product that will evolve faster that my own product and would save me 2-3+ full time engineers and 6month at least.<p>You can still do it yourself but in the end you&#x27;ll be building a dashboard, with a catalog, inventory, delivery management, dispatch, driver app then a customer facing app VS only build a customer facing app, and get the rest from a Saas, customize it so that you can focus on your business. What I like to consider is the following: Why would I spend $$ and time on something for my business if I can have a competitor doing the same thing as me in just a couple of days&#x2F;weeks thanks to SaaS solutions? Where will be the value added of your business?<p>Hope that helps and happy to discuss more if needed!

My advice is to only depend on commodities, not services. a commodity is something so generic and interchangeable that you can hardly distinguish between competitors. In this way, every external thing you rely on is a potential single-point-of-failure, but can be easily fixed.<p>If the thing youre depending on is unique and essential to your business, you&#x27;re taking a risk because your success depends on their continued success and if that service goes down you go down with it because it cant be easily replaced.<p>TL;DR You dont have to reinvent the wheel, but for all of the off-the-shelf products you use make sure there are plenty you can replace it with at any time.

Frankly, SaaS bills add up. The things with all these subscription fees, is they&#x27;re like paying in installments -- it numbs you to the bottom line, you end up paying anyway.<p>Here&#x27;s an example, $9.00 isn&#x27;t much for Photoshop. For it&#x27;s $120.00 a year. That $30.00 server that barely dents your budget? $360.<p>If you have a programmer that has devops chops in chef&#x2F;puppet&#x2F;salt&#x2F;ansible, get on his back and have him automate everything on cheap cloud instances.<p>Chef recipes and such as genius because you consolidate your work in declarative recipes.

Very interesting question I&#x27;m curious to know about this as well, because I just found that there is a company that sells delivery management as a service.<p><a href="https:&#x2F;&#x2F;onfleet.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;onfleet.com&#x2F;</a><p>However, to answer to your question in more detail I would say:<p>User management &amp; Delivery management&#x2F;tracking.<p>Those two are very crucial parts of any SaaS and can be tweaked in many ways towards the business.<p>The tracking can be either cheaper or more accurate etc.<p>And the user manipulation have to be dealt with inside, regardless some generic stuff, there are always industry specific user management stuff.

I think the advice usually is to build everything yourself.<p>Local delivery? Seems like a saturated market. How are you going to compete with Amazon?