Alteryx

BACKGROUND

Can you please provide a brief description of your company and the role that you play there?

My company is called Marquee Crew. I founded it this year to provide professional services in data blending and advanced analytics, primarily through the use of Alteryx. I am the founder of the company.

OPPORTUNITY/CHALLENGE

What was a business challenge your company was facing that incited the need for this platform?

I have a major manufacturer here in Michigan that I met with to discuss some of their logistic planning needs. They're an SAP shop, and they were drawing their data out from their data warehouse and into Access databases, where they were massaging that data and performing their analytics against that data in Excel.

SOLUTION

How does your company implement this platform or software?

In the use case that I worked on with them as a prospect, they had spent more than a month in development of a process that required a senior-level person to execute procedures, and it took roughly three-to-five business days to be able to create their data for analysis. It was highly time-consuming to the person that was actually doing the work, it's vulnerable to errors, and it's not easily transferrable from one person to another. This is the question they wanted to answer with Alteryx: is there something that could make this a repeatable process and, hopefully, make it go faster?

The source of the data was SAP, and they were previously using Microsoft Access and Microsoft Excel. The data needed a lot of touching – copy from one place, create a pivot table, and move the data through the pipe manually.

Was your company considering other platforms? Why this platform?

Another platform they thought about using was SAS Enterprise Miner, but they didn't have the skill set to be able to do that. The infrastructure around the company had suggested that as a potential solution, but the group that was part of this use study saw how intuitive Alteryx was and how it could easily adapt to the problem at hand.

On an annual or monthly basis, how much does your company spend to utilize the platform?

A desktop license is $3,995. If a company were to deploy modules across departments, and have many users feeding off the data and consuming analytics that were created in Alteryx, then they might decide to go for a server version. That version is $45,000, plus there would be some additional licenses for designers, for the people who are creating the modules for others to consume from. To be able to use Alteryx for these solutions, it's very affordable to get one or two people up and running and then to build out the business case to justify the growth of Alteryx.

RESULTS

Can you share any success, metrics, or overall results of your implementation of this platform?

For the use case development, I sat down with the person who constructed the original jobs and, in one day, we went further than he had been able to go in one month. In that one day, we created something that solved the problem, ran for a runtime of seven minutes instead of three to five days, and actually produced more data and higher quality input for the analysis process. Before using Alteryx, if he ran into a record that had a reference for integrity issue, or had any type of an error, he didn't just sum the total quantities up together, he put those into a queue for research later.

He was only able to analyze approximately 80 percent of the available data. In our process together, we put together rules to be able to fix units of measure and other issues in the data, so that almost 100 percent of the data was available for transportation into the analytics.

In the 10 years that I've been using Alteryx, these are the types of stories that come out. This is not an unusual case. Generally, with Alteryx, you're taking processes that were taking hours or days to run, and you're processing them in minutes. At a major marketing service provider not using Alteryx, it would take hours to days to be able to create geographic files that explain to a customer what the three closest stores are, for example. With Alteryx, you can answer those questions in seconds to minutes, depending on how many people you're using.

A marketing service provider had between 100 million and 200 million consumers in their database, and they are able to not only run faster but, now, instead of limiting the stores to three or five, the database could handle up to a dozen store choices. You could take route differences in time, to ask about the drive time distance, instead of geographic distance.

The manufacturer that I spoke to in the first use case gave me another problem, which was being able to audit records that are going through foreign trade, and their current goal is to be able to audit 15 percent of the data. We demonstrated in another use case how that could be 100 percent of the data being virtually audited as opposed to needing to have eyes on each of the audits.

This hasn't been implemented yet, but it can be an automated read of the records that are going to customs. This would enable them to verify those records, and be able to check prices on them for accuracy. If something was being shipped at the wrong rate, it would find errors and reconcile them, and correct the problem before customs and duties were paid.

FEATURES

Were there any software features or tools that really impressed you?  

I think the first one is data blending. Some tools have the ability to do things like match on a hard key. If the shipping IDs between two records are identical, they can match these easily. Alteryx provides the ability to be able to match with multiple keys, it has the ability to have fuzzy matches. If you have John Smith at 123 Main Street, and John Smythe at 231 Main Road or Main Street, you wouldn't necessarily be able to use other tools to be able to match those records. Alteryx does have that capability.

Alteryx can access data from a variety of sources. It can work with flat-files like CSV or Excel as well as Databases or even cloud data. Users can also take in data through APIs [application programming interfaces] and off the web. Alteryx can provide flexibility in the data blending, and it can be used not only by IT [information technology] professionals, but by business analysts as well. Alteryx has the same capabilities of some very advanced IT tools.

The second feature is the advanced analytics tool. Alteryx has the native ability to consume analytics that are built in the language of R. R is the language that statistics are being taught with. You can hire interns and first-time job seekers that have a background in R, and they can be put to use in a business practical problem-solving mode with Alteryx.

Also, within Alteryx, you don't need to know how to code in R to use many of these functions like linear regression models, logistic regression models, and market basket analysis. Those have all been packaged into modules within Alteryx that you can call and be able to run those analyses. These tools within the R Toolbox are all there from beginning blending through advanced analytics. It's all there, one single tool to solve all the problems.

Looking back, are there any areas of the software upon which you feel could be added or improved?

They've just recently started connecting into in-database processing. In-database processing allows Alteryx users to be able to run queries, whether it's in Hadoop, Teradata, or Oracle, you can run those processes in the database. This is something that Alteryx has just begun doing. That's where I'd like to see more support built out. You can connect with Spark, but it's not quite fully matured. The tools can be made simpler. I'm hoping that does happen.

They've just recently started connecting into in-database processing. In-database processing allows Alteryx users to be able to run queries, whether it's in Hadoop, Teradata, or Oracle, you can run those processes in the database. This is something that Alteryx has just begun doing. That's where I'd like to see more support built out. You can connect with Spark, but it's not quite fully matured. The tools can be made simpler. I'm hoping that does happen.

Have you had to interact with the platform's support team or reference their support resources?

I've used Alteryx support many times myself. When I come into an area that I haven't used recently or that I struggle with, I either go to the Alteryx community website, ask questions there, and the users and the user community are very quick to answer, or I call up the toll-free number to get Alteryx support and get a ticket opened. Generally, I'm helped in minutes to hours. They're very responsive. If I have a development question, I have a little bit more patience. If I have a production issue, I let them know it's a production emergency, and they're giving me status updates or getting me fixes very quickly.

OVERVIEW

We have a few quick questions and, for each question, we ask you to rate the software on a scale of one to five, with five being the best. What would you give the software for functionality of the features available?

Five.

What would you give the software for ease of use or ease of implementation into your business?

Four and a half.

For support, as in responsiveness of the team or helpfulness of the resources available?

Five.

For overall satisfaction with the platform?

Five.

How likely are you to recommend the software to a colleague or similar business?

Five. Very likely.

BACKGROUND

Can you please provide a brief description of your company and the role that you play there?

I am the president of CHAMPS Analytics. We are a division of a larger entity called CHAMPS Software. CHAMPS Analytics has been around since 2006. Our company focuses on providing analytics solutions to our customers. We focus in a couple of different verticals, including nuclear power, call centers, and business process outsourcers. Then, we've got some experience with health care analytics as well as law enforcement, and utilities. We've been partners with Qlik the entire time, and we have partnered with Alteryx since earlier this year.

OPPORTUNITY/CHALLENGE

What was the business challenge your company was facing that incited the need for this platform?

One of our customers is a consumer products company, and they are using Alteryx to help them with their demand forecasting. In essence, they have several years' worth of data. A lot of their products are sold during the holiday season, so they want to do a better job of predicting how many units of different types they should make based on historical information. For example, suppose they know in the Northeast they sell a lot more of product X than they do product Y in the Southeast and, every now and then, they run into situations where they don't have enough products on the shelves. They want to avoid that as much as possible.

SOLUTION

How does your company implement this platform or software?

They're also using Alteryx to ask the question, "Based on weather, if we know we're going to have bad weather, how does that impact sales?" They're trying to blend weather-related data, and they're also blending their sales related data using Alteryx to try to help predict what the best number of units they should be producing in their manufacturing operations in China, so they can sell it through Wal-Mart and all the big-box retailers.

Was your company considering other platforms? Why this platform?

They were looking at IBM Watson, but they just don't want to spend millions of dollars. They wanted to go ahead and get their data blended first, which Watson doesn't necessarily do. By the way, they're already a customer of Qlik. The reason they like Alteryx is because they really like Qlik, and they know that what Qlik does, Alteryx doesn't do very well, and then what Alteryx does, Click doesn't do at all. Qlik can do data blending, because it has a full-blown ETL [extract, transform, load] tool inside of it, but Alteryx makes it a lot easier for business analysts – not necessarily IT [information technology] people – to connect directly to their data sources to prep it, and then at the same time do all the advanced analytics as well.

Any of our large customers that have Qlik through the visualization, we think it's very symbiotic for them to also have a relationship with Alteryx because Qlik will do all the wonderful visualizations and all the user experience, along with the ability to interact and work with your data. You have to prep the data first, and Alteryx makes prepping the data a lot easier. Once you've got all the data prepped, then you can leverage and analyze it in Qlik.

If you want to do advanced analytics with predictive and geospatial analytics, Qlik can't do any of that really well. That's where you can go to Alteryx, take all this data, and plot it on a map. You can go ahead and run it through R models so that we can go ahead and predict how much more we need to make, or predict which customers are going to turn, predict how much we should be generating based on previous sales, and so on.

On an annual or monthly basis, how much does your company spend to utilize the platform?

For this particular case, it's about $4,000 per user. It's $20,000 a year for three years.

RESULTS

Can you share any success, metrics, or overall results of your implementation of this platform?

They're relatively new, and we're currently helping them right now to get their familiarization up with regards to the predictive analytics. That shouldn't take very long because we're using models that are already delivered out of box. I think the ROI [return on investment] is going to be very large because we're talking about a company that's a multibillion dollar company. It's a publicly traded company that wants to make sure that they don't run into situations where millions of dollars of product are not being sold.

The business problem they're running into is that they are manufacturing too much of some product and too little of other products. Now, they have repercussions of having more product than they need, which is going to take up warehouse and shelf space, and then not enough product that needs to be sold and pushed through their distribution. We're talking about millions of dollars worth of decisions that they're going to be making based on a less than $100,000 spend. The goal is to have ROI that can be measured in a matter of seconds versus days or weeks or months.

FEATURES

Were there any software features or tools that really impressed you?  

I think in this particular case with our customer, the main thing that is very appealing to them is that they don't actually have to go out and hire data scientists, which are very expensive types of resources. Our client can leverage the models, the R-predictive analytics model that Alteryx already provides, so they can go ahead and analyze all of their data. They know what their sales were in 2000, and they know what their sales were in 2012. They want to go back, run this data through the model, and predict what the sales are going to be in 2012.

They can compare that against the actual results. They have the ability to tweak the model so that they can get very similar types of results. Then, they can try that again with 2012 and 2013, and then 2013 with 2014. Then, you've got 2014 with 2015 and they want to do that going out into the future. They want to be able to tweak that model without having to be a hardcore R developer or data scientist. That's currently what we're working on with them right now, leveraging the out-of-the-box models.

Looking back, are there any areas of the software upon which you feel could be added or improved?

I don't really have anything in that regard because I don't want them to do anything different. We like what they're doing. We like what Qlik can do, and we like what Alteryx can do. I think what Alteryx as a company is realizing is that, if they start building visualization capabilities into their tool, then they're probably going to be out of the good graces of the two biggest visualization vendors that they've already partnered with.

I think they understand that, and they're carving out the data blending niche because Tableau can't do it at all, and Qlik can, but you have to script it in, like a developer function. They want to make their product easier to use for a business user, which they're already doing quite successfully, but then they also want to do all the advanced analytics that Tableau and Click can't do at all.

They know where their place is in the marketplace, and they just want to dominate those areas: data blending and advanced analytics. Some people will tell you that they want Alteryx to get into the visualization side of the house because they only have a relationship with Alteryx. They don't have a relationship with Qlik or Tableau. Because we've got a relationship with Qlik, we like the lines to be delineated because it's easier for us to tell our story. Use Qlik for your visualizations, but use Alteryx to help you prep the data and do all the advanced stuff.

Have you had to interact with the platform's support team or reference their support resources?

We haven't really done a whole lot with their support team. I can tell you that we have interacted with their partner sales organization, and it's phenomenal. From a support standpoint, we haven't had a whole lot of reason or need to interact with their folks.

OVERVIEW

We have a few quick questions and, for each question, we ask you to rate the software on a scale of one to five, with five being the best. What would you give the software for functionality of the features available?

Five.

What would you give the software for ease of use or ease of implementation into your business?

Five.

For overall satisfaction with the platform?

Five.

How likely are you to recommend the software to a colleague or similar business?

Five – very likely.