Product-led growth is a continuum, not a category

It is a continuum between 0 and 100% self-serve.

When we ask questions like, “Should we do product-led growth (PLG) or not?” or “Do we allow self-serve?”, we are only considering two extremes in a continuum of intermediate possibilities.

As with most things in life, there is a messy middle. We can't just build a product, turn on marketing, and sit back and let it sell itself.

There is a continuum between 0 and 100% self-serve.

It isn’t a yes or no question, it is a matter-of-when question:

• When do we intervene in the prospect's journey?

Well, the research says that prospects spend about 5-17% of their time with salespeople during their buying journey. (mostly spent researching independently)

Physical stores face a similar conundrum — salespeople look for signals that someone needs help and we've all had the shopping experience where an employee repeatedly asks us if we need help, to which we respond with, "no, I'm just looking."

So…when do we intervene?

It depends. It depends on the product, the prospect, and how easy our UX makes it to reach value. 

For example, industry research (below) has shown that the higher a prospect’s annual contract value (ACV), the more likely they will need help converting. Whereas, the lower the ACV the less likely they will need help converting.

Referenced from Product-led Onboarding book

Mostly, it’s because high ACV accounts come with more bureaucracy and red tape. But, even if the ACV is high, there’s still an opportunity to enable first-time-to-value.

When prospects are researching independently, which is 83-95% of the time according to Gartner, wouldn’t it be nice if they were experiencing value with your products?

This is PLG. It can be simplified down to one principle: enable time-to-value, on the prospect's terms.

Adhering to this principle requires a strong sales motion and onboarding experience.

Here’s how leading PLG companies are doing it: 

1. Product-led onboarding (PLO) (see the book)

   &

2. Product-led sales (PLS) (see the playbook)

Product-led onboarding is the name of the second book from ProductLed.com. Presumably, after launching their first book "Product-Led Growth" in May 2019, and after consulting with hundreds of different SaaS companies, they quickly learned that product-led growth was a continuum.

And in this continuum, they learned that no matter what % is self-serve, the prospect goes through a journey of value moments:

• They will perceive value → experience value → and adopt value

Getting prospects through these value moments and understanding why, if they have not reached value, is the key for PLG companies. Here’s how this may look:

Modified from the original source: Chapter 1 of the Product-led onboarding book

There are two takeaways from this visual:

1. The journey starts with a first impression and requires strong cross-functional collaboration to reach value (PLO)

2. The journey needs strong analytics & data instrumentation so you know when to intervene (PLS)

The time it takes a prospect to get through these value moments, and how much of it requires intervention, depends highly on your UX and your sales team having visibility into the data.

As mentioned, this is being solved in two ways, PLO & PLS.

For a deep dive into PLO, please check out the book, but here are a few takeaways:

• Product-led onboarding starts with the first impression and ends with a customer sharing your product. This means you need to include your website and marketing campaigns in the experience.

  • Also, strong first impressions lead to strong retention, so make it a priority.

• Don’t forget sales in the onboarding process. For example, Slack increased conversion by 3.5X by getting sales more involved in enterprise accounts.

The second strategy is PLS. For a deep dive, check out the playbook, but here are a few takeaways:

• "What's in it for me?", said the prospect

  • Too often, sales reps reach out to a prospect with information about themselves — their products, marketing #’s, or an assumption about the prospect's needs.

  • Instead, they should understand intent, and lead with value.

  • To do this, sales reps need actionable insights about their prospects.

• To accomplish this, you need a different infrastructure, built for PLG.

  • PLG companies come with large volumes of data and different data models. Traditional infrastructures and tooling have trouble with this.

  • This requires the power of a cloud data warehouse, data engineering, tooling, and a flexible way to surface actionable insights up to sales teams.

• To enable actionable insights, you need the ability to interject product data into your sales team's workflows.

  • Commonly referred to as PQLs (product qualified leads), a tool and/or team needs to sit at the cross-section of all of GTM -- including Ops, BizApps, and your data team -- to help identify prospects who've shown strong signals.

At Matillion, where I reside as Director of Growth, we are a technical B2B SaaS company, which means we automatically start on the left side of the self-serve continuum.

Which means we need an even stronger PLO & PLS motion.

There are no templates, we are trying to harness the principles above. We have started small, experimented, learned, and iterated. 

What you’ve read so far is a summary of our learnings. To help you put this theory to practice, here is a chronological timeline of how we executed it:

1. In April of 2022:

   1. We quantified our customer journey and analyzed where the biggest points of leverage were. (acquisition and activation)

   2. We began experimenting and learning how Matillion could do PLG by running A/B tests on the website.

   3. We did qualitative research to understand the biggest pain points in our PLO experience.

2. In July of 2022:

   1. Matillion hired Julian Wiffen, a Director of Data Science who sits within the product team.

   2. Then, I stumbled upon Allison Pickens’ Customer Success vs. Sales Assist article which introduced me to  Alexa Grabell's PLS playbook

3. By August of 2022, we had increased monthly PLG sign-ups by 108% but our down-funnel metrics had only gone up by 33%.

   1. We learned that we needed a strong PLO and PLS motion to support PLG.

4. By October 2022, we had announced major innovations that improve time-to-value (example 1, example 2.)

   1. We also launched smaller ones like Navattic's interactive demo experience.

5. In November 2022, we implemented a PLS pilot program with eight sales reps (4 SDRs and 4 AEs).

   1. We did this with a tech stack of: Heap.io, HeadsUp, Matillion ETL, Salesforce, and Snowflake.

   2. Our team included GrowthScience (Data Science + Growth), a data engineer, 4 SDRs/4 AEs, and HeadsUp

      1. The data engineer, Susana Cardoso, was critical for setting up data pipelines and handling the complexity that comes with joining data sources like Heap & Salesforce into Snowflake.

6. By January of 2023, we had evolved into eight product usage playbooks (aka PQLs):

   1. Here are a few examples: 🚰 Abandoned a data pipeline, 💰 Buying Intent, 🎓 Academy, 🔌 Data Transformation

   2. We also implemented a catch-all playbook driven by Data Science and HeadsUp’s machine learning. If a prospect has usage that’s predictive of conversion, and it’s not in one of the eight playbooks, we surface it to reps.

   3. Each of our eight reps received product usage data via Slack alerts and HeadsUp’s browser extension.

   4. The combination of Heap/HeadsUp/Snowflake allowed us to adapt at lightspeed. A business user is now empowered to track an event, sync it to Snowflake, and get it in the hands of a sales rep within hours.

7. As of February 2023:

   1. We have committed to rolling PLS out to the rest of the organization. Why?

      1. Reps couldn't live without product usage data as a part of their workflow and 100% said they would be disappointed if they didn't have access.

      2. What’s better — pilot reps showed a 57% higher conversion to OI (opportunity identified)

   2. We’ve also committed to improving first-time-to-value through:

      1. New innovations like our Data Productivity Cloud

      2. Empowering innovation by implementing full-stack A/B testing in our product via LaunchDarkly's experimentation feature.

This is obviously just a summary, and being a Series E startup, we have the privilege of more resources and funding. The important part, as mentioned before, is to start small, begin experimenting, learning, and iterating. 

PLG is a continuum, we all need to find our place in it.