If you want to get ‘political’ in the world of Sales Operations, the easiest way to start a fight is to ask this age old question:
“Does it make sense to use Leads instead of Accounts/Contacts in the modern B2B SaaS context?”
In this article, we’ll share learnings/quotes from interviews with 32 sales/ops leaders (including the original product owners at Salesforce!) to help you develop a strong opinion of how to answer this question, so you can find the right data model for your organization.
Why Even Have This Debate?
A core RevOps mandate is to build the infrastructure needed to help the company understand its revenue funnel. Doing this requires 1) a good database structure that allows for complex and efficient queries and 2) good data to ensure those queries yield actionable results.
The division between Leads and Contacts prevents RevOps teams from succeeding at both of these objectives.
Here’s a scenario that explains why… imagine:
- A sales rep (“Rachel”) creates three leads representing sales directors at ACME Corp
- After reaching out to all three, DirectorA doesn’t respond at all, DirectorB responds ‘not interested’, and DirectorC says ‘I’m interested, let’s book a meeting’.
- At this point, Rachel converts DirectorC into an Account/Contact/Opportunity. She either forgets or chooses not to convert DirectorA and DirectorB because it takes too much time with no clear benefit for her.
- Now, when Rachel’s sales manager looks at the ACME Corp account, he has no context on the other outreach that has occurred to the other Directors, who surely will be key influencers in the deal. And so he gives Rachel the wrong coaching on the deal and the Opportunity is marked closed/lost.
- Rachel changes roles at the company and a new rep Mark inherits the Account.
- Mark takes a look at the account and has no knowledge of DirectorA and DirectorB. He adds both contacts to the Account. Now we have multiple problems:
- Mark doesn’t know information about DirectorA and DirectorB
- There is now duplicate data representing Director A and DirectorB in CRM (meaning, the CRM doesn’t know there is a relationship between the Leads and the Contacts created).
- The new contacts don’t have the full history of all outreach to DirectorA and DirectorB
Imagine repeating this messy cycle 10,000 times over two years and you start to get some key issues that go beyond just a bad experience for the prospect (and a lower win rate for the rep).
Ex: You buy a marketing automation software like Marketo, which defaults to picking up the Lead and starts syncing data to that object, while your reps’ activity tracking software syncs their emails to the Contact.
Now you have data going to different places and your funnel modeling/reporting starts to become more and more inaccurate. Your VP Marketing/Sales start arguing about attribution, lead quality, etc. And by the time you’ve figured out what’s happening, you now have bad data to clean up (a massive undertaking). To get around this problem, you end up having to buy a data warehouse and hire people to run an ETL out of Salesforce so you can aggregate and join the data there. And then you start hitting performance bottlenecks, etc etc etc.
All of this because of a lack of a unified data model of a ‘person’ in the CRM to begin with.
Why Do Leads Exist in the First Place?
To understand the context of ‘Leads’ as a concept, we have to take a stroll down memory lane to the Glengarry Glen Ross days – before Linkedin Sales Navigator, Data Exchanges or similar concepts existed.
Lars Nilsson (former VP Sales @ Cloudera) paints a really good picture of what life was like to be a rep during this time:
When I started as a rep for Xerox in the 80’s, I had a pair of sturdy penny loafers and walked up and down PCH in SoCal walking into businesses and prospected by getting past the front office clerk and making my way to back the copier room to check out the copier service log. Yellow pages were my back up. Would take me 6 months to blanket my territory and index and log the office managers that likely would make buying decisions.
In other words, there was so little access to prospect data that the need for you to build an ‘Account Based View’ for a prospect was almost zero. You were lucky enough to get one person to stay in touch with, so problems like ‘we forgot to convert all the different Leads into the Account after the opportunity was created’ just didn’t exist.
At the same time, Leads offered (and continue to offer) some benefits:
- Data Quarantine: since leads are inherently quite messy (you get different bits of information over time via different marketing programs), marketing ops people like the idea of having an object that has less validation rules to begin with. The idea is that once a lead turns into an account, you can enforce a higher standard of validation rules on the Contact object, since a lot of data has already been captured on the account AND you have a rep who can research to fill in the blanks. Having a different object called a ‘Lead’ is a shortcut to creating different validation rules.
- Clear Ownership: it’s a really simple construct for most to understand that Leads are the domain of Marketing and Accounts are the domain of Sales. In a siloed go to market team, this gives each team operating flexibility to customize their objects and reporting as they like without having to care about the other.
- Backwards Compatibility: this is the one that most people don’t really talk about, and which is best explained from Brett Queener- one of the first product owners at Salesforce:
Some funny history as i ran products for a while. Salesforce copied the siebel data model to start.This leads / contacts model is a mess. Leads should have just been a people object that you market to. And if those people respond to a marketing offer or knocking you door those actions should have been one of many responses tied to that person. Each response in themself should be things you consider following up on and then decide if its a marketing qualified lead that sales should follow up and convert to an opportunity.By the time i looked at refactoring it there were already so many email marketing / marketing automation partners who had architected this solution around this bad set up that it would really messed them all up.Also at that time, salesforce having a marketing cloud was just a glimmer at best in someones eyes.So sorry team – we should have bit the bullet to fix it a decade ago but didnt.
So essentially, a HUGE part that this debate even exists is that someone at Salesforce 25 years ago made a decision (which they now regret) that they couldn’t roll back.
And interestingly, the default choice of modern marketing automation software like Hubspot is to eliminate the concept of a lead completely.
Do these concepts still apply in 2020?
We believe that the concepts that made Leads ‘work’ twenty years ago no longer apply today in B2B SaaS. The average number of DMs in sales motions are now 2+ and the number of influencers only keep growing with ‘land and expand’ strategies bringing more front line people into deals.
Now, there are still some important logistical challenges that companies must consider when deciding to kill leads altogether – specifically addressing data quarantine, ownership and backwards compatibility.
On the topic of ‘backwards compatibility’ – the number of marketing automation software that works in an Account Based model has hit a tipping point. All of the major vendors like Marketo, Pardot and Hubspot support this. NOTE: not all of them do it WELL, but the ones that started with an Account Based model like Hubspot are leapfrogging their competition and their offerings are good enough to power companies into the 10’s of millions of revenue.
On the topic of ‘object ownership’, the reality today is that the marketing/sales silo is continuing to erode. The trend to merge marketing ops and sales ops into a single function called ‘Revenue Ops’ seems to be unstoppable at this point, as companies move into an infinite go-to-market loop of ‘sell >> upsell >> renew >> cross sell’. As long as customers live in a ‘subscription’ model, the traditional silos of marketing/sales are going to have to keep integrating more and more.
On the topic of ‘data quarantine’ – we believe this was always a weak argument because it’s not THAT MUCH easier to apply different data validation models to the same object depending on some state. (eg: you could say ‘validate contacts for conditions ABC if the account is in ‘Prospect’ stage and conditions XYZ if it is in the ‘Prospect’ stage). This is largely a data modeling problem and not a technical/logistical issue.
Some Final Thoughts
The problem we continue to see in the marketplace is that the root cause of all these problems is that Salesforce’s core data model fails to give companies and easy way to model their funnels end to end in the Account Based Model (in this article, we discuss why and how to solve it with a one click install package.)
This Leads/Contacts debate is just one more symptom of the larger problem, and we’re excited to help everyone in the RevOps community better understand how they can get so much more out of their sales stack (while saving $20k-200k/year) if they can just ensure they have the right data modeling strategy in their CRM instance.
Are you struggling to get real-time visibility into your sales/marketing funnel? We’d love to talk to you!