Skip to content

Instantly share code, notes, and snippets.

@jamtur01
Last active October 29, 2018 01:24
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save jamtur01/471d66da69764502859dd9cb6aed748d to your computer and use it in GitHub Desktop.
Save jamtur01/471d66da69764502859dd9cb6aed748d to your computer and use it in GitHub Desktop.

A sense of urgency

As a newcomer to the company, and indeed a returnee to larger companies after ten years in startups, I've been finding my feet and getting a sense of the culture of the organization. I'm really impressed by the principles of the organization. I'm especially taken with One Microsoft and the Growth Mindset constructs. Using these it has been easy to orient myself towards mission.

What was less appealing, and slightly disturbing, is a lack of a sense of urgency and ownership in the org. In a startup it is all about execution, you're a lean, capital constrained organization with limited human resources. You have to get the right stuff done in the finite window available to you. The tempo is fast, your goals are challenging, sometimes even seemingly unrealistic. But in the best organizations, individuals and teams move with a clear sense of mission and with urgency. They feel ownership and are deeply invested in their customer's success.

At Microsoft I'm struggling to see that sense of urgency and ownership. The organization's default mode is towards long email chains and meetings rather than execution. This is not to say we shouldn't be measured and thoughtful in making decisions but rather there is a spectrum of potential approaches and ours is skewed towards discussion rather than execution.

One of the prime examples of this I've seen in my short tenure are Azure support issues. Reading through tickets I don't see customer obsession. I don't see a sense of urgency and ownership, even when it is obvious the customer is hurting. Issues drag on, responses are slow or even dropped. Breaking out of this cycle usually requires some form of escalation, if the customer is lucky enough to know someone internally who can push for a resolution. I worry about how many issues, where the customer lacks this internal sponsorship, simply aren't treated with sufficient urgency and what impact this is having on our churn.

This is transparent to customers too. They know when they are dealing with people who don't feel their pain and aren't invested in solving the problem and making things right. I've read angry and frustrated emails from Azure customers in the last month where it's clear they don't feel Microsoft has empathy for their issues and the impact of those issues on their business. We need to skew more to execution. We need to make customers feel that we are deeply invested in resolving their problem, have empathy for them and a sense of ownership in fixing an issue and making sure they don't happen again. With support, the best vendors are those that feel like a true partner, an extension of your own team.

Some immediate ideas for better outcomes spring to mind for Azure support based on my prior experience. I list these without knowing anything about the current processes but seeing what look like gaps in the process in tickets escalated to me.

  1. Re-visit the severity levels and incident process. A lot of organizations have an internal and a customer severity. It might be a minor incident for Azure as a whole but a Sev 1 for the customer. The response can then be balanced by factoring in both severities.

  2. Retrain teams on severity analysis and incident response protocol to ensure everyone knows how to treat specific tickets and the appropriate response behavior and timing.

  3. Regular escalation/review/reporting of tickets by severity that are open for longer than specific periods. Increasing level of escalation if they stay open longer. This should be tied to an SLO/SLA.

  4. Daily/shift standups and handovers. Teams prepare a brief that details the "state of the union" for their shift, identify problematic tickets and ongoing issues, SLA/SLO trends and items for follow-up.

  5. I don't know if we have an SLA/SLO model for Azure and/or specific services but if we don't this should be considered and enforced. SLA/SLOs should cover availability, response time to customers and mandate regular responses to customers for open tickets.

  6. 1-5 are tactical, it may be that these processes exist already and just need refining, re-education, or reinforcement. Strategically I think the support functions I've seen need more customer-centric support/customer service training. The teams need to default to customer obsessed behavior. When they do that a lot of these types of problems will go away.

Microsoft Azure is a Tier 2 player in the Cloud market. We were late to the market and the organization, perhaps for the first time?, isn't the leader, it's chasing the leader. We need to do better and do better fast. This challenge is one of the key reasons I joined the company. It's going to be a fascinating and challenging struggle to compete and beat Amazon and hold off GCP. But I don't think it's a challenge we can succeed at without having a greater sense of urgency and ownership. I’m also not sure how we should approach this organization-wide, but wanted to put it in your ear in case you have ideas of how to influence this in a broader manner.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment