Mastering CRM Analytics
- 1. Mastering CRM Analytics Email & CRM Marketing Meetup October 11, 2018 SmartData.MarketingExpertise.Connected
- 2. Sharing some of my go-to’s when it comes to CRM analytics SmartData.MarketingExpertise.Connected
- 4. 5 EFFECTIVENESS ENGAGEMENT EFFICIENCY BUSINESS KPIs DIAGNOSTIC METRICS CHANNEL METRICS Market Share Revenue Growth Sales Growth ROI/ROMI Churn Reduction Net Promoter Score Repeat Purchase Adherence Customer Lifetime Value (CLV) Customer satisfaction score Cost per sale Cost per leadCost per response Traffc Sources Keyword Performance Engagement duration First call resolution Lead conversion Basket composition Site visit Bounce rate Call Volume Open Rate Click-through-rate Sentiment # Followers Reach # Opt-in Page per visit Time on site Call disposition Agent quality score OPTIMIZATION OPTIMIZATION OPTIMIZATION Analysis Purposes Understand why you are doing the analysis is key to getting meaningful outcome. 5
- 5. 6 SEMI-ANNUAL / ANNUAL Review channel performance and category research for roadmap planning MONTHLY / QUARTERLY Deep dive into specifc program areas for ongoing optimization WEEKLY / MONTHLY Track KPI and diagnostic metrics from multiple sources for short-term adjustments POST LAUNCH Monitor results of new initiatives to ensure they function as required QUALITY CONTROL INITIATIVE PLANNING BUSINESS PLANNING Analysis Frequency Templated Ad hoc
- 6. 7 Prototype
- 7. 8 Quick Prototypes With features e.g. Google Analytics and API integration, prototypes can be built easily in tools like Google Sheets, Excel.
- 8. 9 Integrated Measurements Reports showing results across channels can provide richer context to CRM performance.
- 9. 10 Audience Preferences Mockups and prototypes are easy ways to gauge if your audience fnds it easy to consume the information presented. Some may prefer no graphical visualization at all!
- 10. 11 Reporting Architecture A centralized data mart may be useful when: ● Data / tactics don’t change frequently ● Team managing the data mart can react quickly to change ● Resources are available to maintain and QA data It often is a hybrid. sources Reports Data martsources Reports vs
- 12. 13 Niche Engaging Noise Open Bait Open Rate Click-to-OpenorClickthroughRate Benchmark or past performance Email Insights MapTM This grid is my favourite diagnostic tool for email performance.
- 13. 14 Niche Engaging Noise Open Bait Open Rate Click-to-OpenorClickthroughRate Benchmark or past performance Email Insights MapTM The 2x2 provides an easy way to spot any trends by type of campaign. Campaigns can be grouped by any features – ofers, language, segments, ... that are coded using colours and shapes.
- 14. 15 Contest No Contest +15% likely -5% likely Likelihood Of Contest Segment To Further Engage Beware of complex visuals However, most audience prefer a simpler interpretation of the results vs the raw diagnostics.
- 15. 16 Benchmark
- 16. 17 Benchmark DistilleryTM Developed in conjunction with RI, this methodology collects and distills benchmarks from diferent sources to produces the most applicable benchmark for comparison. Benchmarks Benchmarks Benchmarks Benchmarks Most applicable benchmarks
- 17. 18 Benchmark DistilleryTM Find diferent benchmarks based on industry, batch vs triggered email, content of emails, geography, target audience, etc. BATCH EMAILS – INFORMATIONAL Min Most Applicable Max Unique Open Rate 8.0% 24.0% 35.3% Unique Clickthrough Rate 1.1% 3.7% 11.1% Source Get Response Global Industry Benchmarks – Retail BATCH EMAILS – ECOMMERCE Min Most Applicable Max Unique Open Rate 8.0% 33.1% 35.3% Unique Clickthrough Rate 1.1% 3.2% 11.1% Source Epsilon 2017 Q1 Specialty Retail – Broadcast
- 18. 19 Open Rate 0% 5% 10% 15% 20% 25% Maximum across benchmarks Minimum across benchmarks Most applicable across benchmarks Benchmark DistilleryTM Can compare results against: ● “most applicable”; or ● min/max range Ultimately, the most important is to improve against our previous!
- 20. 21 Link Email To Other Sources Getting data to track in the appropriate system can be complex. Beware of on-going operational impact to campaign setup. Embed tracking codes in link to pass information to websites http://example.com/page/? cmp=brand:promo:source:email:link:type& utm_source=brand&utm_medium=email& utm_campaign=campaign&utm_content=link Embed tracking codes in link to pass information to websites http://example.com/page/? cmp=brand:promo:source:email:link:type& utm_source=brand&utm_medium=email& utm_campaign=campaign&utm_content=link Embed image pixel to pass information to external servers e.g. Litmus, ad servers, DMP <img src=...> Embed image pixel to pass information to external servers e.g. Litmus, ad servers, DMP <img src=...> Embed additional data for ESP <a href=”link” _label=”Top_CTA”> Embed additional data for ESP <a href=”link” _label=”Top_CTA”>
- 23. 24 Actionable Insights Interesting facts don’t always equate to actionable optimization. Interesting Facts Actionable Insights Those who haven’t interacted with any emails in 8 weeks won’t interact with us at all. ● Reduce / stop emails to reduce CPM. ● Switch users to re- engagement stream. ● Experiment with content in frst 8 weeks to gauge interest.
- 24. 25 Actionable Insights And never forget to keep asking WHY? Interesting Facts Actionable Insights Regardless of recency of purchase, users are equally engaged with new product launch information. ● Continue sending new product launch announcement to all users. ● What does “engagement” mean?
- 26. 27 All A/B test vs Seg A A/B test Seg B A/B test Automated A/B Tests When you use automated A/B test feature, consider how variables that can afect the test results can be taken into consideration by the automation system.
- 27. 28 Email A Email B Continue Drip Campaign A/B Tests A/B tests can be incorporated as part of the design of drip campaigns. This is especially useful when audience size is small.
- 28. 29 Winning Version Winning Version Continue Drip Campaign A/B Tests All test streams can use the same winning version when test is not being run. This preserves the mechanism to launch new tests quickly.
- 30. 31 Accessible Analytics Open source analytics platforms e.g. R & Python, and cheap computing power, makes it possible to do more advanced analysis on your laptop for free.
- 31. Have a favourite CRM analytics tip? Let’s connect: margaret.ngai@connector42.com | @letsfika SmartData.MarketingExpertise.Connected