6 ms·
Launch HN: VoiceOps (YC W17) – Data analysis for sales calls
I'm Daria, a co-founder and CEO of VoiceOps (https://voiceops.com/ https://voiceops.com/). We built a product that analyzes sales calls and provides actionable tips to close more deals.
We transcribe, parse, then analyze calls for what behavior is most successful (e.g. how many probing questions to ask on a call, the value of describing benefits vs. features, when to time an upsell or close attempt, etc.). Once we understand what works, we help managers scale that out, improving performance across their entire team.
Currently, sales managers spend 10-20 hrs/week listening to the call recordings of their reps and then guessing at how to provide feedback. The process is slow, lacks consistency, isn’t based on data, and often fails to motivate or change behavior.
Every other part of the sales stack (email especially) is already data driven. With sales reps spending most of their time on the phone, it should be the most important activity to optimize.
The prototype was built and first clients closed while I was living in a van and Ethan was sleeping on a kitchen floor in the Mission. Between our three co-founders we’ve spent nearly a decade building sales and support tools for companies like LinkedIn, Gusto, Coinbase, and General Assembly. Even there we saw this lack of data become problematic and expensive.
Take a peek - we’d love to hear your feedback!
- ASpring 9y agoHow do you balance what is effective for closing deals versus what is sustainable for the employees? Are you tracking any satisfaction from the employees and how it changes through this process or in response to the variables in the calls? Taylorism has never been great for employees and it would be fantastic to see a company step up into this role positively.
- dariaevdo 9y agoReally good question - the way we help managers coach reps isn't giving them the data to say, "you should be making 75 percent more calls in a day." We show managers and reps what the highest performers are saying and when they're saying it, and then visualize week-over-week trends. Sales reps are thrilled to have this data - it's helping them understand their own call behavior and adjust accordingly, which has a very tight feedback loop for teams with transactional sales (or any high velocity sales cycles). As a sales rep, if I can see that the person at the top of the leaderboard is asking 8 probing questions per call and hardly ever highlighting product features, and I'm only asking 2 probing questions and talk extensively about features, that's extremely valuable and actionable information. We aren't tracking rep/agent satisfaction quantitatively, but anecdotally we're seeing highly positive feedback.
- Caligula 9y agoDo you do some type of machine learning or actually manually process calls and offer feedback? Is each agent call transcribed?
- dariaevdo 9y agoThe current process on most sales teams is managers listening to 2 or 3 call recordings per rep per week (out of hundreds) and basing feedback on that, so having data across dozens and hundreds of calls is a huge win for them. We can transcribe every agent call, but typically don't have to. It's a combination of both ML and human QA'ing when necessary.
- Nilef 9y agoOne of those ideas I wish I'd had
- dariaevdo 9y agothank you :)
- malchow 9y agoI have a MVNO (mobile virtual network operator) selling high-speed LTE plans and smartphones to folks, mostly via phone. Would this be an effective for a small (~15-person) call center?
- Matetricks 9y agoHow does this compare with other companies in the space like Chorus, Gong, and other YC companies like People.ai or VOIQ? Congrats on the launch!
- dariaevdo 9y agoThe biggest difference is single call vs. trend data. We're focused on not just looking at how to dissect a single call (chorus, gong), but how to make sense of a larger dataset of calls. That let's us understand multi-day/week trends and how to best optimize call behavior across an entire team. As I understand it, People.ai is looking at metadata (call length, percentage of time spent talking vs. listening, etc.).
- rogik 9y agoHey Daria, thank you for the shoutout! We do a bit more than just voice metadata tracking. To use a baseball/Moneyball analogy, while voice is the first/entry thing you do in sales, just like catching the ball, People.ai works across all sales activity channels, such as email, calendar, phone, conference systems and even Slack to optimize individual sales rep and team behavior. While voice analytics tools can help produce the best catchers, we are focused on making every rep a 5 tool player.
- otto_ortega 9y agoHello there, congrats for the launch! I have a question regarding your back-end: What speech-to-text service are you using to transcribe calls? I'm guessing that it is powered by one of the major players in the field (Nuance, Google's Speech API, etc) and I would like to know which one you ended up selecting. I'm developing a service myself (on an unrelated field) that uses speech to text as one of its cornerstones, and I'm still trying to decide which service to use, so your experiences while selecting one may be of use.
- dariaevdo 9y agoThank you! We're keeping the particulars of that piece on the DL for now, but feel free to ping me at daria@voiceops.com if you want to talk shop offline.
- pain_perdu 9y agoHi Otto! While you're waiting a response from VoiceOps, I'd love show you what my company can do. We've spent the last few years building an ASR system with advanced features like biometric speaker diarization, homomorphic encryption, multilanguage support and the fastest STT engine online. Ping me at colinh at intvox dot com if you'd like to check it out.
- mzarazov 9y agoPretty sure they use manual transcriptionists according to LinkedIn employees
- tyingq 9y agoCurious about how you deal with the legality of recording calls. The required verbal notice that people get when they call in for support doesn't seem to be an issue, because the caller can see the benefit back to them (improved support, documenting poor service, etc). In this case, though, I don't see that the person being sold to would feel like a recording provided any benefit to them. I could see it perhaps even reducing the likelihood of a sale. Curious where your internal conversations landed on this.
- dariaevdo 9y agoOur clients (for the most part) already use call recording tools like RingCentral, 8x8, or a recording function within their preferred VOIP tool. As call data becomes more accessible it will become more valuable, and my guess is we'll see more companies adopt call recording for their inside sales teams. Regarding legality, some of these call tools have the option of 1-sided recording, meaning they're only recording what the sales rep is saying.
- weishigoname 9y agointeresting,it must be very valuable if voice recognition applied and accurate, it could be huge market value.
- hubrix 9y agoReach out to me, I'd like to test it on 100 person sales floor. I have years of recordings.
- dariaevdo 9y agoif I guessed correctly who you are just reached out to you :) If not, please shoot me a note daria@voiceops.com or lmk what the best way to get in touch would be.