Voxeo CXP Analyzer Report Samples
Analysis of phone self-services including
business
tasks, customer behavior, application
tuning,
TABLE OF CONTENTS
Introduction ... 2
Business Task Analysis ... 3
Personalization Analysis ... 4
Dominant Path Analysis... 5
Menu Navigation Analysis ... 6
Recognition Quality Analysis ...7
Session Input State Analysis ... 8
Utterance Analysis ... 9
Number of Sessions ... 10
Session Partitioning Analysis ...11
INTRODUCTION
Key to delivering effective phone self-service today is the ability to understand what customers are doing and experiencing within the self-service phone portal – in real-time.
Voxeo CXP Analyzer is a service analysis environment enabling instant reporting on key business, operational, and application design questions. In addition to enabling users to create custom reports, Voxeo CXP Analyzer delivers over 50 out-of-the-box reports providing valuable insight into the usage and acceptance of application services by leveraging the power of common business intelligence frameworks such as SAP BusinessObjects, IBM Cognos, and MicroStrategy.
By providing up-to-the-minute graphical views of the application and system status, business analysts, marketing specialists, dialog designers, developers, administrators, and system operators can now efficiently:
• use standard functionality for easy and powerful analysis • create customized reports and distribute reports automatically • integrate with existing data and analysis
• leverage existing knowledge, infrastructure and resources
This document provides 10 report examples that highlight the analytical capabilities of
B U SI N E SS T ASK AN AL Y SI S
Self - serv i c e appli c ati ons most of ten map to one or more u nd erlyi ng b u si ness proc esses. Proc esses li k e tariff c h ang e, prod u c t ord er or c h ang e of payment meth od typi c ally exi st as b ac k end proc esses. T h e ph one self - serv i c es ac t as i nterf ac e to th ose proc esses.
B u t, h ow su c c essf u l are th ese proc esses su pported and d oes th e serv i c e really ad d v alu e? W h at i s th e transac ti on c ompleti on rate and w h at are th e reasons f or i nc omplete transac ti ons?
ï Si mply i d enti f y th e task s th at are ac c essed most of ten or w h i c h are not u ti li zed as expec ted . ï M oni tor th e task c ompleti on rate.
ï I d enti f y reasons f or task i nc ompleti on, i nc lu d i ng analysi s of rec og ni ti on q u ali ty, d omi nant path or d i alog exi t analysi s.
ï Analyze task u sag e and task c ompleti on b y c h annel.
PE RSO N AL I Z AT I O N AN AL Y SI S
D i alog personali zati on i s one k ey su c c ess f ac tor f or ph one self - serv i c es. Personali zati on h elps to b est ad j u st d i alog s to th e c allers need s and ab i li ti es. I t allow s f or i nterac ti v e serv i c es and enab les c orporati ons to g et th e b est ou t of i t. H ow d oes th e c all ow personali zati on w ork f or you r c allers? I s personali zati on h elpi ng c u stomers u se th e
appli c ati on? I d enti f y f req u ent u ser selec ti ons or i d enti f y typi c al u ser pro les f or f u rth er appli c ati on tu ni ng s, to opti mi ze d i alog personali zati on and th e c aller experi enc e. T h i s analysi s c an also su pport mark eti ng and sales i n planni ng mark eti ng ac ti v i ti es.
ï I d enti f y f req u ent u ser selec ti ons. ï I d enti f y typi c al u ser pro les. ï I mprov e th e c aller experi enc e. ï Plan one- to- one mark eti ng ac ti v i ti es.
D O M I N AN T PAT H AN AL Y SI S
D omi nant path analysi s prov i d es d etai led i nf ormati on on h ow u sers b eh av e and h ow u sers nav i g ate th rou g h th e appli c ati on. Analyzi ng d omi nant path s c an h elp to i d enti f y i neffic i ent experi enc es, b arri ers to c ompleti on and v i si ts to i nv ali d areas. I t prov i d es i nf ormati on ab ou t th e d omi nant experi enc es of th e maj ori ty of c allers and h elps to ev alu ate th e u sef u lness of appli c ati ons and mod u les.
D omi nant path analysi s answ ers q u esti ons i nc lu d i ng : W h atí s you r c u stomersí b eh av i or? H ow d o c allers nav i g ate th rou g h th e d i alog ? D oes th e appli c ati on ë w ork í as d esi g ned ? Are th ere spec i c areas of i nterest or are th ere some d ead end s?
ï I d enti f y c aller nav i g ati on patterns. ï Vi su ali ze th e d omi nant u ser experi enc es.
ï I d enti f y appli c ati on aw s and i neffic i ent experi enc es. ï E v alu ate th e u sef u lness of appli c ati ons and mod u les.
M E N U N AVI G AT I O N AN AL Y SI S
Si mi lar to d omi nant path analysi s, b u t on th e more d etai led lev el of M enu ob j ec ts, th i s analysi s sh ow s w h i c h selec ti ons are mad e b y th e c allers. Answ ers c an b e prov i d ed to q u esti ons li k e: H ow d o c allers nav i g ate th rou g h th e appli c ati on? W h at are pref erred c h oi c es? W h at i s th e d omi nant i npu t mod e, v oi c e or D T M F ? D o appli c ati on aw s exi st th at prev ent u sers f rom reac h i ng spec i c serv i c e areas?
ï Analyze h ow c allers nav i g ate th rou g h th e appli c ati on. ï I d enti f y pref erred and less u sed c h oi c es.
ï I d enti f y appli c ati on aw s. ï M oni tor u sag e of h yperli nk s.
RE CO G N I T I O N Q U AL I T Y AN AL Y SI S
M oni tor th e rec og ni ti on q u ali ty on serv i c e lev el and d ri ll d ow n f or a d etai led analysi s to a si ng le i npu t state. T h i s analysi s prov i d es i nf ormati on ab ou t th e su c c ess rate of i npu t states along w i th th e k ey perf ormanc e i nd i c ators to analyze th e reason f or b ad rec og ni ti on q u ali ty.
H ow m any i npu ts w h ere answ ered su c c essf u lly? I n w h i c h mod u les and i npu ts d o w e h av e a lac k i n rec og ni ti on q u ali ty? H ow many w h ere answ ered i mmed i ately or h ow mu c h N o M atc h or N o I npu t ev ents oc c u rred ? W h at w as th e av erag e c on d enc e?
ï M oni tor th e rec og ni ti on q u ali ty.
ï I d enti f y prob lemati c mod u les and i npu t states. ï Id enti f y reasons f or b ad rec og ni ti on q u ali ty.
ï I d enti f y appli c ati on aw s and i neffic i ent experi enc es. ï E v alu ate th e u sef u lness of appli c ati ons and mod u les.
SE SSI O N I N PU T ST AT E AN AL Y SI S
Analyze c aller b eh av i or and rec og ni ti on q u ali ty on th e low est lev el of d etai l: A si ng le c all. W h at w h ere th e steps th e c aller h as tak en? W h i c h answ ers d i d th e c aller g i v e? At w h i c h i npu t states d i d prob lems oc c u r? And w h at d i d th e c aller really say?
ï Analyze si ng le c alls step- b y- step f rom start to end . ï Analyze w h i c h i npu t states w h ere u sed .
ï Analyze w h at th e u ser sai d and w h i c h prob lems oc c u rred . ï L i sten to th e u tteranc e rec ord i ng s.
ï I d enti f y opti ons to opti mi ze g rammars.
U T T E RAN CE AN AL Y SI S
M easu re th e c omplexi ty of u ser responses to opti mi ze prompts and g rammars. W h atí s th e av erag e nu mb er of w ord s u sed w i th i n th e i npu t states? And , w h atí s th e av erag e c on d enc e d epend i ng on th e leng th of th e response? H ow many v ari ati ons d o th e c allers u se?
ï Analyze c aller u tteranc es.
ï I d enti f y i npu t states w i th c omplex responses. ï I d enti f y opti ons to opti mi ze g rammars.
N U M B E R O F SE SSI O N S
T h e Number of Sessions report h elps you to g et a g eneral ov erv i ew on th e serv i c e u sag e and to i d enti f y trend s i n th e u sag e f req u enc y of you r serv i c e. W h atí s th e long term u sag e trend of you r serv i c e? W h at are most ac ti v e d ays of th e w eek ? At w h at d ays w ou ld system mai ntenanc e d ow nti me h as th e least i mpac t? W h at are most f req u ent h ou rs of th e d ay?
ï M oni tor th e ov erall serv i c e u sag e. ï I d enti f y trend s on serv i c e u sag e. ï M oni tor g row i ng c onc u rrenc y.
ï Su pport system mai ntenanc e and system planni ng .
SE SSI O N PART I T I O N I N G AN AL Y SI S
T h e Sessi on Parti ti oni ng Analysi s report h elps you to g et an ov erv i ew on you r c onc u rrent sessi on li mi ts f or serv ers
and serv i c es.
W h at are serv er li c ense li mi ts? W h at i s th e u ti li zati on of c on g u red sessi on g u arantees? W h atí s th e u ti li zati on of li mi ts f or spec i c serv i c es?
ï M oni tor th e li c ense u ti li zati on.
ï M oni tor u ti li zati on of c on g u red sessi on g u arantees, oati ng sessi ons and sessi on li mi ts. ï O pti mi ze th e u ti li zati on of resou rc es.
VO XE O CXP AN AL Y Z E R RE PO RT SU M M ARY
Voxeo CXP Analyzer i nc lu d es 5 0 reports f or b u si ness and mark eti ng u sers, d ev elopers, d esi g ners and ad mi ni strators to g ai n end - to- end i nsi g h t i nto system and serv i c e u sag e. U se Voxeo CXP Analyzer to effic i ently moni tor b u si ness task c ompleti on and analyze c u stomer b eh av i or.
ADMINISTRATION AND MAINTENANCE
ï Ch annel and D ri v er D i stri b u ti on ï Clu ster E ffic i enc y
ï Clu ster E ffic i enc y T rend ï N u mb er of Rec ord i ng s ï N u mb er of Sessi ons b y D ay ï N u mb er of Sessi ons b y
H ou r of D ay
ï N u mb er of Sessi ons b y M onth ï N u mb er of Sessi ons b y
W eek d ay
ï Proc essi ng v s. Sessi on D u rati on ï Serv er Stati sti c s
ï Sessi on Parti ti oni ng Analysi s
APPLICATION DEVELOPMENT
AND TUNING
ï Av erag e N u mb er of Sessi ons ï Caller E xi ts b y I npu t States ï Caller E xi ts b y M od u les ï D i alog L ang u ag e ï D i alog L ang u ag e T rend ï D T M F F allb ac k Analysi s ï E v ent D i stri b u ti on ï E v ents b y D ay ï E xi t T ype D i stri b u ti on ï E xi t T ypes T rend ï I npu t State Stati sti c s ï M enu N av i g ati on Analysi s ï M od u le Analysi s
ï Rec og ni ti on Q u ali ty ï Rec ord i ng s b y I npu t State ï Serv i c e Analysi s
ï Serv i c e Ch ai n Sessi on Analysi s ï Serv i c e D eployment H i story ï Sessi on D u rati on B enc h mark ï Sessi on I npu t States Analysi s ï T ransi ti on T ypes
ï U tteranc e Analysi s ( Voi c e, D T M F , T ext)
ï U tteranc es b y I npu t States ï Voi c e v s. D T M F I npu t
BUSINESS AND CALLER ANALYSIS
ï B u si ness T ask Completi on Rates ï B u si ness T ask D ata
ï B u si ness T ask Sessi on Analysi s ï Caller F req u enc y
ï Caller L oyalty
ï Ch annel T rend Analysi s ï D i alog D epth
ï D omi nant Path Analysi s ï E arly D rop- O ff Analysi s ï L ayer U sag e O v erv i ew ï M od u le Seq u enc es ï M od u le Sets ï Sessi on D etai ls ï Su b seq u enc e Analysi s ï T op 2 0 Callers