Showing posts with label CAA. Show all posts
Showing posts with label CAA. Show all posts

Tuesday, December 16, 2014

Personal Insight into Project Management

[Disclaimer - The article below was written for a Project Management Course. The thoughts are of my personal and does not represent the company's (CAASCO) view points. Many details have been omitted/dramatized]


Project Description:

“You are off the hook for now Ali” – stated Victor, the Finance Director of CAA Insurance. He along with 3 other Managers from different business units (Adjuster, Claims and Reporting) were testing the new Finance Reports while I (Architect/Developer) was on the conference bridge hearing them validate those reports. I have to admit my heart beat was racing. As much I was relieved that I don’t have to ‘face’ them while they are doing their UAT (User Acceptance Test) in Montreal office but at the same time oblivion of what is going on there in the conference. The project deployed was in 2 days and any major defect discovered would put the project schedule in jeopardy.

The project was called ‘Finance Gateway Reporting’ project. The Reports were mix of daily pulls and monthly pulls. The data source of the Reports were ‘cubes’ (aka Multi-dimensional Data entities) from our Enterprise Data Warehouse (EDW) which were to be hosted in SharePoint. The business stakeholders were ‘straight shooters’ – they want accurate numbers in the Report. Difference of 0$ between the Reports was the criteria as Reports would be useless if not otherwise. All the Reports have to reconcile with each other and displays detailed-transaction level details that can be drilled-up to summary level.

The Finance Gateway Report Project kick-off was 6 weeks before. The timelines delegated to the Business were aggressive. In the planning phase, all the stakeholders along with a Project Coordinator (acting as a Project Manager), a Business Analyst, a Quality Assurance Analyst and me – the solo Developer/Architect were present. I estimated timelines based on ‘sunshine’ scenarios and full time dedicated effort for the project. The timelines had to be given during the Planning phase meeting. To give correct work effort estimate while you were on the hot seat was as good as using a black box to generate random number for the work tasks.

However, the project was deemed successful with on time delivery and execution. Measurable acceptance criteria (eg. Bug Tracker for bugs) were set in place. Agile methodology was adopted and we had daily ‘scrum’ meetings hosted by the Project Coordinator every morning. The objective of these Scrum meeting was to give a succinct account of your status: tasks accomplished and task intend to do next and apprise about any obstructions or challenges encountered. The meetings were kept with 10 min run time. This was a platform where business were informed in every step of the project. Any ambiguous (or redundant) business case was raised for clarification during the scrum meetings. Project Coordinator would update the ‘sticky notes’ on the board along with email send out to everyone ensuring any changes or updates we have captured are not missed.

The challenge I experienced was in the planning phase when the Finance Director said “Why can’t the IT get the damn Reports right.” The Reports were simple tabular (with expand/collapse feature along with Export to Excel/PDF), so it was perceived as less development effort. At the same time, not plucking the correct data from the data warehouse and displaying it in the Report was also perceived as a callous attitude with no due diligence of the IT in front of business.  Moreover, there is this popular perception around EDW in Business Intelligence world, “Data goes in to EDW to die” hurting our cause of delivering the Reports. It was my first client-facing project at CAA Insurance and knowing the history of IT vs Business; I saw this as an opportunity as not only to redeem IT but also to establish myself as a force to reckon (sounds cheesy).


The goal of winning business’ heart or better put - to salvage the reputation of IT in front of Business was the driving force that led the project to be successful. The daily scrums were seen as a transparency barometer gauging not only the progress of the project but also gauging the authenticity and intention of everyone involved. I used that opportunity to foster stronger relationship with the Business. Raising the ‘umbrella high’ sets the tone of the project. I would promptly set-up a meeting with a Business person to explain or proof of having this in the Report instead of asking next day at the Scrum meeting. Being proactive and showing ‘human’ side by showing empathy and care to their requirement goes a long way in not only making the project successful but also fostering relationship with Business.

All risks were put on the table and were managed by having placed them in risk registers and having an action items against that risk as part of risk mitigation plan. The inherent risk that was not explicitly enunciated was the business knowledge. Able to understand the business and able to converse in Insurance business jargon was a steep learning curve.

Project Analysis
In terms of following best practices with building trust and transparency, project scope was clearly defined. Work breakdown structure (WBS) and Solution Design Document (SDD) were documented with sign offs from key stakeholders. One of the key scope for the project was that data inception is assumed to be from the Enterprise Data Warehouse (EDW) and not from the heterogeneous data sources that feed the EDW. This was the highlighted clause that was made sure it is well understood across the table and no surprises crop up during the project execution.

If one of the sources feed an erroneous data into EDW, the projects that are in progress for that faulty system must be notified and are responsible for the fix. The onus of that is conveniently transferred to other project. The projects at CAA run in isolation. Project boundaries are high enough for other projects to see.  Raising the issue with the leadership was also futile. Different projects deal with different systems. Different teams work on different systems. In order to tell the Project Manager for ‘SystemA’ to tweak their design was easy said than done. Each Project Manager wants his/her project to succeed. Each project runs in isolation with having no incentive for cross-project learning. This ties to the structure (or lackof) for cross-project learning. Each Project’s objective is to execute on time with absolutely no bargain. In the scope of the project it was conveniently addressed that other system’s issues would not impact our delivery. I had raised the concern of not sharing knowledge amongst projects does not bring the most value to the organization. My concerns were shrugged off by “that’s how Agile works – I know it’s a drawback” remarks from the Project Coordinator.

Adaptive Project Organization is what CAA projects need to apply. In Adaptive Project Organization, all projects are dealt as part of a same body. If any part of the pains, whole body feels the heat. Such integrated project organization allows better resource sharing and also better equipped with facing the stakeholders. Although the overall approach is ‘Chaotic’ but the project was dealt with ‘Adaptive’ mindset. Resources were highly committed to the projects. Personally speaking I had worked lot of long hours every day during the project so that the project could see light at the end of the tunnel on time. We had a ‘Demo’ every Friday to the Business emphasizing engaged project culture and being transparent and open to Business.

The Project Coordinator ensured everyone was aligned with Project and the progress was pacing well towards the finishing line. The Project Coordinator also raised the umbrella by negotiating delivering the ‘functional aspect’ of the Report was higher importance than the ‘cosmetic aspects’ of the Project. Project Coordinator took the Business in confidence that any ‘cosmetic bug’ that arises during UAT will be taken care during the warranty period as delivering Reports with data-integrity was placed in higher value. Cosmetic bugs (e.g. number formatting to 3 decimal instead of 2 decimals, font color and size, or order of columns) will not hinder or impact the delivery of the Project. This was a key win as we certainly did not want the Project to get delayed because of a font size!

Management showing trust after seeing our efforts was also heartening. After all both units (IT and Business) work for the same organization. Both have the same motivation and drive to succeed. Management understand the value IT brings to the table. Management also noticed the off-hours efforts being put in the project. This was also a reason why Management agreed to delay the cosmetic enhancements during the warranty period and focus on the core Report functionality (e.g. Logic of data retrieval and the calculations based on Claims Payments, Reversals, and other charges).

Best Practices:

The three best practices that I would adopt in my career based on my experience and the takeaways from the course are:

-         Raising the Umbrella high: Trust has to be earned and cannot be taken for granted. When Raising the Umbrella high, it sets the tone of the project and gives Management confidence that the project is in ‘good hands’.
-        Identifying risks and plan to tackle it: This is something I should strive more in my career. I have a tendency to follow “sunshine scenarios” and not even account for any potential risk or concerns. Be able to identify risks and putting on the table is half the battle won. To plan and mitigate and abate those risks is the other piece of the puzzle. The biggest risk in my project was the dependency (or intersection points) with other systems (and teams). To coordinate the dependency and escalate it should be included in the plan.
-         Communicate with Intent: This single trait makes or breaks a project. Keeping the Business Management in the loop and making sure everyone is aligned with the requirements and how those requirements are tied into strategy. If the communication is with intent and transparency, any challenge will be dealt with care from all the stakeholders. Be it Scrum meeting, email communication, weekly Demo – communicate with intend is winning over Business through work ethics.

Recommendation

Even though Project execution and delivery was near-flawless but it was much to be desired in terms of graduating from ‘Chaotic’ to ‘Adaptive’ project leadership at CAA.

To have a Project charter underlining Project strategy and Project leadership gives the projects some branding and therefore a responsibility. The onus of delivery was heavily skewed towards the Architect (me in this case) and the Project Coordinator. The responsibility should be shared and resources should be better planned out. To adopt a Project leadership where projects are integrated as single unit and treated fairly rather than me (first project) facing business clients and appointing a Project Coordinator instead of a Project Manager. Projects sharing best practices and lessons learned amongst projects strengthening Project Implementation and Strategy. This practice of leveraging knowledge from other projects can be iterated and included in the Project resource kit. To have a holistic Project implementation plan amongst projects where time estimates can be negotiated and risks be identified and put into the plan schedule. 

Friday, July 11, 2014

Towing Data Analytics?

       
Data Analytics-driven Telematics holds a bright future for CAA

        Winter 2009, when the dispatching unit of CAA South Central Ontario (CAASCO) was getting bombarded with calls from members for car assistance. It was a brutal snowstorm paralyzing half the GTA. GTA had experienced the highest casualties related to automobile in that year[1]. It was also the most financial drain for the company as property and auto claims depleted our pool of reserves. By end of 2009, CAA SCO (South Central Ontario) was in a loss of $29 m. Tethering into bankruptcy and by 2013 in a surplus of $ 1.9 m. Miracle isn’t it? What magic wand did we acquire that turned our fortunes? Ironic how we look elsewhere for solution when the solution exists right in front us the whole time: data and people.

In corporate circles, executives vouch for data being akin to gold. I have a different tale to tell: Data has the potential to be gold. Data is worth stack of hard drives collecting dust in company Data Centers unless we churn data into gold!

“Leveraging customer Data and doing some simple eye-balling was the need for the hour
” Jay Woo - President and CEO CAA SCO


 A home owner claiming for sewerage repairs for three consecutive years without Premium go up a cent1. An auto insurance policy holder heaping demerit points for impaired driving and still manages to retain the same Premium by threatening to dessert us. Under pressure from increasing competition, we recognized that our viability and sustainability depended on taking a hard look at analytics at each of our pillars – customer, operations and finance.

Jay Woo, President and CEO of CAASCO said: “Leveraging customer data and doing some simply eye balling was the need for the hour.” We had reached our inflection point. We identified our ‘Analytically Challenged’ framework and processes and re-evaluated and aligned our strategy to being customer-focus and data-driven one.
     
Business Intelligence and Data Analytics was the key component in reaching our goal and thus keeping us afloat. We did cherry-picking in terms of whom to target for our home and auto insurance. We segmented the market and focused on the piece of a pie that we were most interested. The pilot project was named ‘Risk Segmentation’ which was put in place in 2010 to increase our market share.

Enterprise-wide KPIs were laid out by business which enabled us to exploit those areas in our ‘analytical sandbox’ or ‘Center of Excellence’ for analytics. Same methodology and processes were scaled to other product lines and business units within the organization. We were able to minimize our loss ratio and by 2013 we were leading insurer with the least loss ratio. This by no means was a flash in the pan, detailed and articulate analytics enabled us to see value hidden in the data and realize its true potential via analytics.


By transforming insights into actions, we are now able to prioritize our future investments. We are now aiming to being ‘Analytical Innovators’ from ‘Analytical Practitioners’. This will enable us to be achieve competitive differentiation as well as fine tune our market segmentation. Adopting analytics made us turn challenges such as customer-retention, anti-fraud practices and misaligned marketing campaigns into opportunities. True strength of our analytics will be tested when we start gathering even more data ‘gems’ from our telematics this year (We have recently launched our UBI-tool called Telematics last year). This will shift us from being in ‘What happened” and “Why it happened” mode to “What is happening now” mode. The amount of data we will collect and to realize the data’s value is the essence of data analytics. To exploit the ever-increasing data with degree of rationale in real time is where we will be focusing our attention to. Driven by analytics, we can provide BPO (Best Product Option and BPA (Best Product Action) to the customer.

Data Analytics is by far not the domain for growth seeking companies only. Established companies can leverage data analytics in seeking efficiency, seeking new markets or even fine tuning their strategy based on analytics. Data analytics should be ingrained in a company’s DNA, driving forward its strategy and implementing its goals. Over-time, companies will witness data-driven decision making branches out across companies cementing more value and trust in analytics 
 As I write this article, one should be cognizant that the data is being accumulated at an exponential power and the ability to churn data into gold is the bottleneck. To identify ‘strategic information’ from ‘noise’ that is being flooded from structured and unstructured sources and mold them into ‘single version of truth’   is only half the battle. To ‘predict’ with reasonably amount of accuracy , mitigating risk , providing timely customer engagements are some of the opportunities that will drive our company forward – after all we are in the business of ‘making bad days good  and good days better’ .


[Disclaimer - The article above was written for a Management Course about Analytics practices in current workplace. The thoughts are of my personal and does not represent the company's (CAASCO) view points. Many details have been omitted from the blog.]


[1] CAASCO Portal