How AI will change PLG forever?? ??
Hey Folks!?
I know, I know. You're probably dead-sick of even hearing the word "AI" by now. So am I. I'm waking up every single day to?learn I now need to read up on Bard, then Bing, and feeling guilty of not using ChatGPT nearly enough, wondering if I've boomered prematurely.?
But while we?might be intimidated by this revolution, I can't help but?see?how?it's going to transform PLG completely?(and hopefully?finally?heal my rocky relationship with SaaS products...)?by allowing product managers & marketers to understand and?respond to users' behavioral cues?in real time, provide?100% personalized in-app support?also in real time,?and?work on various?behavioral KPIs on autopilot.
"It's not that we don't have enough data. We don't have?enough?insights."?
This is my standard reply to cold emails pitching me?yet another?data analytics tool. You're solving the wrong problem, bro. Both Product Managers and Marketers that use any analytics tools have in fact?plenty of data?at their fingertips. The crux of the issue is - you have to be able to use the data.??
With the current state of affairs, very few companies can?actually?leverage?Product-Led Growth. Most just open a free trial/ freemium account and hope for the best.?
?
How can AI change PLG in the next 5 years??
1. Identifying hot?PQLs on autopilot to improve trial-to-paid conversion rates?
Today, in order to identify your hottest Product-Qualified Leads in your free trials/freemium accounts and work on them, you need to analyze tons of data, manually create segments, and then build tons of automation workflows in e.g. CRMs to e.g. notify your sales/ CS folks there's a hot lead that needs to be worked on?OR?helped with something. It's possible, but?very few companies?have the resources to do that:?
You can make some (more-or-less data based) assumptions about the key events a hot PQL performs in your product in a specific time period, funnel the data from?Userpilot?to Hubspot, set up workflows to assign the PQLs to an SDR based on their timezones...
...or you could wait for the Machine Learning algorithm to draw conclusions by reverse-engineering the conversion path of your different person in the past, automatically segment your PQLs by conversion likelihood and different JTBDs, and even suggest the most-effective way of approaching them to push for a sale.
?
2. Dynamic auto-segmentation?
Similarly to what I wrote above - keeping track of *all your users' actions*, making conclusions, and creating?the right segments?is still a nightmare.
I believe that in a few years' time, AI-powered product analytics tools will be able to analyze large volumes of historical user data, and?dynamically create relevant user segments?that?require?individual?approach.
?
3. Trend detection & Predictive Analytics?
What do you see below??
Making sense of event analytics and user trends these days requires so much effort and manual work. Imagine an ML model would do that for you, and just send you notifications on worrying trends occurring in your user base, with recommendations on how to counteract them??
领英推荐
In a few years' time,?AI may be used to determine?the correlation between different user actions and various outcomes (both for the user, and the business) - both positive and negative.?
And based on that - it will be able to predict future events for individual accounts, users or whole user segments, and help you take action?before?they happen.
4.?Happy Path Redirects?
Since we'll understand and segment each user accurately, as well as use predictive analytics to determine their "happy path" to the most successful outcomes, we will be able to react instantly (with in-app experiences, e.g.?reactive tooltips) when they wander off that path.?
Imagine?GPT being able to write microcopy in reactive tooltips in real time????- so product marketers don't have to predict and write for all the possible user action scenarios (which we all know is impossible...)?
It will be like having a co-pilot watching you use the product in-real time, and giving you ultra-personalized prompts based on your goals and current behavior so far.?
?
5. GPT Chat-powered Self-Serve Support?
Needless to say, AI will also be able to provide?specific?answers to user questions (based on the information provided in the Resource Center, your help docs, and billions of user events) in real time - and even trigger the relevant in-app experiences based on that.?
“The race starts today, and we’re going to move and move fast”?
This is what Microsoft CEO Satya Nadella said at Microsoft's last conference on Tuesday, announcing the new ChatGPT-powered Bing search.?
The race starts today for you?as well.?
Will your product adapt to this revolution, and use the latest tech to serve its users' needs better, or remain a?product-boomer?
Leaving you with this thought???
See you all next week!
PM @Guvi | building Watermelon ??
1 年Love the word Product-boomer ?? Can I borrow it, please?
Emilia Korczynska Thanks for Sharing! ?