How to Align Sales and Marketing in Enterprise Companies
Highlights from this Episode
Highlights from this episode
Allison Smith Terrey, the former Vice-President of Global Marketing Operation at Pitney Bowes discusses funnel strategies. She also offers great advice to those looking to align sales and marketing. She also talks about the future impact of machine learning (ML) and artificial intelligence (AI). Allison comments on how and why data should be handled in marketing department to optimize selling opportunities.
Key points from this episode:
- How Allison uses the B2B funnel to drive revenue performance.
- Improving the conversion of marketing qualified leads (MQLs) to sales-accepted leads (SALs) at Pitney Bowes.
- Allison’s advice to someone trying to unify sales and marketing teams.
- The usefulness of storing data to create baselines and trends that give funnel measurement context.
- Allison’s projections for the impact of ML and AI on the B2B marketer.
- Scoring models and how they might change with machine learning.
“These things like drops in in conversion rates, I think they’re good flags for where your conversation needs to go.” — @asmithterrey [0:07:28.8]
“Then I usually just start to dive in the data. It’s really pretty fun and exciting what you can see and you learn together.” — @asmithterrey [0:11:46.9]
Links mentioned in today’s episode:
Allison Smith Terrey on Twitter — https://twitter.com/asmithterrey
Allison Smith Terrey on Linkedin — https://www.linkedin.com/in/allison-smith-terrey
Pitney Bowes – https://www.pitneybowes.com/us
SalesForce — https://login.salesforce.com/?locale=eu
AT&T — https://www.att.com/
[0:00:07.2] Announcer: Welcome to the B2B Marketing Jukebox by LeadCrunch. Help us start a movement to make B2B marketers the maestros of shareholder value. On our website, LeadCrunch.com, you can find timestamped transcripts of these podcasts and info about the guests. Subscribe to these podcasts on all major platforms, like iTunes. Send topic or guest suggestions to the host at email@example.com.
[0:00:45.6] AST: Hi, thanks. It’s great to be here.
[0:00:47.4] DG: Tell us a little bit about the background you have in marketing operations, which is usually right in the middle of all this funnel measurement stuff.
[0:00:54.8] AST: I come to marketing ops with a background of engineering and business. It’s enabled me to integrate strategy, technology metrics, see the big picture and see how everything integrates from that analysis technology perspective. I have been in marketing operations for 15 years I started early. I don’t think I realized I was in marketing operations at the time. We were just starting to create all these new reports for business marketing, understanding the funnel, understanding conversion metrics, marketing automation, all that. I have worked for several companies in this role, created several marketing operations departments.
[0:01:40.4] DG: Allison, tell us one big success you’ve had applying the discipline of marketing operations to one of the companies you’ve worked for?
[0:01:48.5] AST: Yeah, I think the most recent is that I led the implementation of the salesforce.com marketing cloud, which use all these elements of funnel marketing capabilities, marketing measurement. We were migrating from another platform onto the marketing cloud, enabled us to have a 360 view of the client, to be better aligned with sales and then also to provide and enable marketing with more capabilities for omni-channel marketing.
[0:02:19.0] DG: Alison, let’s talk about the funnel. The B2B funnel is one of those things where if you do it right, it can really make an enormous revenue impact. Talk to the audience about the top of the funnel stages that you use and how you use the funnel to drive revenue performance?
[0:02:37.7] AST: We measured for the pipeline the funnel especially MQL, Marketing Qualified Leads, Sales Accepted Lead, Sales Qualified Lead, which for us was an opportunity creation and then close one deal. Initially, A, you all have to be on the same page marketing, especially all of marketing and sales. What are the data definitions of those? You get that squared away. What we did in marketing operations is we create a report that we ran every month and then every quarter and in between as well, of course, but we would capture the numbers of MQLs, SALs, SQLs and close one deal. We take and we store it. We take a snapshot.
We put out official report of the total numbers of MQL, SAL, etc., and the conversion rate. The conversion rates between MQL, SAL, SAL, SQL to close one deal. We collected this data. I think you need to collect it at least for our business, probably about a year. I’ll talk about how we use it in a second, and you’ll see that you can – usually, you can start to use it and identify some trends after maybe two or three quarters. After a year, you get a good solid set of data.
Then for conversion rates, what we did is we would create graphs and we would for each one of the conversion rates and what you start to see over time, if your marketing efforts are successful, your conversion rates should increase over time. If they’re not increasing, or if they’re starting – if they start to dip somewhere in one quarter, that’s a highlight that something’s going on.
The conversion rates really measure process. How is the marketing to sales process working? Is marketing sending sales leads that they’re picking up, and it is sales then progressing them to go through pipeline. If that’s not happening, they will show up in your conversion rates. What we would do is when we would see either trends flattening out and we thought they should still be climbing, or if they started to decline, that’s where we would focus our energy.
For example, we saw that our SAL, sales opportunity lead to sales qualify lead conversion rate was falling. Now some marketing might say that’s in sales territory, but what we saw – we work with sales ops on that with demand gen, and really what that told us when we peeled back the layers was that the sales accepted numbers were too high. The inside teams were accepting everything and the quality of lead wasn’t good enough to convert to an opportunity.
Meanwhile, marketing thinking that the leads are getting accepted, great, but they’re not being converted to opportunities. This dip in the conversion rate over time forced us to look at that, and that’s when we started to say, “Okay, we’ve got a marketing issue here. We’ve got to increase the scoring model, make it harder to become an MQL, and then we also have a sales training issue.”
But it’s that dip. Those dip in the conversion rates that I think are flags to process problems, before they get to be huge problems that results in decline in revenue or some other, something else blowing up.
I’ve used them a lot in my career. That’s one example, but we’ve also been able to flag really great things when some conversion rates are going up, because what we would do is we would look at the conversion rate by campaign, or we look at it by industry. We dive into the data a little bit. If we see something going up in one area we say, “Hey, what’s going on?” Sometimes it was a great white paper, or it was a great podcast and we try to replicate that in other areas to see if we can leverage that trend in other areas as well.
[0:06:53.0] DG: Interesting. That conversion from an MQL to sales accepted lead I think is often a pain point. The thing it sounds like that was happening at Pitney Bowes is the sales organization was not being as diligent about what it really meant to be accepted. How did you determine that and then how did the sales organization fix that problem, if you could share a little bit about that?
[0:07:20.2] AST: Sure. What we did is again, really to a conversation. These things like drops in in conversion rates, I think they’re good flags for where your conversation needs to go. We have the conversation with sales and sales operations as we’re seeing this decline. We’re meeting our goals, but these don’t seem to be moving forward. Working on [inaudible] cause analysis, asking those five whys.
And talking to enough sales reps and then determining that the leads coming over, they just weren’t the – they were too early in a buying cycle, they might have been the wrong industry that they were looking at. It is through those conversations that we were able to identify the problem, or the issues.
These are scoring models that we had work on with sales. I think over time as strategies change and businesses change that the – you need to address your scoring models, and if team members change. That showed that it was time to go back. We work with sales again to tighten the model up to make sure it met their criteria, made it harder to become a marketing qualified lead. Also, marketing then started to use different elements to help them better identify who they should be turning over to sales as well.
[0:08:45.4] DG: If you were advising someone and there wasn’t this unified sales and marketing measurement system that you alluded to and that you’ve leveraged, what would be your high-level advice for giving them a little bit of a playbook on how they get that off the ground and operationalized?
[0:09:08.3] AST: I think it’s relatively easy to get it off the ground, because there’s a lot of data. First off, there’s a couple of things you need to do. You need to understand, if you’re really just starting from scratch, you need to understand in your systems or your data what’s a lead, how do you identify a sales accepted lead, is it a lead status? Things along that nature. Get really good definitions for MQL, SAL, SQL and opportunity. You need to understand from the data perspective and then also from a sales perspective. What does sales want in a marketing qualified lead? Is that a title? Is it band? Is it a region? Whatever.
You need to get those definitions together; the data definitions and the qualitative definitions of what an MQL, SAL opportunity and close one deal. On the sales side, that usually if they have a sales methodology probably aligns with that.
Then you get a good analyst, or somebody who understands the data depending on the size of the company and you go into the data and you start to create the reports that filter out your metrics. MQL, SAL, SQL and close one deal.
Then you would start to just run the reports and collect the data, take snapshots of the data. I would recommend once a month if you’re just starting, but you would want to probably aggregate it by the quarter. Then start sharing the information with marketing, making sure that everybody understand the information and the data and structuring it with sales – sales operations, I think. Unless, if you don’t have sales operations, then find a good sales manager who will work with marketing.
Then I would recommend setting up regular meetings with sales, or sales operations. If you have sales operations that’s probably best. We met in the beginning every other week, over time we moved to monthly, but really started to work together to create the definitions and the reports and then we started to look at them together over time. I would also just want to throw one little note in, as you get better, and as you get more information and get more data, I tended to look at the data on what I would call rolling four quarters.
Look at four quarters of data at a time. Because that take out seasonality if you have a seasonal business. That wreaks a little bit of havoc. You know, It’s really not that hard, but that’s what I would do.
Then I usually just start to dive in the data. It’s really pretty fun and exciting what you can see and you learn together. You learn together to identify, to look at the trends, look at the conversion rates, look at them over time, see them change. Say, “Holy cow. What’s going on there? Let’s go find out.”
[0:12:04.1] DG: You mentioned one thing and just maybe unpack that a little bit for the audience. That is start to store the data monthly and quarterly, so that you can go back and reference it over time to see to what degree the conversion rates and things like that are changing. Can you talk a little bit about that from a tools and the reason why you want to be able to do that and you need to do that and the system won’t just do it retroactively?
[0:12:28.5] AST: Yeah. Actually that’s a good point. Yeah, so some systems, especially with lead data, an opportunity data as they progress, the record itself changes and you can’t get history, you can’t get a snapshot. What we would do is we would – once we had our reports created, we would run the report. We had a couple of analysts on our team who would do this. We outsource some of this also, but they would run a report and they would save the data. We saved it.
If you don’t have a big system, you can save it in an Excel file and that’s what I started with, and then every quarter just take a snapshot and then add it and I don’t know the column for your next quarter if you’re going to do it monthly, just keep adding the data over time. Or if you have a system. If you have – we had power BI. There were different systems that an analyst might use to save the data in their system as well and save it in a database. Excel works just as good as anything, especially as you’re starting.
Then you have the data there and you can just – what I would do in the beginning, or I had my graphs and I would just add on the next quarter and I didn’t have to recreate the wheel every time. I just add on the quarter and look at the trends. As marketing operations professional, I would look at the trends first and try to identify if there were some key elements, or key trends to take that to highlight to marketing. I think that helps start the conversation from analyst to do a good review first and provide some insight. Excel is as good as anything in the beginning. Yeah.
[0:14:11.3] DG: Absolutely. I read your excellent blog post about the looming impact of artificial intelligence on this whole marketing funnel that you’ve been talking about. Could you talk a little bit about where you see the impact coming with use of things like machine learning and artificial intelligence for the B2B marketer and how that – you see that that changing things over time?
[0:14:36.0] AST: Sure. Sure. I think there’s a couple of key areas; I think first of all, finding the lead when a lead is qualified to go to sales. I know there are solutions out there now. Machine learning, artificial intelligence can really have a big impact there. As these algorithms learn what are the leads that sales took converts opportunity and then progress through the pipeline. What are the elements of those records, of those accounts, you’re going to be able to look at our marketing data and even back probably to the prospect data and the web data which I didn’t talk about now. They’re going to be able to identify lookalikes and say, “Hey, you know what? I think this is someone that might want to get over to sales more quickly, because these are the types of accounts that have progressed quickly through the pipeline.”
I think that’s going to be a great tool for sales and for marketing and that will hopefully expedite pipeline progression. I also believe that AI will help marketing better understand the mix that is well, I suppose it’s the same – along the same line as the scoring model and what to turn over to sales. It’s going to help marketing understand the mix. What mix of activities is most likely somebody does these things, they’re most likely to become a really good opportunity for sales? That’s very hard to get at right now.
Try it in a month, a lot of different ways with my analysts and data scientists and the tool, without machine learning I think it’s pretty tough. I think that’s going to enable marketing to understand really what catches people attention and then how to convert them to pipeline. I also think that AI and machine learning will help with what’s the best way to score leads and how to identify marketing’s true contribution to revenue. I’m sure we leave a lot on the floor, because we can’t measure it easily right now.
[0:16:40.2] DG: All the marketing automation platforms out there have built in lead scoring systems. What is it that’s going to be different about the way you’re describing the use of machine learning to score the leads, versus how it happens inside of most marketing automation platforms today?
[0:16:55.7] AST: A lot of the scoring models, the weight of that, most of the companies are using them today. They’re scoring based on attributes on a record, a title, an industry, a business need, whether they have budget, whatever it is for that company. It’s static in time and it is based on what we know as people, as marketers and sales. Where AI comes in is it will look at maybe other elements that we aren’t even thinking about. You might be able to – I know some of the systems are integrating with social data and be able to look out at people’s activities, social media, is there intent to look at other systems? Is there intent to buy, they’ll be able to integrate that into the model? Can be able to say, “Hey, this person, their intent is high. We’re going to we’re going to raise their score even higher.”
I think being able to identify the mix of marketing activities and sales activities that a person is doing that would – as much as different algorithm to look through all the data, identify those mixes that are most likely to result in at least an opportunity, if not a deal, and then highlight those for marketing and for sales. That’s not something we can do today, or certainly not anything that can be done easily. I believe AI is going to give that extra edge into looking at intent, at activity that enables marketing to score someone higher. I believe that, AI or whatever the algorithm is, that’ll kick – that’ll be part of the scoring model, and it’ll enable marketing.
Just as importantly too, there might be some leads that marketing are sending over right now. At the algorithm will say, “Not yet. Let’s wait on that one,” because historically, we’ve seen people with these attributes really aren’t quite ready. I think that AI is going to give us better information, things that we can’t do today because of the strength of their algorithms and their learning, the whole learning nature of it.
[0:19:02.6] DG: Right, right. Yeah, I think my experience with the lead scoring is it’s our intuition, right, that we give an open or click a certain value, or a title of somebody, as opposed to looking at it more mathematically and just objectively and saying what correlation relieves us between these various activities and closed one business.
[0:19:26.3] AST: Yeah, and that deed is all over the place. It’s not just in the marketing systems, so you’d be able to combine the sales data and the web data in the marketing automation data and get a much better picture.
[0:19:37.6] DG: Alison Smith Terry, Thank you so much for sharing insights today about best practices in the really exciting and I think important field of marketing operations. Thank you again.
[0:19:47.6] Announcer: Thank you for listening the B2B Marketing Jukebox by LeadCrunch. On our website, LeadCrunch.com, you can find timestamped transcripts and info about the guests. You can send topics or guest suggestions to the host at firstname.lastname@example.org. Subscribe to these podcasts on all the major platforms, like iTunes.
Thanks for joining us, and we wish you happy, data-driven growth.