EMF Analysis
Key Takeaways
Renewal by Decision Type
- NNLOA is exactly at 50%: towards the bottom of all forums.
- Opt Out performance and volume is helping overall renewal rate, as opt outs are ~32% of total decisions.
- Keep the 1.0 Ratio for NNLOA in mind throughout the analysis.
Renewal Ratio by Decision Type
|
DecisionType
|
N
|
ATL
|
Dropped
|
Ratio
|
Perc_Renewed
|
|
Need New LOA
|
42
|
21
|
21
|
1.0
|
50.0%
|
|
Opt Out
|
20
|
18
|
2
|
9.0
|
90.0%
|
|
Total
|
62
|
39
|
23
|
1.7
|
62.9%
|
Renewal Rate by Account Segment
- Some consistency in performance across segments:
- Regional Private and Large P&P renew at higher rates.
- Regional Privates are small N, but still surprising to see a high rate.
- Half of the partnerships are with Large P&P schools.
- Regional Public and Selectives perform equivalently and low
Renewal Ratio by Account Segment
|
AccountSegment
|
N
|
ATL
|
Dropped
|
Ratio
|
Perc_Renewed
|
|
Large Public & Private
|
31
|
22
|
9
|
2.4
|
71.0%
|
|
Regional Private
|
5
|
4
|
1
|
4.0
|
80.0%
|
|
Regional Public
|
14
|
7
|
7
|
1.0
|
50.0%
|
|
Selective
|
12
|
6
|
6
|
1.0
|
50.0%
|
Renewal by Decision Type & Segment
- [Large P&P]
- Performance on NNLOA is roughly what we have seen in other flagship forums
- Performance on Opt Outs is superb - perfect score.
- [Regional Private]
- Only NNLOA and performance is pretty strong.
- [Regional Public & Selective]
- NNLOA is a particular problem for Regional Public and Selective
- Opt Out Performance appears to be OK, even with low N.
Renewal Ratio by Account Segment
|
AccountSegment
|
DecisionType
|
N
|
ATL
|
Dropped
|
Ratio
|
Perc_Renewed
|
|
Large Public & Private
|
Need New LOA
|
21
|
12
|
9
|
1.3
|
57.1%
|
|
Large Public & Private
|
Opt Out
|
10
|
10
|
0
|
Inf
|
100.0%
|
|
Regional Private
|
Need New LOA
|
5
|
4
|
1
|
4.0
|
80.0%
|
|
Regional Public
|
Need New LOA
|
9
|
3
|
6
|
0.5
|
33.3%
|
|
Regional Public
|
Opt Out
|
5
|
4
|
1
|
4.0
|
80.0%
|
|
Selective
|
Need New LOA
|
7
|
2
|
5
|
0.4
|
28.6%
|
|
Selective
|
Opt Out
|
5
|
4
|
1
|
4.0
|
80.0%
|
GRP 3-4
- The performance for GRP 3-4 does not look very different across Large Public & Private Decision Types.
- Perhaps a small boost for Selective opt outs - super low N.
Renewal Ratio by Navigate Bundling
|
AccountSegment
|
DecisionType
|
GRP3-4
|
N
|
ATL
|
Dropped
|
Ratio
|
Perc_Renewed
|
|
Large Public & Private
|
Need New LOA
|
0
|
20
|
11
|
9
|
1.2
|
55.0%
|
|
Large Public & Private
|
Need New LOA
|
1
|
1
|
1
|
0
|
Inf
|
100.0%
|
|
Large Public & Private
|
Opt Out
|
0
|
4
|
4
|
0
|
Inf
|
100.0%
|
|
Large Public & Private
|
Opt Out
|
1
|
6
|
6
|
0
|
Inf
|
100.0%
|
|
Regional Private
|
Need New LOA
|
0
|
5
|
4
|
1
|
4.0
|
80.0%
|
|
Regional Public
|
Need New LOA
|
0
|
9
|
3
|
6
|
0.5
|
33.3%
|
|
Regional Public
|
Opt Out
|
0
|
3
|
2
|
1
|
2.0
|
66.7%
|
|
Regional Public
|
Opt Out
|
1
|
2
|
2
|
0
|
Inf
|
100.0%
|
|
Selective
|
Need New LOA
|
0
|
7
|
2
|
5
|
0.4
|
28.6%
|
|
Selective
|
Opt Out
|
0
|
4
|
3
|
1
|
3.0
|
75.0%
|
|
Selective
|
Opt Out
|
1
|
1
|
1
|
0
|
Inf
|
100.0%
|
Impact Interaction Volume Histogram
- Overall low engagement: a lot of partners in 0, 1, 2 categories, with a peak at 2.
- Thick right tail suggests some heavier users in the 5+ categories.
##### Impact Interaction Volume Histogram Split by Decision
- Opt Outs are spread out.
- Relatively uniform distribution of Opt Outs and NNLOA in the tail.
- NNLOAs represent most of the 0-2 utilization bins.

Renewal by Interaction Volume and Decision Type, Unbinned
- NNLOA performance varies a lot due to small N.
- No apparent threshold as the variation in higher utilization makes a cutoff difficult without binning.
- Utilization at 0-2 for NNLOA is deadly as 10 fo 16 partners dropped.
- Opt Out performance is invariant to Impact interaction volume.
Renewal Ratio by Impact Interaction Volume
|
DecisionType
|
II_Credited_Volume
|
N
|
ATL
|
Dropped
|
Ratio
|
Perc_Renewed
|
|
Need New LOA
|
0
|
1
|
0
|
1
|
0.0
|
0.0%
|
|
Need New LOA
|
1
|
4
|
3
|
1
|
3.0
|
75.0%
|
|
Need New LOA
|
2
|
11
|
3
|
8
|
0.4
|
27.3%
|
|
Need New LOA
|
3
|
8
|
5
|
3
|
1.7
|
62.5%
|
|
Need New LOA
|
4
|
5
|
3
|
2
|
1.5
|
60.0%
|
|
Need New LOA
|
5
|
7
|
3
|
4
|
0.8
|
42.9%
|
|
Need New LOA
|
7
|
2
|
1
|
1
|
1.0
|
50.0%
|
|
Need New LOA
|
8
|
2
|
2
|
0
|
Inf
|
100.0%
|
|
Need New LOA
|
9
|
1
|
1
|
0
|
Inf
|
100.0%
|
|
Need New LOA
|
10
|
1
|
0
|
1
|
0.0
|
0.0%
|
|
Opt Out
|
1
|
1
|
1
|
0
|
Inf
|
100.0%
|
|
Opt Out
|
2
|
4
|
4
|
0
|
Inf
|
100.0%
|
|
Opt Out
|
3
|
2
|
1
|
1
|
1.0
|
50.0%
|
|
Opt Out
|
4
|
4
|
4
|
0
|
Inf
|
100.0%
|
|
Opt Out
|
5
|
4
|
3
|
1
|
3.0
|
75.0%
|
|
Opt Out
|
6
|
1
|
1
|
0
|
Inf
|
100.0%
|
|
Opt Out
|
8
|
2
|
2
|
0
|
Inf
|
100.0%
|
|
Opt Out
|
9
|
1
|
1
|
0
|
Inf
|
100.0%
|
|
Opt Out
|
10
|
1
|
1
|
0
|
Inf
|
100.0%
|
Renewal by Impact Interaction Volume, (Binned)
- [NNLOA] When you take the variation out, you might say 7+ is the threshold.
Renewal Ratio for Impact Interaction Volume
|
DecisionType
|
II_Volume_Bins
|
N
|
ATL
|
Dropped
|
Ratio
|
Perc_Renewed
|
|
Need New LOA
|
0
|
1
|
0
|
1
|
0.0
|
0.0%
|
|
Need New LOA
|
1
|
4
|
3
|
1
|
3.0
|
75.0%
|
|
Need New LOA
|
2-3
|
19
|
8
|
11
|
0.7
|
42.1%
|
|
Need New LOA
|
4-6
|
12
|
6
|
6
|
1.0
|
50.0%
|
|
Need New LOA
|
7+
|
6
|
4
|
2
|
2.0
|
66.7%
|
|
Opt Out
|
1
|
1
|
1
|
0
|
Inf
|
100.0%
|
|
Opt Out
|
2-3
|
6
|
5
|
1
|
5.0
|
83.3%
|
|
Opt Out
|
4-6
|
9
|
8
|
1
|
8.0
|
88.9%
|
|
Opt Out
|
7+
|
4
|
4
|
0
|
Inf
|
100.0%
|
Renewal by Previous Yr Impact Interaction Volume & Decision
- [NNLOA] Renewal rates vary a lot, but it is important to note the heavy part of the distribution is in the 3-5 range and the renewal rate is below average.
- [Opt Out] Invariance.
Renewal Ratio by Previous Year Impact Interaction
|
DecisionType
|
PreviousYr_II_Volume
|
N
|
ATL
|
Dropped
|
Ratio
|
Perc_Renewed
|
|
Need New LOA
|
0
|
2
|
2
|
0
|
Inf
|
100.0%
|
|
Need New LOA
|
1
|
2
|
1
|
1
|
1.0
|
50.0%
|
|
Need New LOA
|
2
|
4
|
3
|
1
|
3.0
|
75.0%
|
|
Need New LOA
|
3
|
6
|
0
|
6
|
0.0
|
0.0%
|
|
Need New LOA
|
4
|
6
|
2
|
4
|
0.5
|
33.3%
|
|
Need New LOA
|
5
|
7
|
3
|
4
|
0.8
|
42.9%
|
|
Need New LOA
|
6
|
1
|
0
|
1
|
0.0
|
0.0%
|
|
Need New LOA
|
7
|
4
|
3
|
1
|
3.0
|
75.0%
|
|
Need New LOA
|
8
|
3
|
3
|
0
|
Inf
|
100.0%
|
|
Need New LOA
|
NA
|
7
|
4
|
3
|
1.3
|
57.1%
|
|
Opt Out
|
0
|
1
|
1
|
0
|
Inf
|
100.0%
|
|
Opt Out
|
1
|
1
|
0
|
1
|
0.0
|
0.0%
|
|
Opt Out
|
3
|
2
|
2
|
0
|
Inf
|
100.0%
|
|
Opt Out
|
4
|
1
|
1
|
0
|
Inf
|
100.0%
|
|
Opt Out
|
7
|
5
|
5
|
0
|
Inf
|
100.0%
|
|
Opt Out
|
8
|
2
|
1
|
1
|
1.0
|
50.0%
|
|
Opt Out
|
NA
|
8
|
8
|
0
|
Inf
|
100.0%
|
Renewals Previous Yr and Current Yr Interaction Volume
Only looking at NNLOAs in the below table.
- Maintenance looks like it is key. 16 fall in the 3+,3+ bucket and renew at a high rate.
- Dropping to 0 is undesirable.
Renewal Ratio by Previous and Current Year Impact Interaction
|
II_PrevYr_Volume_Bins
|
II_Volume_Bins
|
N
|
ATL
|
Dropped
|
Ratio
|
Perc_Renewed
|
|
0
|
1
|
1
|
1
|
0
|
Inf
|
100.0%
|
|
0
|
2-3
|
1
|
1
|
0
|
Inf
|
100.0%
|
|
1
|
2-3
|
2
|
1
|
1
|
1.0
|
50.0%
|
|
2-3
|
1
|
2
|
1
|
1
|
1.0
|
50.0%
|
|
2-3
|
2-3
|
5
|
1
|
4
|
0.2
|
20.0%
|
|
2-3
|
4-6
|
2
|
0
|
2
|
0.0
|
0.0%
|
|
2-3
|
7+
|
1
|
1
|
0
|
Inf
|
100.0%
|
|
4-6
|
1
|
1
|
1
|
0
|
Inf
|
100.0%
|
|
4-6
|
2-3
|
5
|
1
|
4
|
0.2
|
20.0%
|
|
4-6
|
4-6
|
6
|
3
|
3
|
1.0
|
50.0%
|
|
4-6
|
7+
|
2
|
0
|
2
|
0.0
|
0.0%
|
|
7+
|
2-3
|
3
|
2
|
1
|
2.0
|
66.7%
|
|
7+
|
4-6
|
2
|
2
|
0
|
Inf
|
100.0%
|
|
7+
|
7+
|
2
|
2
|
0
|
Inf
|
100.0%
|
|
NA
|
0
|
1
|
0
|
1
|
0.0
|
0.0%
|
|
NA
|
2-3
|
3
|
2
|
1
|
2.0
|
66.7%
|
|
NA
|
4-6
|
2
|
1
|
1
|
1.0
|
50.0%
|
|
NA
|
7+
|
1
|
1
|
0
|
Inf
|
100.0%
|
Simple Regression for Impact Interaction Volume
Note: Looking at NNLOAs, only.
- Increasing Impact Interactions is not associated with changes in renewal probability.
Simple Regression Model with Impact Interaction Volume
|
term
|
estimate
|
std.error
|
statistic
|
p.value
|
|
Intercept
|
-0.41
|
0.594
|
-0.69
|
0.489
|
|
Impact Interaction Volume
|
0.11
|
0.139
|
0.81
|
0.420
|
Penetration by Event Grouping
- Opt Outs have higher penetration in most of the Event Groupings:
- Events (marginally), SL Led, Research Interviews, and Expert Center Calls.
- NNLOAs are higher in Service and PLW/Onsites (marginally)
Percent of Renewals w/ At Least 1 Interaction in a Group
|
DecisionType
|
Perc_Events
|
Perc_Service
|
Perc_SLLed
|
Perc_ResearchInt
|
Perc_PLW_Onsite
|
Perc_Experience
|
Perc_ExpertCall
|
|
Need New LOA
|
50.0%
|
33.3%
|
59.5%
|
42.9%
|
9.5%
|
0.0%
|
42.9%
|
|
Opt Out
|
55.0%
|
20.0%
|
70.0%
|
60.0%
|
5.0%
|
0.0%
|
70.0%
|
Renewal by Event Consumption
- [NNLOA] Small lift, 2.4%, above the average. And roughly a 5% bump in renewal from consuming vs. not.
- [Opt Outs] Attending an event is negatively related to renewal.
Renewal Ratio by Event Consumption
|
DecisionType
|
II_Events_Binary
|
N
|
ATL
|
Dropped
|
Ratio
|
Perc_Renewed
|
|
Need New LOA
|
0
|
21
|
10
|
11
|
0.9
|
47.6%
|
|
Need New LOA
|
1
|
21
|
11
|
10
|
1.1
|
52.4%
|
|
Opt Out
|
0
|
9
|
9
|
0
|
Inf
|
100.0%
|
|
Opt Out
|
1
|
11
|
9
|
2
|
4.5
|
81.8%
|
Renewal by Service Consumption
- [NNLOA] No difference between consuming this interaction and not. Plus, both are at the average.
- [Opt Out] Some evidence of benefit from Service Consumption - Small N.
Renewal Ratio by Service Consumption
|
DecisionType
|
II_Service_Binary
|
N
|
ATL
|
Dropped
|
Ratio
|
Perc_Renewed
|
|
Need New LOA
|
0
|
28
|
14
|
14
|
1
|
50.0%
|
|
Need New LOA
|
1
|
14
|
7
|
7
|
1
|
50.0%
|
|
Opt Out
|
0
|
16
|
14
|
2
|
7
|
87.5%
|
|
Opt Out
|
1
|
4
|
4
|
0
|
Inf
|
100.0%
|
Renewal by SL Consumption
- [NNLOA] Decent lift for having an SL Led Interaction - 10% above NNLOA average, 25% bump above not having one.
- [Opt Outs] SL consumption is negatively related to renewal. Not sure this is worth reacting to.
Renewal Ratio by SL-Led Consumption
|
DecisionType
|
II_SLLed_Binary
|
N
|
ATL
|
Dropped
|
Ratio
|
Perc_Renewed
|
|
Need New LOA
|
0
|
17
|
6
|
11
|
0.5
|
35.3%
|
|
Need New LOA
|
1
|
25
|
15
|
10
|
1.5
|
60.0%
|
|
Opt Out
|
0
|
6
|
6
|
0
|
Inf
|
100.0%
|
|
Opt Out
|
1
|
14
|
12
|
2
|
6.0
|
85.7%
|
Renewal by Research Interview Consumption
- [NNLOA] First instance where Research Interviews are not positively related to renewal. -5.6% below the average.
- [Opt Out] Opposite result from NNLOA. We’d conclude service consumption offers a decent lift.
Renewal Ratio by Research Interview Consumption
|
DecisionType
|
II_ResearchInterview_Binary
|
N
|
ATL
|
Dropped
|
Ratio
|
Perc_Renewed
|
|
Need New LOA
|
0
|
24
|
13
|
11
|
1.2
|
54.2%
|
|
Need New LOA
|
1
|
18
|
8
|
10
|
0.8
|
44.4%
|
|
Opt Out
|
0
|
8
|
6
|
2
|
3.0
|
75.0%
|
|
Opt Out
|
1
|
12
|
12
|
0
|
Inf
|
100.0%
|
Renewal by PLW OR Onsite Consumption
- [NNLOA] Onsite and/or PLW positively associated with renewal! Small N, though. Big lift, nearly 29%.
- [Opt Out] Not enough data to conclude anything.
Renewal Ratio by PLW OR Onsite Consumption
|
DecisionType
|
II_PLW_Onsite_Binary
|
N
|
ATL
|
Dropped
|
Ratio
|
Perc_Renewed
|
|
Need New LOA
|
0
|
38
|
18
|
20
|
0.9
|
47.4%
|
|
Need New LOA
|
1
|
4
|
3
|
1
|
3.0
|
75.0%
|
|
Opt Out
|
0
|
19
|
17
|
2
|
8.5
|
89.5%
|
|
Opt Out
|
1
|
1
|
1
|
0
|
Inf
|
100.0%
|
Renewal by Experience Consumption
- No data.
Renewal Ratio by Experience Consumption
|
DecisionType
|
II_Experience_Binary
|
N
|
ATL
|
Dropped
|
Ratio
|
Perc_Renewed
|
|
Need New LOA
|
0
|
42
|
21
|
21
|
1
|
50.0%
|
|
Opt Out
|
0
|
20
|
18
|
2
|
9
|
90.0%
|
Renewal by Expert Call Consumption
- [NNLOA] No difference between consuming this interaction and not. Plus, both are at the average.
- [Opt Out] Some evidence of benefit from Service Consumption - a bump of nearly 10% vs. not having the interaction.
Renewal Ratio by Expert Call Consumption
|
DecisionType
|
II_ExpertCall_Binary
|
N
|
ATL
|
Dropped
|
Ratio
|
Perc_Renewed
|
|
Need New LOA
|
0
|
24
|
12
|
12
|
1
|
50.0%
|
|
Need New LOA
|
1
|
18
|
9
|
9
|
1
|
50.0%
|
|
Opt Out
|
0
|
6
|
5
|
1
|
5
|
83.3%
|
|
Opt Out
|
1
|
14
|
13
|
1
|
13
|
92.9%
|
Correlation betwen Event Groupings, and Renewal (for fun)
- Correlation with Renewal: Despite what we saw above, volume of impact interaction consumption, across categories, correlates with renewal, positively.
- If we run this for only NNLOAs, then expert calls are the only negatively related impact interaction. SL Led and Service are highest positive correlations.
- Halo Effect: Events highly correlated with the other categories, except Research Interviews.
- Onsites and SL Led are the highest correlated event categories.
##### Multivariate Regression (still kind of simple)
Note: Looking at NNLOA, only.
- Nothing Pops as significant.
- If you control for segment, you’d find that SL Led is a significant impact interaction.
Regression Estimates for a Model Fitted on Event Grouping Volumes
|
term
|
estimate
|
std.error
|
statistic
|
p.value
|
|
(Intercept)
|
-0.83
|
0.774
|
-1.07
|
0.28
|
|
II_Events_Volume
|
0.31
|
0.383
|
0.81
|
0.42
|
|
II_SLLed_Volume
|
0.76
|
0.490
|
1.55
|
0.12
|
|
II_ResearchInterview_Volume
|
0.15
|
0.431
|
0.35
|
0.73
|
|
II_PLW_Onsite_Volume
|
0.09
|
1.364
|
0.07
|
0.95
|
|
II_ExpertCall_Volume
|
-0.09
|
0.360
|
-0.25
|
0.80
|
|
II_Service_Volume
|
0.56
|
0.715
|
0.79
|
0.43
|
|
AccountSegmentRegional Private
|
1.64
|
1.357
|
1.20
|
0.23
|
|
AccountSegmentRegional Public
|
-2.20
|
1.247
|
-1.77
|
0.08
|
|
AccountSegmentSelective
|
-0.94
|
1.153
|
-0.82
|
0.41
|