How We Help
We’ve learned of few lessons that can be applied across a broad range of clients. These lessons fall into three categories of projects. We call these “Build”, “Support”, and “Advise”.
This project started with a problem I was having with my fantasy draft, and evolved into a project for a client in a new industry. I joined a fantasy football league with friends. At the time, I knew almost nothing about fantasy football strategy. A friend recommended I use a fantasy football advice site. This site had amazing information in a “Draft Plan”, but it was difficult for me to use this information in real time to select players in a draft when I had 2 minutes to make a pick. During my first draft, I had dozens of pages of paper I was frantically leafing through, and I made many mistakes.
My solution the next year was to create a tool that would recommend the best player at any time by using expert information from the advice site and the players already drafted to that point. It would display your team and recommendations for the next player to pick based on your current team makeup and projections for remaining players.
This tool completely revolutionized my drafting strategy, so I sent it to the owner of the fantasy football advice site. He loved it and asked to put it on his site. Ultimately, we built several tools for them, for drafting, for auction leagues, and even for daily fantasy sports.
When my contact founded a new company called Fantasy Points, their level of industry expertise was unmatched. However, they needed additional tools for end users. They approached us with this question – “can you make the world’s best drafting tool?” A tall order!
We realized a best-in-class tool would need to incorporate Fantasy Points expertise (projections as well as qualitative knowledge), but also predict what other team owners would do based on their team makeup and likely drafting strategy. For example, if a player is the “best” player in round 4 but is unlikely to be picked by another team owner before round 6, then the best strategy is to hold off on picking that player. This is an example of how to use predictive analytics in a practical application. In addition, we decided this tool should be web-based, where thousands of people could use it at once without any other software required.
Taking the tool to the next level
The Fantasy Points team loved the tool. It became the best fantasy football drafting tool on the market. Once they knew our capabilities they had another request – “we want a mode that makes it seem like our expert is telling you who to draft, not a computer. Like it would know in all situations who we would pick, even if it went against what the ‘numbers’ would say.”
Most “build” consulting companies would have one of two responses:
- “sure – tell me exactly how to build this”, or
- “no, we can’t do that”
However, we look at things differently. Instead, we interviewed their experts and developed additional “rules” for the tool to have it replicate their human thinking as much as possible. We had to figure out a way to turn qualitative insights into quantitative rules a tool could handle. We made a new “expert hints” mode that built these rules in.
The end result was a huge success. We got comments like “it was almost eerie how it picked exactly who I thought it should every round”, or “it feels like there’s an expert in the room”
Final thoughts
While your “build” project will not likely be a fantasy football advice tool if you’re looking for a tool, a process, or a strategy to be built, you don’t have to know the exact design or even what the root issue is; you only need to know the outcome you’d like to achieve. While working with us you’ll find we think strategically – we’re not just coders/analysts building a generic model, process, or strategy – we have expertise in figuring out exactly what you need to achieve and figuring out a way to do it.
In essence, we design and develop models, processes, and strategies to solve actual business problems. Unlike the hundreds of other companies that will build a tool based on your exact specification, we strive to solve your business problem even if the tool or problem isn’t well defined yet.
This is one of my favorite clients. (Well, the consultant in me would say they’re all my favorite clients!) We have a great working relationship. Earlier that day I sent a rough draft of a tool I was building, for feedback. The tool wasn’t “ready for launch” but this is how we work – we don’t hold back our work until a grand unveiling – we work collaboratively with our clients throughout the process, as we’ve learned having an honest back-and-forth leads to better results for everyone.
I say “sure, what’s going on?” while holding my breath (“We’ve got an urgent problem” are words you never want to hear as a consultant!)
“A deadline just changed – it turns out we need to use the tool to do some analysis by tomorrow morning! Can we use the tool you just sent?”
Tomorrow morning?! The tool is still in “rough draft mode” – it’s not ready yet! I hadn’t gotten feedback from the client, and we hadn’t sanity checked the tool. But still, they were in a bind, and I wanted to help. And coincidentally I was 5 minutes away from the client… without my computer… wearing faded jeans and a cruddy old t-shirt.
“Hey, I’m literally 5 minutes away – can I come over and we’ll do updates together in real time? And do you have a spare computer?” I ask.
“Really? Sure, that would be great!”
So I drive over to my client. I check in at the front desk, laughing at how I’m dressed. I’m pretty sure any “how to be a consultant” manual would tell you never to show up at a Fortune 100 company dressed in faded jeans and a cruddy t-shirt!
We get a computer and a conference room – I throw the model up on the screen. My clients and I work through the model, making changes and updating it in real time. It’s an amazing collaborative effort – my clients with tons of feedback, me updating it as they watch. It’s a frantic 90 minutes, but we get it done. In the end, the tool became “usable” for their needs for the next morning. Everything worked out great.
Why do I tell this story? Well, I do like the funny visual of “Partner of consulting firm shows up in jeans and t-shirt to fix a problem” but there’s something more – I think this really does show the ethos of Bellwether Analytics and our relationship with clients:
- When we have a collaborative working process, we can accomplish more
- We can be hugely flexible – while I can’t guarantee showing up at a company’s doorstep in 5 minutes every time there’s an emergency, we do go out of our way to help however we can
- We have a great working relationship with our clients. I’m not going to worry about stuff like how I’m dressed for an in-person meeting, neither are they – we’re just going to get the work done!
We enjoy working with startups and early-stage companies because of the novel technology they can launch into their industry. A recent med-tech company’s founder came to us to help prepare for a capital raise. What started as a project to prepare the company to raise money, turned into an exercise in training, helping, advising, and creating. The founder and his partner had a fantastic new product for the surgical space. They were technically excellent but had never started a company in the med-tech space. The first exercise was to train them up on the industry such as what is required to launch a new product, how much capital it typically takes to make this happen, how long it takes, and the kind of expertise they will need to execute their launch plan.
We were able to connect them with regulatory, quality assurance, and software experts to accelerate their regulatory process. The founder was clearly unsure about next steps and had been provided advice from several other people in his network. Most of the advice was fine, but not relevant for a medical device launch. We spent several hours educating them on what mattered and what did not. Most importantly, we changed their forecast from tens of millions to hundreds of millions based on our assessment of the market. Challenging projects such as this one are incredibly engaging to work on. The founder told us that we had become his most trusted advisor.
