How are you capturing metrics and analyzing data?

Jeff Breunsbach

One challenge we see in Early Stage SaaS companies is the lack of data to analyze. Tools may not be installed yet and manual processes may be implemented - so how are you capturing metrics in a way that you can analyze and make business decisions? 

Any examples @Jung Kim? 

Dave Epperly

When I first started as a CSM (pre tools), I built a custom CRM/HealthScoring "system" using a combination of Google Sheets and Forms.

Using a specific Form for all customer facing teams (filled out upon each interaction) I could control what information we collected (frequency of interaction, sentiment, issues: via full text meeting notes, etc.). The Form was great for controlling quality of data - but also to see who on the team was using it. Thus, I could focus on buy-in internally which improved quality in a virtuous circle.

The form output to a sheet where I had some simple formulas updating things as forms were completed. Sentiment over time (sparklines), GYR current sentiment analysis, frequency, major issues, etc. all trending and searchable (think you Google). Each team member had their own tab for only their customers/interactions. Full text search here was a dream. This all culminated into an overall customer-set health score which was pretty but not super useful.

No real segmentation or sophisticated automation - KISS

I think this method worked well for a long time. We've since bought tools with more capability and prettier UI but I still miss the ability to find things quickly via powerful text search capabilities, and to have everything so clearly laid out that even new employees could understand the system relatively quickly.

 

 

Jung Kim

Happy to share my super manual / so necessary process! For anyone else reading, @Jeff Breunsbach and I have bonded over the painful and glaring need to have data, especially in that peak of need before any decisions for CS software installations are made. 

In spirit of being respectful of resource allocation and budget - we measure once, measure twice, build/buy once. 

A few key details for our process and what we use: 

• BI data tool! You should be able to have access to customer behavior and your entire team should also be able to have the autonomy to slice and dice data. Which cards/dashboards are the most useful for us? I will note that this manual process and oversight is only available to use because we have access to platform usage data. Without this, not so sure how CS dept would even work  
• 5 key success metrics for our SMBs (Enterprise clients are high-touch all of the time and a different pace, so that's still in progress). KPIs surrounding $ROI (key measure of success for our clients) + onboarding success metrics + behavior/platform utilization + sentiment. Having too many metrics won't help guide you in your decisions. 
• Invoice master list - connect ROI with utilization and your customer scores
• CS & leadership sync

I like the breakdown of product versus relationship (let's face it, churn is rarely CS team's "fault" and is usually a sign of product market fit/best customer fit/product). 

when I first built out CS scores, I found that I really aligned with Rachel Jennings from Assignar and built out a manual score with her baseline template. Really great if you have just your excel sheet and the spirit of hustle. 

https://docs.google.com/spreadsheets/d/1e5czbWQAL_Y631h8cqm9P9xZijG8vxJzhYvUu7k6EDI/edit#gid=429454996

Even with the score above, we got SUPER clear on: 

• What do these metrics mean - aka what are the action items/playbooks once we have the scores?
• What is the cadence that makes the most sense?
• Iterate on the definitions of this customer score every month/quarter

Happy to chat further as always and iterate if anyone else has some favorite tricks of the trade. 

Jeff Breunsbach

This is awesome @Dave Epperly .  Anything that you'd do differently if you were to go back to that point in time? 

Jeff Breunsbach

@Jay Nathan Jung has got the goods on early stage data capture. Thanks @Jung Kim this is good stuff. 

Dave Epperly

Knowing what I know now - I'd highlight those customers that score(d) highest on our ICP and focus on those narratives to drive product feedback in lieu of some of the incredibly noisy customers that may not have been the best fit for us. That would have saved a lot of headache and stress. I suppose that this would go hand in hand with incredibly high confidence in how you're determining ICP.

I'd also push this data more forcefully back to the sales team than we did. Simple stories like those we captured during regular syncs are gold to the sales team.

 

Gautam Sawhney

Hi @Dave Epperly , thank you for sharing. How did your team react to filling out a form after each interaction and was this for the SMB/Mid-Market or Enterprise market? I am looking for a way to create a similar structure with enterprise customers but I want to mindful of how time is spent and where it is spent. I would love to chat more if you have some time in the upcoming few weeks. Thanks!

Dave Epperly

@Gautam Sawhney You're welcome. 

There were definitely some raised eyebrows at first. However, after they begun to utilize it more and simultaneously encounter issues that were solvable with information from this system (easily discoverable) they quickly became converts. Said another way - they were able to quickly see the value once they hit some "real world problems".

We've always been a high-touch shop so this kind of attention to SMB/Mid-Market was not looked down on.

Gabriel Fallas

@Jung Kim Great stuff. Glad I scrolled through some of these older posts!

I've been building out a client health score and thinking of the best way to make it clean and not overloaded with metrics - love the template you shared. I'll definitely incorporate certain elements of it into our new process. 

We currently lack an automated solution so I have specific reports that are pulled from Tableau and Salesforce, and through that create the Health Score. 

 

Out of curiosity, how did the transition from spreadsheet to an automated cloud system play out for your team and how long did it take? Did you adjust or incorporate any metrics when you migrated?

 

 

Jung Kim

I'll let you know how it goes! Got some cold feet after doing some networking/stakeholder interviews with some of the software that we were targeting. Doing some manual transitioning to the free version of Totango just to see if some of the elements of utilizing the platform would help justify some of the cost in addition to the conversations that are already in place with C-level. 

 

The team is already feeling the pain of manual upkeep (even with the great support that we have from our data science team) - it's a matter of priority and just putting out which fires right now (not just putting out fires, lol). I definitely have a plan to adjust some of the metrics per segments - I was hoping that I wouldn't have to do it manually, to be honest. For example: 

1. successful clients segment - what do they have in common? What are the aspects of these clients that we can double down on and offer to other clients?
2. size/specialty - getting a bit more specific to the variety of customers that we offer, especially for tier 1 and tier 2 
3. advocates/referrals - clients that are successful and are always a 5/5 for sentiment; what are the elements of adoption that we can bring here as well? 
4. high CLV - what are the extra renewal steps/processes that we can apply? 

Then some other metrics from there that I'd like to measure: 

• rate of billing (behavior for us that connects to ROI) and trends
• patient engagement (the C portion for our B2B2C company) and its relationship to other success metrics

Don't want to go on and on - best of luck to you as you lay down your roots!

 

Ronald Krisak

@Jung Kim Appreciate you sharing your health score template.  The more I read & research all things CS, I keep adding more & more metrics to our plate.  A bit of metric overload.  Your example does a nice job streamlining some of the health score elements I have been evaluating.  Thanks again.