This week, we take a look at my interview with Noam Haberfeld, Sales Strategy & Operations at Amplitude. From avoiding costly mistakes when setting up Salesforce to calling BS on unlikely deals, Noam's approach to sales ops is worth reading about. After chatting about Noam's strong background in systems and processes, the interview began...
Rory Brown (RB): Can you tell us about you background and your career to date.
Noam Haberfeld (NH): My background and career space is with the systems and process piece of sales ops. I started my career early with Salesforce implementation, working for a small European consultant partner. During the first couple of years we focused on driving these three-day quick initiations for clients who just purchased Salesforce but needed some assistance.
Those first couple of years really gave me a pretty wide lensinto different industries, different use cases, project management, how tolaunch Salesforce and what to look for. That was the foundation of me gettinginto sales ops, because you’re off to a good start if you have a strong commandof systems and processes.
Later on, I joined a company called App Annie where I gotmore involved with territories, quota, MQL, SQL, com plans, forecasting, etc.
RB: What advice wouldyou give to a sales ops person who has landed at a well-working growth businesswhere processes are probably either non-existent or a little bit all over theplace? Where do you start to get yourself somewhere close towards a cleanslate?
NH: You start by interviewing sales managers, sales reps anddifferent colleagues in different departments to uncover where the gaps are. Youtake both a top down and bottoms up approach. Top down would be C-level: CRO, VPSales, CEO. Ask them, “What is your mission statement for sales? What do youwant to see from the revenue org as a whole?”
Those become your marching orders. And then you start fromthe bottom up. You interview the field to find out what is working and notworking. As you go along you find places to bridge those gaps, places to automate,places to improve. Although that’s easier said than done! That exercise could take6 - 9 months but there is no one template that fits everyone.
You need to be open-minded and creative. Anyone joining anorganisation comes with their bag of tricks and ideas of what they want to do butit might not necessarily apply or be relevant. Be cautious, but also be creative.Start small. I would never start at anew place and immediately do a revamp or introduce CPQ.
RB: What would be acouple of examples of quick wins?
NH: Any small company will not have approval process andorder for launch good to go so that’s a quick one. Order form automation is anotherquick win, as well as lead assignments, whether that’s round robin based orterritory based for example.
Removing, or automating, the deal closure process is usuallya second step. But any small SMB or even big SMB will have a million steps toclose an opportunity and it’s usually extremely manual. So just streamlining thatafter you’ve been there for 2-3 months is a huge win.
RB: Really good. Whatare the key things to consider when building sales processes?
NH: Two things. Firstly, build to scale. A lot of thesecompanies will change every 3-4 months. So it is extremely important when youlay that foundation that you build something that is dynamic and scalable. Dynamicbeing no hard coding, no creating process just to create process. The wholeidea is to enable your salespeople, helping them spend less time on process.
When you build and adapt your sales process, always ask yourselfwhether anything would break if 50 new people joined tomorrow.
Secondly, if this is going to change at least three times inthe next year, when you build it make sure that nothing is holding you back.
It will literally be a definition change, or field change andnot a complete process change that will involve four other departments anddevelopment. Those are the two mainthings to keep in mind.
Also it doesn’t change because we did something wrong, itchanges because we are doing well, and we are trying, and we are going into newterritories and we’re hiring a different type of people, and the messaging willchange – and it is a good thing that it is changing.
But we from the back end – from the systems and process standpoint– need to be prepared for that. We don’t want to be the ones holding this wholething back.
RB: Nice. So dynamicand scalable for anything that you build.
NH: Exactly. Those are the two most important things I wouldsay.
RB: What is a sensibleplace to start with setting up Salesforce (or other CRM) from a lead toopportunity perspective?
NH: I think the one big costly mistake is when you startcustomising too heavily too soon. I’ve seen companies that don’t use leads oropportunities and create all custom. But that level of customisation can catchup with you. For as long as you can, stick to the standard objects and processes.There is so much logic, code, building, UI and UX behind the standard objectsand processes, that if you start customising early on it will catch up and noone wants to do a ‘rip and replace’ just before you go hiring 30 people. Sothat is the one very big pitfall. I stick to the standards.
When it comes to leads, another important point is therelationship between marketing automation (Marketo or Hubspot for example) andsales.
There is a tendency for sales ops people to have everythingin our control and put everything in the system. But over the years I’ve learntthat it is not necessarily the right thing to do. Having a very distinct definitionof ownership, of what a marketing qualified lead and what a sales qualifiedlead is, is very important. It sets the boundaries between marketing and sales,holds the different teams accountable, and also sets a ‘line in the sand’ forsystems.
What happens early on is that sales want all leads in theCRM and then later on when marketing starts building campaigns and nurturingprogrammes, most of the leads that they are working on have already beentouched, with a negative response.
You need to have a clear definition of an MQL and only move leadsinto the CRM when they meet that MQL criteria because we also don’t want ourISRs or SDR team to do a lot of the investigation. I don’t want them to go andspend a lot of their time on just combing through leads and combing throughwebsites. I want them to talk to leads who already know who we are.
Opportunities are the core of your database. As much as youcan, build it so everything is reportable and trackable. And keep it simple. Becauseover-complicating the opportunity object is what will get your repsdisengaged. And I think working with methodologieslike MEDDIC or BANT is really useful.
Salesforce should be your system of truth, your North Star. Youneed to be able to rely on Salesforce for anything and everything.
RB: In a high-growthcompanies, processes can be a bit scrappy. It’s all about getting deals acrossthe line in any way possible. With that mentality, what is the best way to getpeople to see value in having a process like Salesforce?
NH: A great point. You always have the 20-year veterans thathave been around since before Salesforce and it will be very hard to get themgoing. You have to explain why it will benefit them, for example, “I’m going tosave you up to two hours per deal through automation and accurate forecasting.”
But you need buy-in from above. If the CRO or Head of Salesis not on board, then it is not going to work. No matter how strong a leaderyou are, it is not going to work.
However, I have yet to see a sales leader that is not on board.
RB: Let’s talk aboutover-complicating. One of the areas I see people over-complicating isforecasting. If you are a high-growth SMB, what are some of the straightforwardquick wins you can get in forecasting in terms of building a process around it?
NH: Forecasting is a very tough process to master. I thinkof it as setting very clear, almost mathematical roadblocks. First of all, youhave to be able to forecast by the pillars of the business, which are new, upsell,renew, churn. It is not one bucket. You forecast each and every one of themseparately.
Then I basically call BS on deals that look unlikely. If weknow that a deal on average is 40 days, and we are in stage 3 of 5 20 days out,I am going to ask questions. It’s about being realistic and making sureeveryone speaks the same language. What is a commit? What is the best case? Acommit is a deal at stage 5 that is closing in the next 30 days. We have aneconomic buyer, procurement has been vetted. Have a super clear definition andfor the first two or three quarters be as strict as you possibly can.
You would much rather be a little lower on your forecastthan 200% above and then on the last day of the quarter everything startscoming down. I think setting those realistic guidelines is what really helpscompanies go from average to good. From good to great takes a ton of experience,a ton of mentoring during one-on-one’s, doing a bottom up. But off the bat,forecast based on pillars, and have a super clear definition on what a commitis, what a best case is and what pipeline is. Often you find in a team of 20reps you have 21 ways of forecasting. So just aligning on the pure definitionwill be able to get you from 20 reps in 21 ways to 20 reps in two ways. And thenwe can fine tune those two into one and then we move forward.
RB: This ‘calling BS’thing I quite like. What kind of pre-analysis can you do before a forecast issubmitted to be able to ‘call BS’ and bring people into line?
NH: There are a couple of things. There is the closed date,the stage, the amount. Some companies also use forecast categories so commitbest case and pipeline. That is the core five.
Then underneath that you have MEDDIC, BANT and whatever youmeasure internally. But just on that core five you could say, “If your closedate is this week, and we are in stage 2, I am calling BS. If it is stage 2 andit is in your forecasts, for this quarter, I am calling BS. If the amount is$20K below floor and I don’t see an approval, I am calling BS.” Or the other way around, if it is anextremely high number that we’ve never seen but the economic buyer – or procurementhasn’t been vetted – I’m going to call BS.
It’s common sense. Nothing too mathematical or complicated. Simply if you showed this dealto anyone on the street and tell them it’s going to close next week, they wouldcall BS.
RB: Thinking aboutbefore opportunities are added, we prospect someone, we social sell and then wenurture them until there is an opportunity. What is the best way to try to setup that contact op mechanism?
NH: I think, again, keeping it simple. Most of marketingautomation already has a built-in integration with CRM and there is a supersimple way of setting up a process where a lead goes directly into marketingautomation, rather than straight to Salesforce.
So in Salesforce, instead of having 20,000 leads you should have200. You should have the 20,000 inmarketing automation for marketing to work on. You use a lead scoring system todetermine when they move into Salesforce. Then we talk about lead assignmentrules.
RB: And what about onthe outbound side? What is the smartest way, particularly for growing companiesto try and a simple process for that ‘pre-funnel’?
NH: It is one of the areas that I hate most. It is so saturatedwith vendors. Every single SMB has at least three of those tools. Unfortunately,I think we are a couple of years away from having Salesforce own it and do itall. Get whatever tool does the trick for you. At the end of the day you needto be able to track and report on the activities that your team is doing and beable to report on it – not only now at the top of funnel – but all the way throughthe opportunity. Because what will happen eventually, you close an opportunityand you want to say, hey: What was the firsttouch, when was it and what influenced it?
So that is what I would be looking for. Make sure that it isreportable and trackable, all the way from A to Z – and that’s that.
I think, again, in a couple of years Salesforce willprobably build or buy and implement. But it is just too saturated at thispoint.
RB: Going back tostrategy a little bit. As you pointed out in the beginning, strategy in a growthcompany can change and pivot and move and shake. With all that going on, howdoes someone in sales ops decide what to implement and build with that movingbackdrop?
NH: So again, the way I like to see it is top down andbottom up. Top down is what are the marching orders? What are we trying to achieve?Where do we want to be next year? Is it going to be headcount focused? Is itgoing to be ARR focused? Is it going to be product focused? If it is headcount,you want to build your onboarding programme, you want to look at comp and thatterritory.
If it is product, maybe you want to look at CPQ. Maybe youwant to put something in the comp plan that will incentivize shipping out thosenew products. Again, the priority is what are you feeling from the field? Youdon’t want to be too removed.
You are never just looking at the silo; you always need to looka little bit ahead. Build the skill and build dynamic. It is not only systems. Itis comp plans, re-routing, and territory assignment for example. Know that itis going to change. And just having that notion in the back of your mind willmake you act more responsibly and a bit different.
RB: What are yourfavourite and least favourite parts of the job?
NH: Favourite is feeling part of that team, knowing that I’m helping them be better and stronger; enabling them. Least favourite would probably be fire drills, which is something we didn’t think of - a huge error that becomes our sole focus. And being a sales ops person you always have that.
Want to get more insights from sales ops leaders? Check out our other posts in the sales ops interview series.
At Kluster, we're big fans of sales operations...
We recognise the growing importance of sales operations. No longer seen as the function that provides spreadsheets, sales operations is integral to building a repeatable, scalable sales machine.
That's why we built Kluster. We make analytics and forecasting systems for you so you can spend time doing what you do best: uncovering trends and delivering growth defining insights.
Kluster gives you total visibility into the effectiveness of your sales machine and helps you generate credible forecasts to revenue leaders and the board.