Interview with Barrett Kelly, Global Sales Operations Manager, Exari

A man of the people. This week I got chatting with Barrett Kelly, Global Sales Operations Manager at recently acquired Exari. Amongst the usual sales ops topics, I admired Barrett's consistent consideration for the people around him and how best sales ops could best serve, empower and challenge them. At a busy time for his business, Barrett squeezed in a WebEx with me from breezy Boston. The interview began...

RB: How did you makethe transition into sales ops?

BK: It all started in my role at Wayfair after I graduatedcollege. 3 months in, my manager asked me to be the analytical team lead, whichinvolved diving deep into the data, learning how to use SQL, excel, Tableau etc.I spent a lot of time playing around and teaching myself and it was here that Iformed a passion for analytics.

I then transitioned to Exari where I did a management training program in my first year, rotating through five different departments. I spent a few months in operations working with Legal and Finance. Then I jumped into the Product team where I was the project manager for a new product release. After that I went to sales where I was doing purely sales ops. After that program, I figured out that the sales operations role was a good fit for me, especially on the analytical side. So, I spent another year doing field operations, which then grew into my current role as a Sales Ops Manager, which is mostly sales operations while still supporting the field operations teams.

RB: What are the keyresponsibilities in your current sales ops role?

BK: I help manage the teams. We have two Field Operations GMs at Exari, one runs the NA team and the other runs the EMEA team. I support them and help coordinate with those teams. Strategy and analytics is a huge part, especially on the forecasting side. Forecasting has been a big focus for me for the past year.

RB: What doesanalytics mean to you?

BK: For me, analytics and reporting are in the same bucket.Essentially, it’s about leveraging historical data to enable data-drivendecision making. Instead of saying we should do X because “I think” or “I feel”,analytics enables you to say “the data is showing me Y”. Once you understandthe data, you can then ask questions to the relevant stake holders and take theappropriate next steps.

RB: What’s your role in helping turn these insights into something good commercially?

BK: Originally, I was finding insights and reportingupwards. While it was effective, what I found was far more successful was goingto the stakeholders and using the data insights to ask questions.

For example, I saw that we had a high win rate for existingcustomer deals. So, I went to the relevant stakeholders and asked them why theythink we're successful with existing customer deals, which then started aconversation rather than just giving my prescription straight away. Getting buyin from the stakeholders is huge. If I were to say I think we should do X, Y, Zbecause that’s what the data is saying, it has less impact than if I talk tothe relevant people and ask them what they think the data is suggesting. Byhaving them draw the conclusions from the data and come up with next steps,they become a stakeholder in that idea. Rather than it being my idea, it becomesour idea.

RB: What would beyour advice for anyone starting in sales ops and are there any key metrics tostart with?

BK: The first piece of advice is to be hesitant in making big assumptions and always talk to the relevant stakeholders before making big decisions.

Secondly, use questions to arrive at an answer. It is easy to use data to come up with a plan, but that plan is only as good as the data. If the data isn’t clean or context is needed (which is often the case), you need to sanity check the insights with the people who drive the data.

In terms of key metrics, pipe generation, win rates, average deal size and average deal length are great places to start. Under each one of these metrics are subcategories which are equally important. For example, win rates are typically better for existing customers and looking at both count and value win rates is very important.

RB: Moving onto forecasting, what forecasting methods and processes have you adopted over the years?

BK: Firstly, we have top down forecasting. It's more assumption heavy and uses historical trends to try to predict future outcomes.

This model is more of a November exercise for the following yearbut we also use it to justify new hires. In Exari especially, our ramp is solong for sellers that if you want to make an impact by adding a seller, youhave to do it a year in advance. Unless they're really senior of course.

When we're actually in the year itself, I use the bottom upapproach. When we’re in-quarter or just about to go into the next quarter, theGMs go through all the opportunities in that quarter and assign their opinionsto them. Essentially, they go through all the opportunities and figure outwhat's going to close this quarter. And then we have two different buckets forthe bottom up forecasting - won deals plus GM yes deals and then won deals plusGM yes and maybes. Then we use an average between those two buckets.

RB: How do you makesure that model is accurate?

BK: That's a huge piece for us at the moment. We’ve beenusing our forecasting model for almost a year, so now we have enough data topick out trends. We found, for example, that new customer deals in stage 4 hada low percentage of closing whereas existing customers in stages 4 or 5 workedwell. And then we also looked at stage durations. The quicker the deals movethrough the pipeline the better. These historical insights then get built intothe forecasting model.

We use the bottom up model for the current quarter. Then forthe following quarters, we use a top down approach based on pipe. I look at thewin rates in the pipe with a few variations and then use a weighted pipe. Stageis a big factor in win rates, but we also go deeper, for example looking at newversus existing customer. We also assign specific win rates to the reps if wehave enough data on them. Then in Q2 when we effectively have top down andbottom up, we check the two models against each other.

RB: Who are the keypeople you work together with on a forecast and who are you giving the forecastto?

BK: We like to have real time forecasting, which is veryreliant on good data hygiene. The GMs are responsible for the bottom upapproach and they’re responsible for making sure the reps have good datahygiene. The biggest stakeholders are the GMs, the finance team and myself. Ibuild the forecast for the GMs. Then it goes to our CFO who will maybe make oneor two adjustments before passing it to the CEO. It then forms part of ourmonthly management pack which goes to the board of directors.

The forecast also really helps the professional servicesteam plan and forecast their own projects. Rather than the professionalservices team just getting handed the deal when it closes, they know it'scoming down the pipe so they can start planning accordingly for those projects.

RB: Do you think asalesperson's ability to forecast tells you about the way they manage their ownpipeline?

BK: I think it says a lot. Consistently high forecasts canlead to questions such as, are they not having pricing negotiations earlyenough? Have they not made sure that the customer wants that price? Are theycommunicating well with other teams? We've seen some reps have a really longsales cycle because they're not communicating with other teams, so the otherteams aren’t helping them as much. Also, the pipeline might not be as real asit seems sometimes. You might get reps who aren’t doing as well as they’d likeand so keep a meaty pipeline. But when you get into the weeds, you can see thatmost of those deals are dead.

RB: How can we measure success in sales operations?

BK: We follow the agile methodology. We have quarterly goalsand everything we do feeds into those goals. When new tasks come in, you throwit into the backlog and constantly evaluate the backlog to see if anythingneeds to be brought into the current quarter. It's having a focused approach.And then you can start measuring your achievements against your goals.

RB: It's taking theagile way of working and bringing it into sales operations, so you're focusedon the goal for the quarter and you manage those individual tasks accordingly.

BK: Exactly. That’s the first element of measuring success. Secondly, I spend a lot of time managing software and working to improve our data hygiene to ensure better visibility and reporting. Communication with the teams and trying to get feedback from the management team are very important as well. Lastly another big piece for me is personal satisfaction, making sure I like what I’m doing and I’m going after goals that are going to make the company better.

Want to get more insights from the sales ops leaders? Read 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.