« Liespotting | Main | A Moral Man »
July 29, 2010
Redemption
On Monday, Edelman’s US operations moved onto a new Global Financial System (GFS). Four thousand separate transactions were recorded, from time sheets to expenses. Bills are scheduled to go out early next week to clients. A pretty boring blog post is in store, you might think. Well think again.
In 1998, in my second year as CEO, Edelman was growing quickly and becoming increasingly global. We recruited a new chief financial officer from a division of a Fortune 500 company. She concluded that our home-made financial reporting system based on spread sheets and manually produced bills was antiquated. Since we had PeopleSoft as a client in Silicon Valley we engaged them to install the financial backbone that would carry us into the 21st century. I was content with the salesperson’s pitch, promising immediate access to profitability by client, outstanding receivables by client, productivity of each operating unit. I went back to my day job, counseling clients, managing people, winning new business.
What I failed to recognize was that PeopleSoft hadn’t fully evolved from a human resources package. It needed a partner software vendor to offer a complete financial solution. That partner was a small Midwestern operation called Proamics, which “specialized” in professional services firms in law and accounting. Time and expenses were collected in Proamics, but the accounts payable interface from Peoplesoft to Proamics was so bad, out-of-pocket expenses couldn’t be billed. The Edelman financial team assured me that everything was going well in the test phase, so we shut down the legacy system and moved onto the new platform.
All seemed to be going well, as we sent out our bills to retainer clients. Then the system came up with bizarre bills for hourly clients, which the account teams rightly refused to approve. Within the month, it became obvious that the Proamics and PeopleSoft interface was flawed. We were a ship with a crippled rudder, unable to bill clients correctly and incapable of processing expenses. We could go neither backward to the legacy system nor forward to a dysfunctional new system.
In short order, I concluded that I had a full-blown, company-at-risk crisis at hand. I wrote down a list of action items, from finding a new CFO to bringing in programming assistance from our audit firm to meeting with our accounting staff to tell them we would have to go to manual processing of bills for a period of time to calling our bank to inform them that we would need to tap our credit line to notifying Proamics that they were in breach of contract to calling clients personally to tell them what was happening. I put a time line together and gave myself a deadline for each action.
So in order, I found a new CFO and fired the incumbent. We gave Deloitte sixty days to fix the system by disabling Proamics and enabling manual bill producing functionality. I called our account leaders twice a week to update them. Our internal financial team worked like Trojans to get bills out in consultation with the staff and clients (who were understanding and paid upon presentation of accurate bills). After running our credit line up from zero to several million, we stabilized the ship and paid it back over the course of the next twelve months. Nothing like a near-death experience to prepare you for running a company!
Fast forward a decade, with Edelman’s revenues more than $400 million and our PeopleSoft system now so obsolete that Oracle says it will no longer provide support for the version we have. We began the process of selecting a software vendor and implementing a new system. Hearing the presentations by the sales people, I heard, “This is going to be a piece of cake. It will take only six months to install. We have a full offering now. There will be no customization. Just talk to our customers who are so satisfied.” On and on the palaver was endless. We stayed with PeopleSoft.
This time I was a lot more cautious. We would not go live on the new system without a way to retreat (keep the old system on line just in case). We would test the system in a smaller unit than the massive US division—in Edelman Canada as it turned out. We would hire PWC as outside counsel from the start, with the senior partner reporting to me each month on progress. We would not have a specific budget nor commit to a timeline to complete the installation. The cake would be baked when it was ready, not a minute before.
So here is what happened. I was told it would take six to eight months to complete the work, then two months to test in Canada. In fact it took over a year and a half to customize the system, then a year of testing in Canada where it became clear that there were system processes that didn’t meet our needs. Here is one example: if one senior person disapproved time of a junior staffer, then the whole time sheet went back to the start, requiring re-approval by each senior person, holding up billing. The software consulting firm recommended by PeopleSoft was disappointing at best, with a constantly changing cast of temper-prone experts. Half way through the project, one of them told me to scrap the whole thing. Over time, we brought more of the expertise in-house. I spoke with account people in the Canadian office, went up to Toronto to see how it was being used and why certain executives just would not enter time. At Christmas, I had to make a decision about whether to delay the implementation four more months to install “bundles” of improvements which PeopleSoft programmers had made to the 9.0 version given customer complaints but enduring the higher consulting costs necessary (I opted for the safety of installing the bundles).
Here are the lessons for all of you considering a comparable voyage across the river into the unknown. Put yourself in a position to make informed decisions, not just to approve recommendations by your CFO or your outside accountant. Go into a test phase that is live production of bills and paychecks, not just a simulation. Take the financial estimates provided and double them—this is just like renovation of your home—it will take twice as long and cost twice as much. Make sure that you have a path out of the forest if the worst happens. Get personally involved with your account people, particularly the squeaky wheels who may overstate their complaints but will offer critical kernels of truth. I want to thank all of those at Edelman who have gone on this adventure for the past three years (Vic, Derek, Barb, John, Kristin, Paul, Anna and Gene) and our consultants at PWC. Dogged persistence and commitment to excellence is the Edelman way. And for me, getting a second chance as CEO to do it right has been a blessing.
Posted by Edelman at July 29, 2010 4:15 PM |
Trackback Pings
TrackBack URL for this entry:
http://www.edelman.com/mt/mt-tb.cgi/1089
Comments
Wow, not a boring post at all. I recently realized that this will be the third time entry system I've used at Edelman (my 10-year anniversary is in December). While I am one of our new financial systems "change champs," I'm being completely honest when I say how exciting this launch week has been. I'm glad we waited to get it right, but knowing the story behind it shows how it really was the best decision.
Posted by: Bree Flammini at July 30, 2010 12:43 PM
"Everyone makes mistakes. The key is to not make the same mistake twice."
Richard, you learned your lesson well the first time, and knew just what to do to avoid a full-blown catastrophe the second 'go round.
Of course, there were things that happened that are very difficult to plan for. As you said, you eventually learn to plan for things costing twice as much as they should. You learn to listen to the sometimes 'overstated' complaints for the grains of truth scattered between.
And isn't it nice that this potential headache has finally come to rest? Glad to know the US operations are now on a new fully-functional GFS.
Great takeaway points here, Richard.
Posted by: Robert Burns, II at August 1, 2010 1:00 AM
An experience, very well written.
There are so many who go through the same experience.
A lesson for consultants who need to strive and make life easier for USERS by first understanding that USERS know their processes best.
I had recently visited your GURGAON office.
Rgds
Posted by: Anil Jain at August 14, 2010 2:40 AM