Thank you for standing by, and welcome to Schrödinger's conference call for the fourth quarter and full year 2020 financial results. My name is Sarah, and I'll be your conference operator today. After the speaker's presentation, there'll be a question and answer session. To ask a question during the session, you will need to press star then one on your telephone. Please be advised that today's conference is being recorded at the company's request. Now I'd like to introduce your host for today's conference, Ms. Jaren Madden, Senior Vice President of Investor Relations and Communications. Please go ahead.
Thank you, good morning, everyone. Welcome to today's call, during which we will provide an update on the company and review our fourth quarter and full year 2020 financial results. Earlier this morning, we issued a press release summarizing our financial results and progress across the company, which is available on our website at www.schrodinger.com. Here with me on our call today is Ramy Farid, President and Chief Executive Officer, Karen Akinsanya, Executive Vice President, Chief Biomedical Scientist, and Head of Discovery R&D, and Joel Lebowitz, Executive Vice President and Chief Financial Officer. Following our remarks, we'll open up the call for Q&A.
I'd like to remind you that during today's call, management will make statements related to our business that are forward-looking and are made pursuant to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995, including without limitation, statements related to our future financial performance, including our outlook for the full year 2021, the potential advantages of our platform, our strategic plans to accelerate the growth of our software business and advance our collaborative and internal drug discovery programs, risks related to the COVID-19 pandemic, our expectations related to the use of our cash equivalents, and marketable securities, as well as our future operating expenses. These forward-looking statements reflect our current views about our plans, intentions, expectations, strategies, and prospects, which are based on the information currently available to us and on assumptions we have made.
Actual results may differ materially from those described in the forward-looking statements and are subject to a variety of assumptions, uncertainties, risks, and factors that are beyond our control, including the demand for our software solutions, our ability to develop our computational platform, our reliance upon drug discovery collaborators and other risks detailed under the caption Risk Factors and elsewhere in our most recent Securities and Exchange Commission filings and reports. Except as required by law, we undertake no duty or obligation to update any forward-looking statements discussed on this call as a result of new information, future events, changes in expectations, or otherwise. These forward-looking statements should not be relied upon as representing our views as of any date subsequent to today. During this call, we will also discuss certain financial and operating metrics, which are further described in our financial results press release and SEC filings.
With that, I'd like to turn the call over to Ramy.
Thanks, Jaren, thank you everyone for joining us this morning. At Schrödinger, we have developed a world-class computational platform that is accelerating drug discovery and materials design by predicting key molecular properties with very high accuracy. We license our software to biopharma and industrial companies worldwide, we are also leveraging our platform to advance a pipeline of collaborative and internal drug discovery programs with the aim of bringing new medicines to patients. We are extremely pleased with our execution across our business in 2020. Total revenue was $108.1 million, a 26% increase over the prior year. As you'll hear from Karen shortly, we also made excellent progress on our discovery pipeline and expect to initiate R&D enabling studies later this year for multiple programs. We also achieved important milestones in several of our collaborative programs, with three moving into the clinic in 2020.
Additionally, we entered into a strategic collaboration with Bristol Myers Squibb to discover, develop, and commercialize therapeutics in multiple disease areas. The agreement provided $55 million to Schrödinger upfront, and we are eligible to receive up to $2.7 billion in preclinical development, regulatory, and commercial milestones, in addition to royalties on net sales of each product commercialized by Bristol Myers Squibb. We are very pleased to be working with BMS, a proven leader in the areas of oncology and immunology. Turning to our software business, we saw very strong revenue growth in 2020. Software revenue was $92.5 million, a 39% increase over 2019, and we expect continued growth this year. We continue to receive excellent feedback from customers on the impact our platform is having on their discovery programs from both biopharma and materials companies. We are also continuing to advance the science underlying our platform.
We extended FEP+ to more accurately model binding affinities of metalloprotein inhibitors, an important class of inhibitors, better support macrocycle design and optimization, and more accurately predict binding selectivity, which is a major way of reducing potential toxicity of drug molecules. We also released a new active learning workflow for structure-based hit discovery, which can screen massive libraries of compounds with greatly improved computational efficiency. We plan to continue to invest heavily in the underlying science of our platform with a focus on increasing accuracy of the predictions and expanding the domain of applicability to a wider range of targets. As you'll hear shortly from Joel, we ended 2020 in a strong financial position with cash resources of $643 million.
In addition to the continued investment in our computational platform, we are investing in our internal discovery programs to key value inflection points. We are adding new talent to support our R&D initiatives, and we recently hired our first head of early development. Our large number of drug discovery collaborations will continue to play a strategically important role for us as well. We believe having a mix of partnered and wholly owned programs provides us with the flexibility to create long-term value for our shareholders. We're excited by the many advances we've made, and we expect continued progress across all aspects of our business, our internal pipeline, our collaborative and partner programs, and our software business to support both drug discovery and materials design. We continue to navigate the challenges of COVID-19 and working remotely, and we're extremely appreciative of the dedication and resilience of our employees at Schrödinger.
I'll now turn the call over to Karen for an update on our Drug Discovery programs.
Thank you, Ramy, good morning, everyone. In 2020, we continued to make important advances on many fronts across our internal pipeline and portfolio of collaborative programs. Throughout 2020, we reported a significant increase in the number of collaborative programs in lead optimization. By the end of 2020, several of these programs had entered preclinical development or phase I. We expect additional collaborative programs currently at the lead optimization stage to advance during 2021. We have made significant progress in our internal oncology programs targeting solid tumors and hematological malignancies. As Ramy mentioned, last year, we entered into a strategic collaboration with Bristol Myers Squibb for the discovery of small molecule compounds directed to 5 biological targets in oncology, neurology, and immunology. We are pleased that two of our internal programs, HIF2A and SOS1-KRAS, are part of this collaboration.
I'll focus the rest of my remarks on our three wholly owned programs, MALT1, CDC7, and WEE1. Based on the strong data we've generated to date, we plan to move forward with IND-enabling studies for these programs. Subject to completion of the preclinical data packages, we expect to submit up to three IND applications in 2022, with the first submissions expected in the first half of next year. Starting with MALT1, we have made significant progress in the last year. MALT1 is a protein that is downstream of BTK in the NF-κB signaling pathway. Constant activation of NF-κB is a hallmark of several subtypes of lymphoma. We believe that inhibiting MALT1 could be an effective therapeutic strategy to treat certain relapsed or resistant B-cell lymphomas or chronic lymphocytic leukemia. In December, we presented preclinical data from this program at the American Society of Hematology Annual Meeting.
Our compound showed potent in vitro inhibition of MALT1 enzymatic activity. We also reported in vivo antitumor activity in mouse xenograft models of diffuse large B-cell lymphoma, a very difficult to treat blood cancer. Additionally, in in vivo models, our MALT1 inhibitors demonstrated dose-dependent antiproliferative effects in combination with ibrutinib and venetoclax, which are approved BTK and BCL-2 inhibitors respectively. Now I'll turn to CDC7 and WEE1, two programs that target cancer through replication stress and DNA damage repair mechanisms. CDC7 is a protein kinase that has been shown to be required in DNA replication initiation. CDC7 is thought to be linked to cancer cells' proliferative capacity and ability to bypass normal DNA damage responses. Targeting proteins that play important roles in DNA replication and replication stress is gaining momentum as a new therapeutic approach for cancer.
With our program, we believe we have an opportunity to develop a best-in-class inhibitor by improving on key properties, including binding affinity, selectivity, and the pharmacokinetic profile. We expect to report preclinical data from our CDC7 program in the first half of this year. WEE1 is a tyrosine kinase regulator of the G2M cell cycle checkpoint and is a well-validated biological target. The therapeutic objective of targeting WEE1 is to reduce cell viability by inducing G2M phase arrest and apoptosis of cancer cells. Others have shown clinically meaningful tumor regression in uterine serous carcinoma, ovarian, and small cell carcinoma through WEE1 inhibition. However, existing inhibitors have profiles that may make dosing and combination therapy more challenging. We have identified tight-binding, highly selective molecules with optimized drug-like properties, including no observable inactivation of CYP3A4, a key liver enzyme.
As these programs advance and transition into development, we expect to initiate new programs. We have prioritized several new program opportunities with human genetic support and emerging pharmacology data in oncology and immunology. We expect to launch these programs later this year. In summary, our diverse portfolio of collaborative and internal programs is rapidly advancing towards the clinic. Activities to support expansion of our pipeline into additional disease areas are well underway. It has been extremely gratifying to realize the power of our computational platform as we advance our own pipeline. I look forward to updating you on our R&D activities throughout the year. I will now turn the call over to Joel to review our financial results.
Thank you, Karen, and hello, everyone. This morning, I'm pleased to discuss our 2020 financial results, and I'll also provide our outlook for 2021. I'll start with a review of the fourth quarter. Total revenue was $33 million, up 28% compared to the fourth quarter of 2019. Software revenue was $25 million, representing 42% growth compared to the fourth quarter of 2019. As was the case throughout the year, the growth in software revenue was primarily driven by increased adoption of our solutions by large customers, as well as the addition of new customers during the quarter. Drug discovery revenue was $8.1 million, compared to $8.3 million in the fourth quarter of 2019. Revenue this quarter included $1 million from our collaboration with Bristol Myers Squibb, announced in November.
The agreement included a $55 million upfront cash payment, $54 million of which is reflected in deferred revenue and is expected to be recognized over the next 3-4 years as we progress the BMS programs to development candidates. Operating expense was $35.6 million, compared to $23.4 million in the fourth quarter of 2019, reflecting our investment in R&D to advance the science underlying our platform and to progress our internal drug discovery programs, as well as costs required to support a public company infrastructure. We recorded a net loss after adjusting for non-controlling interests of $11.1 million, compared to a loss of $6.8 million in the fourth quarter of 2019. For the full year, total revenue was $108.1 million, a 26% increase over 2019. Software revenue was $92.5 million, up 39% versus 2019, with strong growth in both life sciences and material science.
Discovery revenue was $15.6 million, compared to $18.8 million in 2019. As we've stated before, discovery revenue fluctuates from period to period, as it is dependent on the timing of project milestones. Additionally, discovery revenue is impacted by the timing and revenue recognition of certain transactions, such as the BMS agreement. As I mentioned earlier, $54 million related to the upfront payment from BMS in November was reflected in deferred revenue at year-end. As a result, total deferred revenue was $86.6 million versus $27.3 million at the end of 2019. Of this total, deferred revenue related to software was $30.2 million, up 22% versus the end of 2019, also contributing to the overall growth. Full-year operating expense was $124.4 million, versus $87.8 million in 2019, reflecting our increased investment in R&D and increases in G&A to support our operations as a public company.
In 2020, we recorded other income of $34.6 million, compared to $12.7 million in 2019, driven primarily by the increased market value of our equity holdings in Morphic and Relay. These results demonstrate the value creation opportunity of our collaboration strategy. Net loss, after adjusting for non-controlling interests, was $24.5 million, compared to a loss of $24.6 million in 2019. We ended 2020 with cash equivalents, marketable securities, and restricted cash balances of $643.2 million, up from $599.5 million at the end of the third quarter of 2020, primarily due to the $55 million upfront payment from BMS. For the full year, operating activities generated $16.8 million in net cash versus using $26.1 million in net cash in 2019. In addition to the financial results we just reviewed, I'd like to report on some key software performance indicators for 2020.
Total software Annual Contract Value, or ACV, reached $92.1 million in 2020, representing annual growth of 22%, which is an increase from the 18% annual growth rate we saw in 2019. The number of customers with ACV of more than $1 million increased to 16, up from 10 in 2019. Customers with an ACV over $100,000 increased to 153, compared to 131 in 2019. This customer cohort represented 79% of our total ACV in 2020, and our retention in this customer segment was 99%. Finally, the number of total active customers was 1,463, compared to 1,266 in 2019. We are pleased with the performance across our business, and as we look ahead to this year, we are focused on executing on our strategy and generating long-term growth. At this time, I'll provide our revenue expectations for 2021.
We expect total annual revenue to be in the range of $124 million-$142 million, which includes software revenue of $102 million-$110 million, and discovery revenue of $22 million-$32 million. With regard to software, we are excited about the momentum we've established and the opportunity for growth ahead. In 2020, we saw a large increase in the number of customers spending over $1 million per year, which helped drive our strong revenue growth. We continue to work with our customers to demonstrate the benefits of deploying our solutions at an even greater scale, and we believe there is opportunity for ongoing significant growth over time. Due to the increasing size of the individual contracts and the timeframes associated with larger scale deployments, we anticipate this growth will vary from quarter to quarter and even year to year.
For 2021, specifically, we expect that software growth will be higher in the second half of the year than in the first half due to some of these same factors. With respect to drug discovery revenue, there are a few elements that impact our guidance. The first is revenue recognition related to the BMS transaction. As I've mentioned, we received $55 million in the fourth quarter of 2020, of which $54 million was reflected in deferred revenue. We anticipate this revenue will be recognized over the next three to four years as we advance the programs to development candidates. The second element that impacts our guidance is our strategy to progress our lead internal programs into the clinic ourselves. As a result, we do not currently anticipate licensing revenue in 2021 related to these programs.
Finally, as we previously indicated, a portion of discovery revenue is driven by the timing of collaboration programs achieving certain milestones and can therefore vary from period to period. Finally, I'd like to comment on how we expect operating expense and software gross margin to trend for the year. We anticipate that operating expense growth will be higher than the 42% annual growth rate we saw in 2020, driven by our commitment to fund R&D to advance our technology and our internal drug discovery pipeline. We also anticipate that software gross margin will be lower than the 81% reported in 2020, reflecting investment to drive and support large-scale adoption by our customers. We are very pleased with the results we've achieved over the past year, and we are excited about the opportunities that lie ahead to advance our strategy and drive long-term growth across our business.
I'll now turn the call back over to Ramy.
Thanks, Joel. 2021 has the potential to be another strong year for Schrödinger. We anticipate continued growth in our software business and progress with our collaborative and partnered programs. We are building our internal discovery capabilities to support the addition of new programs in 2021 and IND submissions for our most advanced programs in 2022. We have an exceptional team committed to transforming the way therapeutics and materials are discovered, and we look forward to providing updates on our progress throughout the year. At this time, we'd be happy to take your questions. Operator?
Thank you. As a reminder, to ask a question, you will need to press star then one on your telephone. To withdraw your question, please press the pound key. Our first question comes on the line of Michael Yee with Jefferies. Your line is now open.
Hey, guys. Thanks for the question. I think I have two, and I think it would be very helpful because I think investors are quite confused, and I'm sure you can see where the stock is trading. Number one, on 2021 guidance, can you clarify your software guidance, $102 - $110? I think it implies like, I don't know, 13, 14% year-over-year growth, yet you just came off a whopping 40%. Are you putting some conservatism in there? I know Joel made some comments about second half better than first half. Just help us out with that a little bit. You had a great 2020, people are thinking, "Well, why do we drop to 14%?" Talk about that a lot and tell me if there's conservatism in there. Secondly, can you talk about the drug discovery portion?
I think you made some comments about that. I think some of that lower than consensus revenue is due to amortization of milestones and things of that nature. You can talk to those two different lines, that would be great.
Great. Joel?
Sure. Thanks, Mike. In 2020, we're obviously really happy with the progress that we made, and the fact that it's underpinned by larger contracts and the addition of new customers in a really challenging environment. One impact of contracts getting larger, and we're talking about contracts that are now multi-millions of dollars, is that there's also the potential for more revenue growth variability from quarter-to-quarter and year-to-year as well. What we see going forward is that we're really very excited about the continuing opportunity for growth in both sides of our business, both life science and material science. If you look at what drove the business in 2021 and the increase in annual contract value or ACV from an 18% growth rate in 2019 to a 22% growth rate in 2020, it's the same things that we've been talking about all year.
Continuing increase in large scale adoptions by our customers and the addition of new customers. We know the scale at which we are deploying our solutions on our own internal programs to great impact, and it's still quite a bit larger than even our largest customers. On the addition of new customer side, we are still, we believe, in the early phases, for instance, of our growth cycle on the materials business. We really see significant growth opportunity over the long term in both those areas, and we're investing to capitalize on that. You will see variability in revenue as our contracts get larger, and that is driven by the timing, the kind of contract that it is, and other factors. We're providing guidance to help you understand 2021, but our expectations over the longer term are for continued significant growth.
Joel, you want to emphasize also the second part of my question about the BMS revenue recognition?
Thank you for reminding me about that. Yes, thank you. On the drug discovery side, in the fourth quarter, as you know, we received $55 million in cash from BMS late in the year. It was in November. We recognized $1 million from that $55 million in revenue, $54 million of it went into deferred revenue.
We will recognize a much larger amount related to that in 2021 as we work on the programs and progress them towards delivering development candidates. That $54 million will be recognized over the next three to four years as we do that. That may explain the gap that you were talking about.
Okay. Thank you.
Thank you. Our next question comes from the line of Do Kim with BMO Capital Markets. Your line is now open.
Hi. Good morning. Thanks for taking my questions. I wanted to ask about the drug discovery revenue guidance portion of your guidance. What, besides the BMS upfront payment, are you expecting to drive that guidance? The collaboration programs that you have ongoing, were there any discontinuations or delays that play a part in that guidance? Did you say that you're not expecting a partnership for an internal program this year?
Sure, Do, I can answer that question. With regard to 2021 discovery guidance, as I mentioned, obviously, BMS revenue will continue to be recognized in 2021, obviously to a higher degree than it was in the fourth quarter of 2020. We do expect to continue to earn milestone and research funding revenue in 2021 related to our collaborations. I will say that with regard to receiving those, the programs are advancing broadly. Karen has talked about this. I think most recently you saw a great news release from Morphic on MORF-057 that we worked on earlier this week. In addition to the milestones, you need to also consider the increase in equity value that we have been deriving from these collaborations.
Last year we saw a significant contribution in terms of other income from our equity stakes in Relay, Morphic, as well as cash distribution from Petra. So far this year, that has been continuing. With regard to our internal programs, what I was referring to there is that our three lead programs, as we've talked about, MALT1, CDC7, and WEE1, we are intending to take into the clinic ourselves, at least towards the clinic and into the clinic. Therefore, we're not planning at this point, we're not guiding to any licensing revenue related to those programs in 2021 as we continue to advance those ourselves. With regard to possible other transactions that may or may not occur this year, we're not guiding to that right now. We're not including that in our guidance.
Okay. Got it. Additionally, on the internal programs, you're expecting your first IND filing in the first half of 2021. Is that a delay to your prior expectations? If so, what's driving that?
Karen, do you want to address that?
Yeah. Hi, Do. That does not represent a delay. In fact, I think we've, in the past, explained that our programs have actually moved faster than we expected. In the case of MALT1, we're just around the 2-year mark. By the time we initiate GLP tox, it will be about 2 years actually since we started the program. There are no delays at this point. All of our programs, the data packages are looking good, and we're pretty confident about the opportunity to initiate that first GLP tox study in the first half of this year. As Joel said, also the opportunity for the additional programs over the coming year as well.
From the sound of your comment, it seems that MALT1 will be your first IND.
Yes. Based on all the information we have right now, we did present data at ASH, and I think you've had an opportunity to see that. As we discussed, the package really does look very supportive, and we are on track, as we've discussed, to file that first GLP tox in the first half of this year. We're expecting to initiate clinical trials in the first half of next year.
Great. Thanks for taking my question.
Thank you. Our next question comes from the line of Michael Ryskin with Bank of America. Your line is now open.
Hey, thanks for taking the question, guys. Not to belabor a point, I want to go back to the software guide a little bit, just to parse it out. Given how critical it is to the model. Historically, you've had really good visibility on future year revenues, just given the nature of the license contracts, the fact that there's a pretty substantial amount of deferred revenue every year, the fact that a lot of the contracts are annual, the renewal rates all the visibility that that provides you.
Just to go back to that 10%-19% software outlook, you commented on more and more large contracts as the mix shifts. Is there something that's different about the visibility in these contracts? Is there something that makes you a little bit more cautious about the renewal rate, the pacing of them coming through? Any color you can provide on how that affects your forward visibility on the deferred revenue, on the renewals, and the conversion there? I've got a follow-up question on the drug discovery side.
Sure. Mike, I can answer that. Thanks for the question. Really when you think about driving larger and larger contracts, and again, we think the opportunity is still very significant. The larger commitments that customers are making, there's a lead time, there's a deployment time. We will enter into these contracts, we anticipate, at larger and larger levels over time. The timing of when we convert customers to these larger levels does play a more outsized factor the larger they get. With regard to retention rates, there's no issues there. Obviously, you saw that we have 99% retention in our large customer base over $100,000 ACV. It's really about pretty significant decisions that our customers have to make, multimillion-dollar commitments that they have to make to upsize.
Sometimes that takes time, and we work hard to demonstrate to our customers the value that we're seeing on our internal programs in applying our solutions at much greater scale than they're currently applying to their own programs. That's an ongoing discussion that we have. There's both sales cycle and deployment time, and that does in fact impact the variability in the growth rate. We are confident that over the long term, the value of deploying our programs at much larger scale is large and significant. We are working to achieve that over a multi-year period. Mike, let me emphasize that we still continue to receive really extraordinarily good feedback on the software, on the impact it's having, and that's increasing. That's a really important thing that we continue to see.
Still lots of optimism about being able to convince customers to scale up to the level of usage that our partners and us internally are using.
Would it be fair to say that, again, you saw about 40% software growth in 2020. A lot of that was, as you noted, was a move up in terms of ACV from medium to large to very large. You're not anticipating that those were one-time moves and some of those people may scale back down or may take a pause. Is that fair to say?
Absolutely.
Follow-up question, if I could squeeze a second one in. On the balance sheet and sort of the strength you have there and how that's going into the business, obviously very strong balance sheet. How should we think about that being deployed across the business? I mean, you've got more than enough to support your cash flow and your cash needs as it is now. Are we anticipating accelerating spend to support the clinical trial, cash will be phased out through the course of the rest of this year and next year as those three lead compounds move forward? Or is it sort of the back end of the portfolio reinvesting the software business? Could you comment on that?
Absolutely. I mean, that's what we're obviously most excited about is the progress of the internal programs, the three that Karen talked about. I think that should be viewed, of course, as a very positive thing that these programs have progressed to the point that they have, that we've been able to build internally the organization that's required to be able to go into IND-enabling studies and then into the clinic, and that there's so much excitement about the mechanisms. That's definitely one of the very important part of our strategy is to take these programs further into the clinic to significant value inflection points. We are also, as you know, we have a really differentiated platform and a significant lead over anybody else in the space.
In order to maintain that lead, it is important to continue to very aggressively invest in that platform, maintain that lead, and also to continue to advance the technology. We're seeing the impact it's having at our customer sites, at our own internal programs, at our collaborative programs. We've demonstrated that that investment in the science and getting the science right has a really profound impact on the way programs progress. We'll continue to lead in this space and to advance not only the accuracy of the methods, but the domain of applicability, which allows us to work on even more targets than what's possible today.
Okay. Thank you.
Yep.
Thank you. As a reminder, to ask a question, you will need to press star then one on your telephone. Our next question comes from the line of David Lebowitz with Morgan Stanley. Your line is now open.
Hi. Question. Hello, thank you for taking my question. Given the ACV is at $92, does it suggest that the ability to actually gain new customers is somewhat limited at this point? I guess, what type of customers at this point are you actually trying to add?
Yeah. It depends on what area of the business we're talking about. On the life science side, as you know, the main source of growth there, as Joel talked about and as we've talked about before, is to increase usage. We see huge opportunity there because, as Joel said, the level of usage of the software is much, much lower, even at our largest customers, than what we're using internally and what we're using with our partners. On the material science side, there, a lot of the growth is coming from new customers and new verticals that we're getting into. Hopefully that answers your question.
I guess if we look across the client base, I know that you definitely have had a lot more customers added to the 1 million users, ACV users.
More added to the 100,000 ACV users.
That's right.
With that in mind, are there a proportion of users in the 1 million that are disproportionate and actually affecting the lion's share of the revenue among all the users? While there might be 16 in the +1 million users, are there a smaller proportion that are disproportionately affecting the entire revenue base?
Yeah, I can answer that, Ramy, if you like.
Yeah.
So-
Thanks, Joel.
I think in our K, you can see the top 10 accounted for 34% of our revenue, our top 10 customers. It's up from, I believe, 29% the year before. It is up, and that's an important element of our growth strategy. Obviously, as I mentioned, our large customers getting much larger remains to be a big opportunity for us going forward, as Ramy just discussed. There's plenty of runway left for growth there, we believe. Really, we saw growth across, a really broad-based growth across the business. Importantly, we also saw it was driven by uptake in our core solutions like FEP+ across the business, our enterprise solution, LiveDesign, that connects the entire discovery workflow, and really supporting very solid growth in both life science and material science.
That led to the broad acceleration in ACV in 2020, and it's why we are very excited about the multi-year opportunity that we believe we have to both increase contract size across the business, as well as increase the number of customers as we continue, particularly to advance the material science business.
Thank you for taking my questions.
You bet.
Thank you. Our last question comes from the line of Robert Brooks with Brick by Brick Capital. Your line is now open.
Hi, guys. Thank you for taking my call. I just wanted to talk and touch base a little more on the sales strategy of rolling up those medium-sized customers to large-sized customers. I also wanted to clarify on the 99% retention rate. Is that just for the annual contracts of over $100,000, and is that including the contracts of $1 million and over? Thank you.
Yes. On the second part of your question, yes, it includes all customers over 100,000, including the ones over 1 million, which is, as Joel said, 79% of the revenue. Then you were asking the strategy for growing those customers over 100,000. Is that what you were asking?
Yeah. Also the strategy for bringing in new customers, because I see that your guys' sales and marketing went down from 21- 17 year-over-year, 2019- 2020. I was wondering why is that dropping when you guys are trying to get in more customers in the door and grow those customers?
Right. The strategy on the life science side is quite simple, and we've talked a lot about this. We of course have quite a number, close to 25 programs or so that we're running, where we're essentially using the software at scale, at the appropriate scale to see the impact that we're seeing in our programs, and Karen touched on that. That's the story on the pharma side, is helping customers see that impact, but also deploying it. This requires a pretty significant change in how drug discovery projects are run. Shifting from trial and error and just making a lot of compounds to really deploying computation on a large scale. That's the process there.
That's something that is a scientist to scientist sort of conversation, and we have a lot of people in the company, including people in our drug discovery group, that get involved in that process. On the materials science side, where we're sort of a newer player in that space and the awareness isn't as high as it is on the pharma side. There, it involves that. What is the potential role of computation in molecular design of materials? The challenge there is, unlike in drug discovery, where essentially every pharma and biotech company is effectively doing the same thing, and the vocabulary is the same, in materials, it's a lot of different verticals. We have to talk to companies that are designing organic light-emitting diodes, designing battery electrolytes, designing polymers, electronics industry.
It's a lot of different verticals, and we're building up that part of the organization to be able to get the word out about the impact that computation's having on the design of all types of materials.
Ramy, if I could add on the sales-
Yeah, please.
and marketing expense progression that you referenced. 2020 was obviously an interesting year for a lot of reasons. One of the impacts of COVID, obviously, was a change in how we sell and market our programs, our software. Part of that result was a reduction in the sales and marketing effort, in terms of expenses during the year, travel, conferences, things of that nature. The really exciting thing about what we learned from that was, obviously, a very difficult situation for everybody. What we did learn was other ways for us to effectively communicate and stay in front of our customers and engage with even new customers. You saw the addition of new customers across the different customer size segments.
Going forward, what we anticipate is a return to, hopefully, some semblance of normal at some point during the year where we can re-engage with the customers in person, we can be in person at conferences. We are planning for those reinvestments, if you will, at some point during the year. We'll also retain what we've learned about efficiently reaching out this year, which was very effective. I think you need a combination of those. We do plan to be investing in sales and marketing and also scientific and technical support that Ramy alluded to that are necessary to really pull through and support these larger and larger scale deployments that are so critical.
Thank you, guys. That answered it really well.
Great.
Thank you. There are no further questions. Ladies and gentlemen, this concludes today's conference call. Thank you for your participation. You may now disconnect.