Photo credits >>photopin

I started using a data-driven model to manage my team — here are the results

There is a famous quote by W. Edwards Deming, the father of modern quality management. It goes like this:

In God we trust, everything else bring data

Over the last four months, a lot has changed at the iHub. For me, the main change was that I took on a different role that involved managing more people. Previously, I only used to head the Software & Design team only. I now head the Consultancy unit which is comprised of the Software & Design team, iHub Research, the Data Science Lab, and the iHub UX Lab. Currently, the team is made up of 15 full-term employees and 12 consultants.

When I was managing a smaller team, I knew exactly what each team member was doing. On a weekly basis, I could tell you exactly what we had spent our time on. I couldn’t give the exact hours for each team member but I had a pretty good idea of where we stood.

With a larger team, this approach wouldn’t work. I had to figure out a better way to track this.

Luck has always been on my side. Honestly, I didn’t know how I would go about this. Luckily, one of iHub’s new investors has extensive experience in this particular kind of thing. By working with him, I identified that I needed to track two key items:

  • Our sales pipeline
  • What the team spent time on

Tracking our sales pipeline proved easy once the right tool was identified. I can’t say the same for our time — every employee resisted this!

Managing the sales pipeline

To make sure that the sales pipeline is better managed, I use Insightly — a CRM tool for small businesses — to track all our opportunities. Using the tool, I capture the following key information:

  • Opportunity name
  • Value of the opportunity (the $$$)
  • Pipeline stage — new, value proposition, proposal, negotiation, close
  • Probability of winning — new (5%), value proposition (10%), proposal, negotiation (40%), close (100%)
  • Forecast project start date if won
  • Duration of the project

Every Monday, the sales team meets to go through the sales opportunities. In this meeting, we check how far each opportunity has moved down the pipeline. The idea is, on a weekly basis, to move each opportunity to the next stage in the pipeline. We also use the meeting to update each opportunity. An opportunity might have been converted, lost, suspended, or seen its value change amongst other things.
The Monday meeting also presents a good opportunity to check how healthy our pipeline is. In the meeting, we identify where we need to generate more leads as we sell different services. 
Within the time I have been tracking the sales pipeline, I have noticed some interesting trends.

  • Most of our opportunities stay the longest in the proposal stage. With this information, I have devised strategies to move them faster down the pipeline.
  • There is a clear correlation between the opportunity value and how fast the deal is closed. This has allowed the sales team to focus our efforts on opportunities that have a higher chance of turning into projects.

Tracking how the team spends their time

This, so far, has been one of the most challenging things I have implemented in my team. When I first made the announcement, there was a lot of pushback from the team. The concerns raised included:

  • Trust — most, if not all, team members thought I didn’t trust them hence the need to track their time
  • Basis for firing — team members were afraid that if they didn’t log ‘enough’ hours, it would be used as the basis for firing them

This reaction was expected especially considering that previously iHub had never tracked anyone’s time. I had a long discussion with the team to quell these fears. Even after this, some team members were still skeptical about my motivation for tracking their time.

To track time, I needed a tool that wasn’t overbearing but would still give me the information I needed. After looking at several options, I chose Tickspot.

The first thing I did was to add all existing projects complete with the budgeted hours for each. To derive the budgeted hours I used two different approaches depending on available information.

  • For projects that had detailed the hours properly in the work plan, I added this directly to Tickspot
  • For projects that didn’t have a work plan broken down into hours, I took the total budget and divided it by our theoretical billing rate of $50/hour hence giving me the total hours the team could spend on that project

For each project, I also indicated whether it was a billable project or not. Activities such as internal meetings, business development, sick leave, vacation days, and similar activities are not billable. Working, for example, on the iHub website isn’t considered as a billable project. However, team members still need to track time spent on such activities. I added these activities as nonbillable projects.

Once I added all projects , the team could now track their time. Based on our working culture, team members aren’t expected to log their hours on a daily basis. Instead, this is done on a weekly basis as Tickspot is only checked once a week. This allows individuals to retain the flexibility they had previously.

Initial insights from time tracking

After tracking time for about three weeks, I started noticing some interesting trends:

  • On a number of projects, the team spent more time than the budgeted hours. After probing this further, I realized that in most cases the project had budgeted for one person and yet two people worked on it. In some situations, the project work plan had underestimated the time certain tasks take
  • The teams’ billable hours billed were very low compared to nonbillable hours. We were not striking the 80–20 balance. Ideally, 80% of a service based unit should be billable and 20% should be non-billable
  • The teams’ utilization rate was really low. On average, the team was busy about 60% of the time. I had more employees than work :-(

This presented a good opportunity to show the team how the data collected is going to be used. The issues the data brought to the limelight were symptoms of deeper problems.

  • Proposals — our proposals weren’t very accurate. We were not estimating tasks properly. As a result, the team would spend more time than budgeted for
  • Resource planning — as the leadership team, we were not planning this well enough. We had instances where two people were working on the same thing at the same time yet another person (more skilled) could do it alone within the same timeframe. Busy<>profitable
  • Business development — we didn’t have enough projects. That wasn’t the team’s fault but a problem with our sales team

Better resource planning

With the insights from time tracking, I added a resource planning session to our weekly meeting.

After going through the sales pipeline, with the rest of the leads, I spend a significant amount of time reviewing what the team worked on the previous week. This involves going through each team member’s timesheet on Tickspot. We know, on average, people work 36 hours a week. Sometimes, some people log very few hours. In most situations, that particular team member didn’t have enough work. This means the sales team has to sell that particular service more aggressively.

Once this is out of the way, we work on activities for the week. What is each team member doing? What percentage of it is billable? Who isn’t busy? Who needs help? Instead of just waiting for the next meeting to review how busy everyone is, this exercise gives us an opportunity to get this information in advance. In the next meeting, we check against the planned activities.

Taking it a step further — tracking profit/loss weekly

Typically, profit/loss is tracked on a monthly basis. As a result, you can be doing something wrong for a full month and not know until you get the P&L statement at the end of the month.

Before I can track my profit/loss, I first needed to establish how much I spend on a weekly basis. This is something that took a while to figure out. However, once I had a clear idea of my operation cost it was fairly straight-forward. I derived it in the following way:

  • I calculated all the indirect costs incurred by my team — leases, admin expenses, support staff e.t.c
  • Next, I added all the direct costs in terms of salaries being paid out
  • I then took this figure and divided it by the total number of working hours in a day for the team. Assuming each employee works 8hrs/day and I have 15 employees my number is 120. The resulting number is my cost rate

This was only possible with the help of our finance department and so if you want to try it you will have to work closely with your finance department.

I had one more thing to do before I could get the profit/loss number on a weekly basis. Seeing as this approach doesn’t use actual cash flow, I figured out a way to calculate how much each employee makes for me per hour. To get this number, I took the cost rate and applied a mark up that would give me a good profit. This is my billing rate.

Since I have team productivity data on a weekly basis and I know how much it costs the company to have each employee (cost rate) in my team, I am able to track profit/loss on a weekly basis. By taking the total billable hours for the week and multiplying it by the billing rate, I know how much I made that week. If for example I billed 300 hours in a week, I know for that week I made $11,400 (assuming the billing rate is $38). If, for example, my cost rate is $20/hour, I have 15 employees, and we work 4o hours per week then the total cost per week is a whooping $12,000 (15* $20 *40). To break even, I need to clock at least 315 billable hours (12,000/38). To get the weekly profit/loss, I get the difference between what I made and how much it cost me to have my team per week. In this scenario, I am making a loss of $600 (11,400 less 12,000).

Companies seek to grow over time. As a result, hitting the target billable hours isn’t sufficient. A good organic growth rate is achieved once you can make a margin of 35%. Therefore, I need to do 425 billable hours every week to grow organically. Any additional hours can go into my bonus kitty. I can now use these numbers to communicate clearly to my team.

  • To break even we need to bill at least 315 hours a week. Anything less than this means we are making a loss
  • To grow organically we need to bill at least 425 hours a week
  • If we want a bonus at the end of the year we need to bill significantly more than 425 hours every week

Predicting the future

I am lucky to have, over the past four months, worked with a very experienced and talented person who has acted as a mentor and provided guidance on how to run a business better. Left to me, I would have stopped at the weekly profit/loss analysis. Why stop there if you can predict your profit/loss for the next 6–8 months?

If memory serves you right, in the sales pipeline I captured a number of things. I can predict future revenue using a formula that factors in the opportunity value, duration, probability of winning, and the probable start date. The start date is necessary so that I can spread the revenue over the project duration. I am then able to know how I will be doing in the foreseeable future. With this information, I also know whether my pipeline is healthy over a longer period of time.

In conclusion

There are many different data-driven models for managing a company. I have found this works well for my team. With this model, I have total transparency on what is going on. I know where I need to channel more of my energy to yield increased sales. At each point in time I know who needs more work in the team. Consequently, I am able to plan more effectively and that is great.

Would this work for you? It depends on how your organisation is set up. The idea is to identify what works best for you. You can borrow bits and pieces so as to customise it to suit your business. Data doesn’t lie. The more data you have, the better the decisions you can make.


Enjoyed this piece and keen to share this with the world, colleagues and friends? Hit ‘Recommend’ a.k.a the lovely little heart at the bottom left of this post.


One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.