Custom Cloud Solution for Enterprise Scalability

We provided enterprise scalability and an efficient, robust custom cloud solution for an online ticket reseller. Our solution eliminated more than 98% of the load on its inventory system, or more than 300 million transactions daily for optimum enterprise scalability.

Summary

The Problem

---

How We Helped

---

The Outcomes

How we Helped

UX Design + Prototyping

---

Development

Gallery

Client Review

Background

Introduce your business and what you do there.

I was the chief technology officer for an online marketplace based in Austin, Texas. We connected buyers and sellers to offer a kind of standard retail experience. We internally ran all retail e-commerce, wholesale, and inventory management, and we had a full integrated supply chain with about 2,000 brokers uploading items to us for inventory. 

---

Opportunity / Challenge

What challenge were you trying to address with Praxent?

We hired Praxent because we had a couple of projects that we wanted some help with. They were able to help our internal team with our supply chain in order to speed up throughput and process some CSV [comma-separated values] files.  

---

Solution

What was the scope of their involvement?

Praxent ended up delivering a web service that became the new entry point for all inventory coming into our system. The system existed before, but it wasn’t scaling. We were only able to take an inventory update from each individual supplier every 5 minutes, but Praxent helped us remove that bottleneck altogether.

How did you come to work with Praxent?

If I remember correctly, I came across Praxent while looking at different consulting firms in Austin. They had something on their site that mentioned expertise in open-source tools, which was a conscious decision that I was taking to move my team away from the Microsoft stack. We were considering other companies at the time. I can’t remember which ones those might have been, but I know that we did talk with 3–4 others.

There were two factors that ultimately sealed our decision. One was when I first sat down with Kevin [Managing Partner] from Praxent, he seemed to have a good sense of the scope of what we were trying to do, and I think he had a good perspective on trying to control that scope and not let things get out of control. I think he also saw quickly what I did, which was that if we address this particular service properly, we would essentially set up an entirely new architecture and data pipeline that we would be able to evolve the overall architecture of the business onto.

How much have you invested with them?

We invested more than $75,000 but less than $100,000 with Praxent.

What is the status of this engagement?

I think the project took us 3–4 months over the summer of 2016. 

---

Results & Feedback

What evidence can you share that demonstrates the impact of the engagement? 

I don’t know how I could quantify Praxent’s success. I actually left the company at the end of March, so I’m fairly disconnected from the current results. There was some tweaking after their delivery on our side to get it exactly where we wanted, but I think it helped us bring on more and bigger suppliers. Their work certainly helped augment the overall inventory for sale.

I think Praxent was probably one of the fastest companies to go from a concept to a rough prototype. It only took them a few weeks, and that was a collaborative effort between our teams. We had an extended day or two at the frontend, and then we would check in every other day. They were trying to get to the point where we had something working. Once we had that, we realized the model would work going forward, and it was just a question of fine-tuning it and working out all of the deployment issues.

How did Praxent perform from a project management standpoint?

I think we had 2 people from Praxent’s side working on the project, plus administrative folks, so you could probably call it 4 altogether. We interacted with them probably 3 times a week. They had a project lead named John. We used Slack and JIRA to communicate, and I thought both were pretty effective. I like how they did a code review every week of what their changes were; after the review, they showed us the working code in demo scenarios every week, and that really allowed us to iterate quickly. We could say, “Okay, I see what you did there. What if you did this?” Both I and my team liked that and thought it helped a lot. Chris, the project manager we were working with, did a good job of setting the scope for the next week at the end of those sessions. Collaboratively, we would agree on the priorities for the next 5 working days, and then we’d come back and see where they got based on that feedback.

What did you find most impressive about them?

I think Praxent was as good as or better structured than anybody else I’ve worked with. I’m not sure if that makes them unique, though. I didn’t have a lot of visibility into the other parts of their business practice, so I can’t say for sure.

Are there any areas they could improve?

We probably should have done more load testing on the frontend to validate the model because there were certain things they implemented that we had to change internally after the fact to get them to work.

Do you have any advice for potential customers?

Maintain daily communication until you have a working prototype. I would also recommend limiting the scope of the project as much as possible to get to that first working prototype. In subsequent engagements, I’ve seen the scope creep so much that you don’t actually ever get to anything working. Nearer-use cases that will solve the majority of the problems you’re trying to fix.