Two years ago, reps at AMF Bowling Products had good reason to shudderat the very mention of sales-force automation (SFA). It is notoriouslyhard to get right, and the company's first attempt failedmiserably.
AMF reps sell everything from shoe spray to the back-office softwarethat runs a bowling center. They need a way to track theircustomers - owners and managers of some 6,000 U.S. bowlingcenters - and theequipment they already own. Knowing which customers have ancient ballreturn machines or out-of-date automatic scoring systems helps themzero in on the best sales targets. The company's first major SFAeffort - a series of homegrown Lotus Notes databases of customerinformation - didn't take long to earn the sales reps' ire. Reps couldn'ttake their laptop into bowling centers and enter customer data withoutlooking like spies to the alley owners with whom they were trying toestablish trusting relationships. So instead, they'd scribble notes onpaper. But after schlepping through four to six bowling centers, thelast thing reps wanted to do when they got to their hotel at night wasfire up their laptop and spend up to an hour logging their sales callsinto one database and entering data on what kind of equipment eachcenter had into another. Because it usually took another 30 to 45minutes and sometimes as much as two hours to replicate their notes,most reps gave up in frustration. "You're talking about an extra twohours at the end of the day," says Chris Keller, Long Island,N.Y.-based Northeast district sales manager and a 10-year AMF veteran."The application was just too difficult. We're salesmen. We need touse our time to make money." Keller confesses he abandoned the Notessystem after about three months. Only half of the sales force updatedNotes religiously, and as time went on, usage fell, says Jay Buhl,vice president of North American sales. By 1997 the reps had revertedto their own - often paper-based - systems for tracking customer data. As aresult, "We didn't know how to get in touch with our customers anddidn't know what equipment was out there," says Buhl. "If a districtsales manager left the company, we had no clue what was going on inthat geography."
Although the lack of a centralized customer database was tolerablewhile AMF's global business was booming, the boom ended abruptly whenthe Asian market collapsed in 1998. Suddenly, AMF could no longer relyon outfitting new bowling centers in Asia for its growth. To makematters worse, an aggressive expansion campaign increased the numberof AMF-owned bowling centers in the United States from 125 to 415 intwo years, making it the world's largest owner and operator of bowlingcenters. Because AMF sells equipment to its own centers at cost, thecompany effectively decreased its universe of profitable salesprospects. Between 1997 and 1998, worldwide sales of AMF bowlingproducts plummeted from $300 million to $150 million.
Aiming for a Perfect Game
John Walker, who became vice president of North American sales in June1998 (and is now senior vice president of global sales and service),wanted CRM for three reasons. First, he was eager to improve customerservice by facilitating internal communication among reps, thetechnical support help desk and the credit department. Reps were oftengetting blindsided on sales calls to customers wrestling withtechnical problems; other times, customers targeted for big sales werebeing hassled by the credit department. Second, Walker wanted topursue targeted marketing of customers instead of blanket promotions.Third, he was convinced a CRM system could help reps better managetheir time and accounts - and facilitate smoother transitions when repsleave. But for CRM to do any good, the system had to be easyto use.
With Y2K looming in 1999, CRM took a backseat to replacing legacyback-office systems. The company did, however, have to pick softwarefor the tech service help desk - and did so with CRM in mind, says JohnFottrell, vice president of IS. Upon evaluating several products, hesays, AMF chose Applix on the basis of its strong help desk functions,strong CRM functions (the ability to keep track of contacts, maintaindetailed customer profiles and manage quotes) and its capacity formultilingual, multicurrency support.
Once the legacy systems were replaced, Walker pressed for a CRMinitiative and won the go-ahead to invest $325,000 on the SFA project.From the beginning, it was not an IT project but a businessinitiative, says Fottrell, who points to Walker's active involvementas critical to the system's success. Walker had watched a previousemployer waste $1 million developing a CRM system that users hatedbecause the project was driven by what management and IT liked ratherthan what users wanted. To avoid that trap - and to address theskepticism of veterans of the ill-fated Notes system - Walker andFottrell involved sales reps from the start. In October 2000, theyconvened a kickoff meeting, which included six sales reps, SystemsArchitect Harsha Nagaraj and an Applix project manager. Walker set thetone by making it clear that it was the reps' opinions that countedmost.
Reps were encouraged to be frank about what they hated about previoussystems and what they wanted from the new one - and they didn't alwaysagree. "We grunted it out for an entire afternoon - after which I wassure we were not going to be successful," says Fottrell. "It shockedme that they didn't have consensus about what they wanted."
During three or four sessions, the reps, with the help of IT, hammeredout specs for the system and gave feedback on rough prototypes. Andthe reps themselves came up with the idea to use PDAs; putting Applixon Palm devices would let them collect customer data in the bowlingcenters easily and discreetly.
But using the Palm meant overcoming some technical hurdles (such asgetting the data to sync with headquarters) and upgrading to thePalm-enabled version of Applix. The system is also limited by thesmaller screen size, the amount of information that a Palm can holdand the need to minimize sync time.
Nagaraj says that Eric Weimer, a district sales manager based inRichmond, Va., who had been using a Palm to track customer data on hisown, understood the constraints and convinced the other reps that theyneeded to limit the number of fields and the size of the drop-downmenus. For example, to collect data on pinspotters (the machines thatreset the pins), some reps wanted to include every model that AMF hadever made in the drop-down menu. Weimer argued that a salesperson justneeds to know whether a customer has an AMF pinspotter; in most cases,the model doesn't matter since the main selling proposition is whenit's not an AMF product. So the group whittled a drop-down list of 22choices to eight critical items. Reps also asked to see the status ofall of their customers' sales orders. But because a typical orderincludes 80 to 100 line items, that would have dramatically increasedsync time and overloaded the Palms. Instead, they decided to give repsjust information on shortages (when AMF can't immediately supplyeverything ordered) and warranty orders (when warrantied equipmentbreaks and replacement parts must be shipped quickly). This keeps repsin the loop when customers have problems.
To help fine-tune the application, Fottrell and Nagaraj enlistedWeimer to represent the tech-savvy rep's viewpoint. They also gotinput from Dave Cykoski, a longtime AMF rep based in Jacksonville,Fla., "because we thought he would never give up his Daytimer," saysFottrell. "We knew that the rest of the salespeople were somewhere inbetween."
At the December 2000 sales meeting, Weimer led his colleagues througha hypothetical day in the life of an AMF rep using Applix, showingthem how they could use the system to focus on the customers mostlikely to spend money, as well as how to update prospect or customerinformation. He also illustrated how they could use their laptop(which they'd keep in their hotel room) to view messages from thecredit department or see their customers' unresolved calls to the helpdesk. Weimer told his fellow reps, "You'll have the opportunity tosay, 'I know you talked to AMF a couple days ago, and we didn't getback to you, but this is the answer.' You become more valuable, andcustomers will want to talk to you more than someone doing a coldcall."
Weimer deftly fielded reps' questions and challenges. "He was rightback at 'em with some real, specific business reasons as to why it wasdone the way it was done," recalls Fottrell. "There's no way an ITperson could've had the same credibility."
The reps got their Palms and training on the Applix software at theMarch 2001 sales meeting. Because AMF was about to enter Chapter 11,reps were understandably distracted. AMF's aggressive investment inbowling centers in the late '90s had left it with what Buhl calls "abroken balance sheet." Before filing for bankruptcy in July 2001, thecompany laid off 171 people, including two district sales managers.(It emerged with a restructured debt in March 2002.)
Even so, Buhl made collecting customer data a priority. By December2001, reps had collected data on 80 percent of AMF's customers, heestimates. Through a series of pull-down menus on their Palm, theremaining 21 North American sales reps could easily fill out or updatea customer's equipment profile "on their way from entering the bowlingcenter to the front desk," says Buhl. "It's less than a five-minuteevent." Syncing data with headquarters proved to be a breeze: Whenreps dial into headquarters, they just put their Palm in the cradleattached to their laptop and push the Hot Sync button. Thatsynchronizes data on their Palm with the SQL server inRichmond - aprocess that takes 10 minutes tops, according to Nagaraj.
Although Cykoski still hasn't thrown out his Daytimer, he gives theApplix system a B grade (he admits that if he devoted more time toexploring what it can do, he might give it an A).
No More Gutter Balls
Now that AMF has a central repository of updated customer data, thecompany can send targeted promotions to customers most likely torespond to them. Buhl reports that two out of three promotions runthis year using data gathered through Applix have boosted sales.Following a January promotion on pinspotters, AMF's year-to-datepinspotter revenue reached $403,300 in April, up from just $9,000 forthe same time period in 2001. Likewise a February promotion onautomatic scoring equipment yielded $601,000 in revenue for thatcategory through April, up from $252,000 last year.
The ability to target promotional mailings saves the company money andmakes customers feel like AMF understands them. In the past, allpromotions went to each one of the 6,000 U.S. bowling centers. SinceAMF can now send a mailing to just the 500 or 1,000 accounts for whomit would be appropriate, the cost savings per promotion isapproximately 60 percent, according to Buhl. Perhaps more important,AMF is no longer annoying customers by sending them irrelevantoffers - or alienating those who'd just purchased the product in questiona week or two before.
In the field, reps are using Applix to focus their efforts on sellingcustomers what they're most likely to buy and to zero in on thecustomers who are most likely to be the biggest spenders. BeforeApplix, the Northeast district's Keller would have wasted time tryingto sell a product that was on special to bowling centers that didn'tneed it. With Applix, he says, "I can focus on the correct customerwith the correct product." Using this strategy, Keller managed to sell$212,014 worth of bumpers in a single month - surpassing his fourthquarter 2001 quota of $110,206 by 92 percent. Cykoski too has had goodresults with the Palm. Midway through 2002, his sales were alreadyapproaching his 2001 total.
Because reps can gather customer data much more efficiently, they cancall on more customers. Keller, for example, says he can now visit onemore account each day he's on the road. And sometimes that extra stopcan really pay off. One day Keller decided to visit a military basebecause Applix data showed that all of the equipment in its bowlingcenter was old and no AMF salesperson had visited recently. Turns outthe bowling center was about to close and move to a newlocation - andKeller ultimately closed a $400,000 deal at the new facility.
"This gives us an advantage over the competition," Keller says."Knowing what's out there, and who has what and what they need savesour time. The proof of the pudding is that we use it."