SLIDE 1
P A R T O F
O P T I M I S I N G S E A R C H A N O N -T E C H N I C A L P E R S P E C T I V E A n d r e a s B r u c k n e r S e n i o r M e r c h a n d i s i n g C o n s u l t a n t
Luxury brand 1 & Luxury Brand 2
O P T I M I S I N G S E A R C H A N O N -T E C H N I C A L P E R - - PowerPoint PPT Presentation
O P T I M I S I N G S E A R C H A N O N -T E C H N I C A L P E R - - PowerPoint PPT Presentation
O P T I M I S I N G S E A R C H A N O N -T E C H N I C A L P E R S P E C T I V E P A R T O F A n d r e a s B r u c k n e r S e n i o r M e r c h a n d i s i n g C o n s u l t a n t Just some of the brands and retailers we are helping
SLIDE 2
SLIDE 3
1 . B U I L D A D E D I C AT E D S E A R C H T E A M
SLIDE 4
- 1. BUILD A DEDICATED SEARCH TEAM
4
Merchandising Team Tech Team Own product performance Own product data Own infrastructure Own features Should do: Good at: Know users and trends Know the shop features
SLIDE 5
- 1. BUILD A DEDICATED SEARCH TEAM
5
Merchandising Team Tech Team Own product performance Own product data Own infrastructure Own features Should do: If they own search: Make many small scale changes and create problems that didn’t exist Make too few big changes and solve problems that don’t exist Good at: Know users and trends Know the shop features
SLIDE 6
- 1. BUILD A DEDICATED SEARCH TEAM
6
Merchandising Team Tech Team Own product performance Own product data Own infrastructure Own features Own user journey (How to use shop features to find the right products) Should do: If they own search: Make many small scale changes and create problems that didn’t exist Make too few big changes and solve problems that don’t exist Good at: Know users and trends Know the shop features Analyse and generalise
SLIDE 7
- 1. BUILD A DEDICATED SEARCH TEAM
7
No dedicated search expert 14%
Ownership of search at top Fredhopper customers:
SLIDE 8
- 1. BUILD A DEDICATED SEARCH TEAM
8
Dedicated search expert(s) 19% No dedicated search expert 14%
Ownership of search at top Fredhopper customers:
SLIDE 9
- 1. BUILD A DEDICATED SEARCH TEAM
9
Dedicated search expert(s) 19% Dedicated finding expert 67% No dedicated search expert 14%
Ownership of search at top Fredhopper customers:
SLIDE 10
2 . S E T R E A L I ST I C E X P E C TAT I O N S
SLIDE 11
0% 20% 40% 60% 80% 100% Precision Recall Score
- 2. SET REALISTIC EXPECTATIONS
11
SLIDE 12
0% 20% 40% 60% 80% 100% Precision Recall Score
- 2. SET REALISTIC EXPECTATIONS
12
Benchmark
SLIDE 13
0% 20% 40% 60% 80% 100% Precision Recall Score
- 2. SET REALISTIC EXPECTATIONS
13
Benchmark
SLIDE 14
0% 20% 40% 60% 80% 100% Precision Recall Score
- 2. SET REALISTIC EXPECTATIONS
14
Benchmark £1 m to £50m p.a.
SLIDE 15
3 . B U I L D AWA R E N E S S A C R O S S T H E O R G A N I S AT I O N
SLIDE 16
- 3. BUILD AWARENESS ACROSS THE ORGANISATION
16
Wrong data
SLIDE 17
- 3. BUILD AWARENESS ACROSS THE ORGANISATION
17
Wrong data In Inconsistent data
SLIDE 18
- 3. BUILD AWARENESS ACROSS THE ORGANISATION
18
Wrong data In Inconsistent data Mis isle leading data
SLIDE 19
4 . M E A S U R E T H E R I G H T T H I N G S
SLIDE 20
- 4. MEASURE THE RIGHT THINGS
20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 0.00% 1.00% 2.00% 3.00% 4.00% 5.00% 6.00% 7.00% 8.00% 9.00% Search CVR
Started changes
SLIDE 21
- 4. MEASURE THE RIGHT THINGS
21
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 0.00% 1.00% 2.00% 3.00% 4.00% 5.00% 6.00% 7.00% 8.00% 9.00% Signficant visits CVR Search CVR
Started changes
SLIDE 22
- 4. MEASURE THE RIGHT THINGS
22
0.00% 1.00% 2.00% 3.00% 4.00% 5.00% 6.00% 7.00% 8.00% 9.00% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 0% 20% 40% 60% 80% 100% 120% 140% 160% Search NCVR Signficant visits CVR Search CVR
Started changes
SLIDE 23
- 4. MEASURE THE RIGHT THINGS
23
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49
PDP clicks per search session Week
Search PVR
Started changes
SLIDE 24
- 4. MEASURE THE RIGHT THINGS
24
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51
PDP clicks per search session Week
Search PVR
Started changes
SLIDE 25
- 4. MEASURE THE RIGHT THINGS
- Online evaluation
- Search result click rate
- MRR of first click
- Search refinement rate
- Offline evaluation
- Precision@1
- Precision@4
- Precision (all results)
- Recall
+ additional low level metrics
25
SLIDE 26
5 . B U I L D A M E A S U R E M E N T F R A M E W O R K
SLIDE 27
- 5. BUILD A MEASUREMENT FRAMEWORK
28
74% 13% 8% 5% € - € 0.10 € 0.20 € 0.30 € 0.40 € 0.50 € 0.60 € 0.70 € 0.80 € 0.90 € 1.00 Value per search Exact match Spellcheck Partial match No results
SLIDE 28
6 . M I X O N L I N E A N D O F F L I N E E VA LUAT I O N
SLIDE 29
- 6. MIX ONLINE AND OFFLINE EVALUATION
31
Weekly: Operational analysis
- Find areas to
improve
Monthly: tactical analysis
- Evaluate
performance
Ad-hoc analysis
- Evaluate
impact of new features
Bi-annually: strategic analysis
- Find pattern
- Find shifts in
behaviour
SLIDE 30
- 6. MIX ONLINE AND OFFLINE EVALUATION
32
Weekly: Operational analysis
- Find areas to
improve
Monthly: tactical analysis
- Evaluate
performance
Ad-hoc analysis
- Evaluate
impact of new features
Bi-annually: strategic analysis
- Find pattern
- Find shifts in
behaviour
Online evaluation Offline evaluation
SLIDE 31
7 . E C O M M E R C E I S D I F F E R E N T
SLIDE 32
- 7. ECOMMERCE IS DIFFERENT
34
dress
Name Summer dress Category Casual dresses, Beach dresses Style Skater dress Name Evening dress Category Evening wear Style Formal
SLIDE 33
8 . STA R T W I T H U S E R N E E D S
SLIDE 34
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% Product Type Brand Product Features Individual Product Per ercentage of
- f sea
searches
- 8. START WITH USER NEEDS
37
SLIDE 35
78% 44% 34% 5% 14% 0.2% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% Product Type Brand Product Features Individual Product Gender Size Per ercentage of
- f sea
searches
- 8. START WITH USER NEEDS
38
SLIDE 36
9 . D O N ’ T F I X W H AT C A N ’ T B E F I X E D
SLIDE 37
- 9. DON’T FIX WHAT CAN’T BE FIXED
41
SLIDE 38
1 0 . Q UA N T I F Y A L L C H A N G E S
SLIDE 39
- 10. QUANTIFY ALL CHANGES
46
Increase revenue UX: improve search results Add searchable attributes +16% to 28% search PVR +8% to 14% search CVR +3% to 6%
- verall CVR
+£15m to £30m annual revenue £500m annual revenue 40% of revenue from search +1% PVR = +0.5% CVR
SLIDE 40
S U M M A RY
SLIDE 41
SUMMARY
- 1. Build a dedicated search team
- 2. Set realistic expectations (no 100% search quality)
- 3. Build awareness in the wider company (with buying team, etc.)
- 4. Measure the right things
- 5. Build a KPI framework
- 6. Mix online and offline evaluation
- 7. Ecommerce is different
- 8. Start with user needs, not features
- 9. Don’t fix what can’t be fixed.
10.Quantify all changes
48
Organisation Measurement Optimisation
SLIDE 42
C O N TA C T U S
F O R G E N E R A L E N Q U I R I E S sayhello@attraqt.com www.attraqt.com +44 (0) 20 3675 7800 C O N T A C T M E Andreas Bruckner abrueckner@attraqt.com