Ask HN: Any good resources out there on putting together consulting packages? | Hacker News Reader

Ask HN: Any good resources out there on putting together consulting packages?

15 pointskfk1 comment

I am considering doing some consulting on the side around data management, analytics and business process optimization/automation. I have few contacts I could leverage but I am not sure what to sell exactly. I am concerned with selling time as it puts me into the delivering business requirements area and I would rather work directly on the business problem without some intermediary "translating" it into a list of deliveries. I tend to work better when I can deal directly with management and end users and build a solution from the ground up. One type of consulting package that worked for me was "Current State Assessment" in which I did few interviews, reviewed few systems and then did a write up of the risk areas and the possible next steps to improve. This works as a relationship starter, but I am lost on what comes after. As a further note, I am dealing mostly with big companies using consulting services from deloitte, accenture, etc., most of the writing I have found on consulting is for smaller clients with a completely different purchasing process. Basically in startups and small companies you convince 1 guy and you are mostly done, in big ones your offerings need to make sense to multiple teams and departments and require funding approvals.

I wrote a tiny bit about this in a twitter thread[0].

Context:

We do consulting exclusively with very large enterprise clients for data projects. We do however deliver turn-key custom solutions: from helping them figure out the problem when they haven't necessarily formulated it, sometimes bringing them to the state of the art in their field, write custom connectors for their data sources, build machine learning models, then create applications to enable their domain experts to use these models, and sometimes train models on new data.

We've worked in different sectors and industries: energy, employment, telecommunications, banking, retail, rail transportation, etc. Different types of problems: predictive maintenance, sound event detection, churn reduction, etc.

We meet the client where they are on the "data maturity" spectrum. We've helped organizations that were in the "what's this AI thing?" camp, and organizations that were in the "We have an internal machine learning team but we bid on a project and you've worked on a similar project. We don't have expertise over this, so we need you to help us when we win this bid", basically augmenting their internal teams.

>Basically in startups and small companies you convince 1 guy and you are mostly done, in big ones your offerings need to make sense to multiple teams and departments and require funding approvals.

If you go high enough, you mostly deal with "one guy and you're mostly done" because the purchasing/decision power they wield is far greater than the quote for your project. With some organizations, your mid six figure quote is a rounding error green-lit over a phone call.

However, you will have to push to include the teams and departments and onboard them to the project to ensure it succeeds. Many years ago, we practically only dealt with executives. Now we push to include their domain experts and stakeholders and bring them to the table.

This is good for trust. Repeat business is good, because you don't have to go through many things over as you know many of the people in the client organization when you work on the next project.

Also, it helps to deliver consistently.

- [0]: https://twitter.com/jugurthahadjar/status/131066829330549965...

Reply on news.ycombinator.com