Recently we had the pleasure of talking to our (ex) Head of CX, Jonny Davies, about the journey any Customer Experience Manager walks, and how to get insights that power data-driven decisions. Data, data, data…from so many different places. Aggregate, segregate, distill, and there it is…the magical potion of knowledge.

In this short article, you will learn something a little bit different from what you’re used to—a CX approach to your mobile apps, and how to make data in your capable hands sing.

Determining the data-driven approach

So, let’s imagine you've been busy throwing everything into the mix to catch what customers are talking about. Too many channels, too much noise, too few words that make sense. What can be used to gather information from all those chats—from reviews and tweets to phone rambles and support notes—into one big, messy summary to try and make some sense of it?

That's your big puzzle at the moment. Let’s address that first.

Types of data to pull

Everything that you get about your app or business…and at the same time, not all of it.

• App reviews—App Store, Google Play, or any other if you use them
• Support tickets—probably your Zendesk business
• Social media mentions—Insta, Reddit, Quora, wherever you are
• Sales—if you have a team doing that
• Any other channel where someone talks and you’re meant to listen

Normally, it’s a suite of SaaS tools helping you out and a whole lot of time you spend on it. Lately, you’re in luck—all the wonderful AIs out there can crunch the data for you and get you some insights if you know how to feed it.

Thus, let’s say you’re kinda using AI to chew through the info dump from everywhere, but it's not all slick yet; you haven’t pulled it together on one screen. Maybe a few spreadsheets, maybe a few templates.

How do you clean the data?

The tech involved

Simple. Lean on AI to skim the cream off the top—picking out the bits that matter from all the customer talk, like reviews or support tickets. Ditch the junk and zero in on what can actually help tweak your products or services. You’re mostly tossing this job over to AI tools, including stuff like ChatGPT, though it’s definitely no walk in the park and takes a ton of effort to get real results out of it.

You're wrestling with a ton of data, trying to funnel the important bits to your product team. You're pulling in everything—Net Promoter Scores, sales chat, and customer helpdesk drama—to spotlight what really ticks off or pleases your customers across every way they touch base with you.

Then, you’re gearing up to get a system that doesn’t just scoop up all these data streams but also scrubs them clean and spits out the gold on top. Once you’ve got those insights, you start slinging them across other teams. The juicy bits from sales calls? Blast them on Slack automatically so the right folks can jump on it, especially if a big client is on about something new.

What we can recommend is that not every little piece of data should get this VIP treatment. Stuff like random support tickets? They get benched until the bi-weekly product team huddle, where you decide if anything's actually worth sweating over. You also lean on tools like Product Board to sort out which customer whines should get priority, based on how big the customer is and how often they’re nagging about the same thing. It’s all about making sense of the chaos.

Recorded calls are a sanity check for what your product team has on their plate. With a mountain of customer wishes waiting in the wings, these calls help make sure you’re building stuff that not only fits the market but also amps up your product’s worth.

Get Outreach for taping sales calls and running email campaigns, which is all cozy with Salesforce (or Pipedrive if you’re into it more). Your support crew can lean on Zendesk.

Chat and swap notes on Slack about customer chit-chat, leads, and deals, although keeping up with the chat flood in a bustling startup scene is a headache. Then there's Amplitude for spying on how users mess around on your platform, with all its fancy heat maps and stats to show you who’s doing what.

Salesforce is your best bet for keeping an eye on customer details, like how much money they’re likely to drop on licenses, especially the newbies. Tailor your offensive based on what Salesforce tells you. Plus, tie Jira into the mix for your product and engineering teams to help them stay on their toes. Create a weekly Jira dashboard to keep tabs on any customer issues still on the loose after ten days.

That’s the top trick: a list of issues for your devs that must be fixed, with their names on it, and as a public Dashboard. No one is forgetting about that tiny little thing that needed to be fixed a few days ago—not anymore.

The structure to driving action

Make product committee meetings every other week where your go-to-market, product, and engineering teams huddle up. Comb through the big insights pulled from all over the place to connect the dots and speed things up.

It's serious business focused on tweaking your processes. Also, start a revenue committee that's all about pumping up your products.

While you're hustling to improve how you handle feedback from every which way, aim for solid improvements by the quarter’s end.

As an option, when it comes to addressing issues, sort your customers into big fish (XL) and not-so-big fish (L, M, S, whichever) in Salesforce, and tie it into Zendesk. When someone kicks up a fuss with a ticket, it checks who they are in Salesforce and then sends the hot mess over to Jira to keep an eye on any big problems. This setup will keep things transparent and on track, which you obviously need if you want to keep your customers around and happy.

When it comes to handling your customers, see their usage stats to try and catch any that might bail before they actually do. A good start would be a health score for each account based on how much they’re using and what they’ve paid for—like licenses, subscriptions, etc. If someone’s barely touching their stuff and ignoring key features, slap a 'red' tag on them because they’re just not into you. But if they’re maxing out their plan, you flag them as 'green' because they might need to level up.

Your churn score isn’t just some boring number either. You can color-code everything—green for chill, yellow for watch out, and red for danger zone. This makes it super easy for your team to see who might be at risk of leaving. Say someone’s only using 30% of their apps, come renewal time, they’re probably thinking twice about paying for the whole shebang, which could mean they bail or drop a few services.

North Star Metrics

The ultimate!

These focus on how much your customers poke around and use your stuff, which helps you tighten up your keep-‘em-happy strategies. Figure out what’s yours—is it LTV, the best-selling feature sold, a powerup that someone keeps buying every week? This is what you aim for.

When it comes to tracking who's really getting into your features, you label them: 'active users' hit it once a week, 'power users' are on three times a week, and 'NSM users' are your everyday die-hards. Plus, count the days they’re active in a month to really get the full picture.

All this info, from how much they use to how often they log in, helps you figure out a health score that tells you how likely they are to stick around. This score is big news for you—something you can use!

And use you shall—with all the feedback that comes from this kind of client, from your power user—you grab every single piece, scrutinize, analyze, and bring it to the Product Team on a silver platter. That’s the recipe for success.

Afteword

To wrap it up, your churn risk indicators like usage and how chatty your customers are with both manual and AI responses are something you should focus on. They link your product-making and market-storming teams, helping you sell it well. This teamwork is what's pushing your numbers up. Make a solid dent in your market—listen, digest, and act. Good luck!

FAQ

How do you collect client feedback from multiple sources?

Aggregate data from app reviews, social media, support tickets, and sales communications. That will help to centralize feedback into a single platform. AI tools like ChatGPT will help you get the summaries you need.

What role does AI play in processing client feedback?

AI tools process client feedback by filtering out irrelevant information. These insights can then be used to prioritize product updates based on real customer inputs.

How should client feedback influence a product roadmap?

Regular product committee meetings should integrate user feedback to align product developments with customer needs.