Archive for the 'Learning from Zayo Bandwidth's Busts' Category

Below is a letter I sent over the weekend to our Service Activation Managers:

With the help of several folks, we made major progress this past week in designing and implementing a process for tracking our install pipeline and actual. I believe this will prove to be extremely valuable to our entire organization in four ways.

First, it will greatly simplify the process in which we maintain a view of our pipeline and installs. This will save our entire organization an incredible amount of time and will reduce stress level and anxiety. Beyond just those directly involved in service activation, it will help those involved in financial forecasting, capital planning, and customer relationship management.

Second, the installation pipeline data will be more accurate and more accessible. Our ability to forecast our financials will improve dramatically, and thereby help us to navigate through the difficult macro-economic environment.

Third, it will provide our sales team with greater visibility to the status of their orders. They will be able to pull up salesforce.com at any time and see what has been installed and, for those orders in the pipeline, when they will be installed.

Forth, those of you who are doing a great job of managing your orders will benefit from recognition. We will quickly learn who has command of their portion of the pipeline and is able to use salesforce.com to communicate this to those who need to know the status of their orders. We will also learn which individuals are struggling—either because of training or workload—and get them the help they need.

I know you all are working extremely hard and are under a lot of stress. I need you to trust me that this tool and process will relieve you of stress and free up time for you to do the real work of managing order activation. However, to get there, you will need to learn a few new skills and, more importantly, develop new habits in your daily work routine. I hope you accomplish this within a few days as opposed to it taking a few weeks. With this in mind, I will send you a follow-up email that will help you get up to speed quickly and make this tool everything Zayo needs it to be. I will also personally be following up with you to see how you are doing with your patch of the pipeline.

So Now What?

  Leave a response (0 so far)
  Subscribe via RSS
  Subscribe via by Email



‘Twas a hectic and stressful week at Zayo Bandwidth. As I wrote this past week, ZB has been having a few challenges in perfecting the pipeline processes. In my past, I have been known to get a bit impatient. It used to be that my impatience would lead me to get–oh, how would you say—“helpful”? Yeah, helpful is a good word for it. So, in memory of my younger days, I was impatient and “helpful” this past week. It was time to define a more eloquent way to track our installation pipeline.

Salesforce.com was the chosen tool. Our orders are already captured in Salesforce.com during our sales funnel process. So we added a couple of fields—one called Pipeline and one called Installed. We then allowed for the name of the service delivery owner to be assigned to each order and, finally, provided an “install date” field.

Then we wrote a couple of simple reports. The first report was named “Total Pipeline”. The second was named “Gross Installs per Month: MTD Completions + Committed Pipeline”. These reports will tell us 90% of what we need to know on this topic. Provided, that is, the information is accurate.

The beauty of Salesforce.com, if implemented properly, is how visible the information can be. More so, by clearly associating the data with the name of the person responsible for its accuracy, the likelihood of having extremely accurate data increases dramatically. Beginning this past Friday, I will look at these two reports every day. Total Pipeline will tell me each order that each service delivery owner has in their pipeline—and it will tell me what month the service delivery owner expects the order to be installed. As I type this, $879,566 is in our pipeline, of which $226,558 is expected to be installed in October. I can see which orders belong to each service delivery manager, as well as each owner’s total pipeline and how it spreads by month.

When an order is installed, it moves out of the Pipeline and is designated as Installed. When I look at “Gross Installs per Month: MTD Completions + Committed Pipeline”, the first thing I see is what orders are in the Pipeline for current month install. They get captured in this report because the service delivery manager has committed that they will be installed this month—and they show this commitment simply by reflecting the install date in this month.

The second column shows those orders that made it into the Installed stage during that month. The Committed Pipeline and Installed are added together—and this tells us how much we expect to install during the month.

The result of this is hard to convey in words. But the impact is enormous. I will pick this up in subsequent posts.

So Now What?

  Leave a response (0 so far)
  Subscribe via RSS
  Subscribe via by Email



I had a whole bunch of posts on the topic of churn.  Did you read my long series on disconnects?  The difficult macro-economic environment makes a refresher on this topic an unpleasant necessity.

Given all the calories I expended on this topic, you could imagine my chagrin when we discovered Zayo Bandwidth’s other bust was a time lag in identifying an uptick in churn.  How could this be?

A well functioning churn process has its headlights pointed several months forward.  Churn, more than anything else, can help (if low) or hurt (if high) the near term financials.  This is particularly true in the Zayo Bandwidth business, due to the lumpiness of disconnects.

Many telecom companies have a lax churn process.  Perhaps when things are going well, this matters not a lot.   However, when events go south, a lax process masks this early indicator of what is happening.   Critical months are lost.  Actions that could be taken are not, because the problem is buried in the organization.

Does the company have a service issue?  Does the account management process need to be strengthened?  If related to upselling service, are the overall upselling dynamics well understood or is money being left on the table?  Are competitors stealing business?  Is the economy weakening?  Is it simply the ebbs and flows of lumpy disconnect orders?

The point is this.  If the process is tight, these questions will be asked months prior to the churn actually taking place.  If there is corrective action that can be taken, this is done at the early signs of the problem.  If there is not corrective action, at least the financial forecasts are accurate.

In early August, ZB had sufficient insights to detect that churn between September and December would be higher than in the prior months.  Instead of reflecting this intelligence in the forecasts, it let the forecasts continue to reflect the historical churn rates.  As a result, the senior team had a false sense of security around its 4Q08 revenue.  Moreover, the team did not have the opportunity to diagnose the problem and determine what, if any, corrective action was necessary.

The phase “we learn from mistakes”, though overused, is critically important.  What is Zayo Bandwidth doing to learn from these mistakes?  Stay tuned.

So Now What?

  Leave a response (0 so far)
  Subscribe via RSS
  Subscribe via by Email



Yesterday I mentioned two forecasting busts at Zayo Bandwidth.  One pertained to the Installation Pipeline.

Let’s start with a definition.  Pipeline is a simple concept.  New orders enter one end of the pipeline at the time they are sold.  They exit the pipeline when they are installed.  They move through the pipeline as the various work steps between sale and installation are completed.  So the definition of the installation pipeline is all installation orders that have been sold but not yet completed.

(Note: another important pipeline is the disconnect pipeline.  As you could guess, the disconnect pipeline is all disconnect orders that have been received but not yet completed.  More on this in subsequent posts.)

In Zayo’s August financial report, the ZB team reported its pipeline as ~$890K of MRR.  In reality, it was only about $660K.  That’s a difference of $230K.  The difference should be zilch.  That is, a well functioning process should never allow for two different views of how much has been sold but not yet installed.

Why is this such a big deal?   $230K of MRR is $2.6M/year of revenue; nearly $2.0M/year of EBITDA and about $2M of success based capital.  Our near term forecasts will be off by this magnitude of numbers. Said differently, a company that has this poor a grasp of its pipeline will have an equally inadequate grasp of its forward-looking revenue, EBITDA, and FCF.  Besides the normal hidden costs of unpredictability, such a company is exposed to a looming disaster.   This is not theoretical–I’ve seen this bring down numerous telecom companies.

The good news for ZB is that it already had processes in place that should have enabled it to be on top of its pipeline.  Therefore, it was fairly straight-forward to ensure the day-to-day processes don’t get detached from the monthly forecast.  However, ZB should take this as a wake up call–how do we ensure something like this never happens again?  We are taking steps now to increase the visibility, clarify accountability, and tighten the processes.  I will cover more about my thoughts on pipeline in subsequent posts.

So Now What?

  Leave a response (0 so far)
  Subscribe via RSS
  Subscribe via by Email



Recent Comments

Categories