Apply to Register Here

Founded in 2015, Activant is a global investment firm with $1.5 B AUM that partners with high-growth founders and teams that are transforming commerce.

Activant

We’re always looking to learn about new perspectives, new research, and new ideas.

Get in touch

Disclaimers & Terms of Use

Privacy Policy

323 Railroad Avenue
Greenwich, CT 06830
USA

Greenwich

New York

Berlin

Constantia Uitsig Wine Estate
Spaanschemat River Road
Cape Town 7806
South Africa

Cape Town

Harish is providing the Amazon Prime fulfillment experience to everyone else through a combination of AI and an asset-light network of warehouses across the United States.

Alex is automating workflows for the world’s largest enterprises by using process mining to build the connective tissue for businesses.

Mayan is delivering Autonomous IT to the largest companies in the world through systems of intelligence.

Carsten built one of the largest commerce platforms in the world, which powers enterprise merchants and retailers across the globe and now serves as the SAP commerce engine.

Soups is stopping financial crime in its tracks by building the first end-to-end platform for fraud, risk, and compliance.

Adam is enabling restaurant owners to take back their customer relationships by digitizing their entire online and offline operations in a single platform.

Activant is a research-led growth firm that invests in relentless founders

Read past issues

Monthly Newsletter

Browse Research Library

Research

Perspectives

Home

Activant is a research-led global investment firm that partners with high-growth companies.

We partner with founders who have won their initial battles and are ready for the next challenge.

We typically invest when two things come together. First, the company has the potential to achieve escape velocity and transform an entire industry. Second, we believe the founders can lead the company through hyper growth, and our support can significantly increase the probability of this success. 

The critical time window to escape competitors and transform an industry is not open for long.

That’s why after we invest, the real work begins.

About

We only partner with a few founders per year. But when we do, we’re fully committed.

Our portfolio

Companies

Read our investment rationales and market perspectives.

A team built to support and advise yours.

We invest behind insights.

View all research

Founders who aspire to build something bigger than themselves chose to partner with us at their most critical point.

Team

Raised in Cape Town, South Africa, Nina studied at the University of Cape Town with a triple major in Applied Statistics, Economics, and Human Physiology. She nurtured her passion for the investment world by diving into her MSc in Mathematical and Financial Statistics, focused on developing new algorithmic scenario-based approaches to investment portfolio construction. She went on to begin her industry experience as a quantitative analyst at a boutique hedge fund and investment management firm sharpening her data analytics skills. At Activant Nina forms part of the research team, working towards identifying the next game-changing technologies and software-obsessed founders.

Growing up and to this day, I always ask the nagging question "But why?"... Approaching complexity and novelty in this way drives my curiosity to understand markets and founders. Relentless curiosity is a cornerstone of the entire Activant team, enabling us to identify the next disrupters in the world of startups. A broad knowledge base in statistics, economics, computer science, and as far as human physiology shapes how I approach research. It enables me to view things from different angles and to apply my technical knowledge when digging into the crux of the data and software world.

In the race to outpace competitors, companies across industries are turning to data to speed up and improve decision making, roll out better products, and – most importantly – leverage modern AI models and tools. 

, for example, generates ~$130 billion in advertising revenue with dynamic, targeted advertising powered by data from ~4bn 

 broke into the saturated food delivery sector thanks to the insights from ~9 billion annual trips, where more accurate delivery estimates keep 

 AI models also power music recommendations on 

, protect against fraud when consumers tap 

 credit cards, and power your first article draft through 

 – all originating from the troves of data used to train those models.

While these may just be anecdotes, the global economy is pouring investment into the infrastructure that underpins data analytics and AI, creating a burgeoning $100 billion 

 This is a mature market that has brought us scaled public companies like 

, leading outsiders to think that data teams operate in a highly sophisticated, flawless manner, but that couldn’t be further from the truth.

Most investments into data are a failure. In fact, 87% of data analytics projects never make it to production

 Data platforms are chaotic, manually operated, and highly error-prone. In this article, we dive into these issues and introduce a new approach – DataOps ­– which is ready to not just solve those problems, but provide a 10x improvement.

of data science projects never make it to production.

The amount of data that we’re generating is exploding. New data volumes are 

expected to grow by 80% in just the next three-year period 

and will soon be nearly 100x their 2010 level. As we generate and capture more data, the opportunities for data-driven companies like 

 naturally expand, but so does the pressure on data teams and their infrastructure. Data teams can run into processing bottlenecks, scaling limitations, and skyrocketing costs, with these issues made all the worse if their data platform was architected 

 a 100x jump in data volumes. For data-driven business models, this volume growth is the gift that just keeps giving, but for data teams, it’s the hamster wheel that just keeps spinning.

Worldwide Annual Data Generation Forecast, in 

Increasingly, companies expect and need data to be pushed through their systems 

. Executive teams want live dashboards of their key business metrics, and technology companies like 

 need their decisioning systems to be updated for real-time data including traffic, driver capacity, and more. The shift from batch to real-time processing is both a significant trend and a challenge for teams already struggling with batch infrastructure.

Expected Change in Data Budget Over the Next 5 – 10 Y

And it’s not only executives demanding access to more data - 

means that employees at all levels are expected to leverage enterprise data in their day-to-day jobs. But that also means that data infrastructure needs to be more accessible, data more user-friendly and collaboration fostered without requiring years of technical expertise.

, arguably the strongest driver of new data use, is forecast 

to jump from its current 5% to 80% by 2026 

as companies race to unlock the value promised by large language 

 However, 70% of organizations have already found data to be a challenge to capturing value from Generative AI tools and cite it as their most common 

Enterprises are using more data, more timeously, with increasingly sophisticated AI systems. These trends are here to stay and data teams will have to ensure that their infrastructure is up to the challenge. It likely isn’t. In a recent study 

96% of analytics leaders reported being held back by data management challenges

 - dealing with the storage, access, quality, state, and flow of the 

Practically all organizations are struggling to manage their data, as reflected in the following commonly occurring themes in the IT landscape

1. Data and teams are siloed and fragmented:

Organizations use an average of 187 different software systems. For large enterprises, this number can reach 

 The data in these systems is trapped unless teams can build and manage the pipelines that connect all these systems. However, with on-premises, legacy, and closed systems in the mix, building and maintaining these pipelines can be a highly complex and resource-intensive task. Compounding this issue, each system might have its own departmental owner (a 

 challenge), creating a bureaucratic nightmare that could delay executing on data projects by as much as 12 to 18 

2. The “Modern Data Stack” itself is fragmented and complex:

 exemplify the modern data stack (MDS) – a modular network of best-of-breed tools built in the mid to late 2010s to empower data engineers by being easy to set up and use. Unfortunately, these characteristics meant that teams quickly become inundated with numerous, sometimes duplicative MDS tools, aggregating into an infrastructure that is costly and complex to manage. Worse still, MDS tools targeted at business users might have thin APIs, making it tough to integrate them into one single control plane for data platform visibility and automation. 

Hence, the MDS has no single source of truth

3. As a result, almost half of all data workflows are still manual:

The fragmented nature of data necessitates significant intervention by data teams for tasks such as data discovery, lineage tracking, governance, quality checks, and ETL (extract, transform, load) development. These processes are time-consuming and error-prone, introducing data quality and reliability concerns. As we can see below, nearly half of these are still handled manually using pieced-together code to build and maintain pipelines.

Percentage of Processes That are Manual (Moderately and H

4. Data outputs are error-prone and teams lack visibility for root-cause analysis

Consider an AI project that aims to enrich the quarterly financial reporting for a CEO’s board meeting. The project would need to pull data from various sources such as 

. When this is done by a data engineer in an ad-hoc and manual fashion, errors often arise and when the CEO says that one of the figures “looks off,” it can take days or even weeks to trace through the system to test validity and correct any errors. If this sounds like an exception, consider that 33% of business users are not confident in the quality of data provided to them and 44% don’t have visibility into their data 

Moving from the status quo to the infrastructure that our future requires will drive a massive upgrade in both data tooling and practices. We think that DataOps presents that upgrade.

, but also encompasses several key technologies. At its core, DataOps is about making data teams more efficient - breaking down silos between themselves and IT staff and then accelerating data analytics and the AI lifecycle. Applied correctly, these tools improve the quality and reliability of final data outputs.

From a technology perspective, DataOps products augment the data infrastructure to achieve these goals. Using our Activant Value Chain below we define five core capabilities in the context of the data pipeline infrastructure and summarize how we view the DataOps market:

Our View on DataOps

capabilities enable automatic systems searches to discover new or previously overlooked data across all sources. This capability is rarely seen as a stand-alone solution but is rather coupled with data cataloguing capabilities.

serve as a single source of truth for exploring datasets across fragmented infrastructure and are essential for facilitating data lineage. Lineage visualizes and traces data flow, recording usage details throughout. 

 enhances cataloging capabilities by using AI for data tagging and has a built-in AI copilot, focusing on creating and analyzing metadata to categorize datasets and present data in a user-friendly way. Vendors like 

, however, are more focused on lineage - making use of the catalog metadata. Users are able to understand how data is being leveraged and can assess how changes may impact downstream systems, preempting potential disruptions.

tools implement automated validation/testing at various points along the pipeline to ensure the accuracy and reliability of users' data. 

 is an innovator in this space, providing out-of-the-box quality checks and empowering users at all technical levels by implementing AI to convert natural language processing (NLP) into RegEx/SQL for generative AI quality checks. Comprehensive quality tools such as these quantify reliability levels, reduce time spent identifying errors, and automate manual testing procedures.

 tools collect and analyze logs and performance metrics across the entire pipeline to flag errors/failures in the execution of pipeline jobs. Providers collate and present the metrics in dashboards to provide visibility into data pipelines. 

s solution stands out in this segment thanks to their dual focus on the monitoring of data pipelines as well as the data itself, creating a unified assessment of the health of a customer’s data lifecycle. They also offer a built-in co-pilot, as well as platform audit reports for vendors such as 

automates the pipeline end-to-end, triggering workflows based on schedules, events, or other preconditions. Orchestration platforms act as an authoritative source of truth by defining workflow execution and implementing optimizations such as parallel processing to increase data throughput. Orchestration, traditionally a tech-heavy task, is moving towards more accessible tools that incorporate user-friendly low-code options. 

 is a leader that strikes a good balance – providing user-friendliness while still maintaining high-code capabilities to offer flexibility to data engineers.

In addition to the key areas of our value chain outlined above, there are a few key trends that drive the market direction. These include notable platforms, the proliferation of open-source and the rise of data products.

 stretch across the value chain with the most full-featured products on the market. 

 adopts an asset-orientated approach (rather than a task/workflow-oriented approach), which enhances debugging, reusability, visibility, flexibility, and data management. Following 

 the company is implementing AI-enabled solutions that offer cost optimization, discovery of security vulnerabilities, improved architecture, and AI-assisted orchestration.

Throughout the value chain (notably in orchestration), open-source software (OSS) tools (like 

) have found significant traction, but increasing adoption of managed orchestration platforms is underway as they decrease labor overheads and provide more advanced capabilities than their OSS counterparts.

Providers are not only embracing core DataOps capabilities but also integrating data products and low-code/no-code (LCNC) functionality into their strategies as a way of liberating data for non-technical users and achieving data democratization. Vendors like 

 are working hard to create data hubs and marketplaces, offering a user-friendly “online shopping” experience for data to be accessed by all 

As we can see below, DataOps augments and improves the existing flow of data. Rather than surfing thousands of data sources, data users simply go to their 

 to explore data, their ELT & ETL jobs are executed autonomously by 

As data flows through to its end location, it can be tested for accuracy by 

tools both before it is loaded in the analytical store and when it is queried by the target system. Finally, data consumers can discover analytics-ready datasets as 

without worrying about all the integration and transformation that took place before that. 

Previously, data teams were suffering from significant manual work, a lack of visibility, and poor-quality data. Now, their workflows are automated, monitored, and any data quality issues are detected automatically.

Data Infrastructure with DataOps 

While the traditional data flow can exist without these DataOps tools, the value proposition that DataOps brings to the data team is becoming clear.

Gartner believes that data teams who make use of DataOps tools and practices will be 10x more 

 Given the scale of issues that data teams are facing, such an immense gain 

achievable, and the research backs that up:

reduction in data pipeline development time.

is achieved by teams who use DataOps tools to reduce time spent on manual work like pipeline builds and error troubleshooting. Users are reporting an 80% reduction in the time taken to develop data pipelines, equating to a 

 found that implementing DataOps practices for a customer had the potential to reduce time to market from 100 days to 

 found that through these improvements, engineers using DataOps tools are 5x to 7x more 

Reduced infrastructure costs and greater scalability 

are the primary benefits of automation. In some cases, users have reported a 10x increase in pipeline throughput and a significant reduction in processing times, from 8 hours per client to 3 hours for all 

 case referenced above, they also found a 

, and where customers leveraged cloud computing, they were able to achieve approximately 

can save organizations up to an average of $12.9 million 

 By automating data quality solutions, companies are rebuilding trust in AI models, dashboards, charts, and statistics that are relied on for decision-making.

With such immense gains available on the table, we see DataOps tools becoming a mission-critical element of every enterprise tech stack, driving efficiency, scalability, and accuracy that modern data teams can't afford to overlook.

With the rapid adoption of generative AI and the increasing need for data on demand, DataOps will undoubtedly shift from a "nice to have" to a necessity. Its ability to streamline operations, automate workflows, and ensure data quality will make it essential for modern enterprises aiming to stay competitive in a data-driven world. Reflecting on all that technology has achieved with broken and inefficient data practices, it’s incredibly exciting to look forward to what we can achieve with modern analytics, generative AI and 

If you have a different view or are building in this space, we’d love to hear from you.

The information contained herein is provided for informational purposes only and should not be construed as investment advice. The opinions, views, forecasts, performance, estimates, etc. expressed herein are subject to change without notice. Certain statements contained herein reflect the subjective views and opinions of Activant. Past performance is not indicative of future results. No representation is made that any investment will or is likely to achieve its objectives. All investments involve risk and may result in loss. This newsletter does not constitute an offer to sell or a solicitation of an offer to buy any security. Activant does not provide tax or legal advice and you are encouraged to seek the advice of a tax or legal professional regarding your individual circumstances.

This content may not under any circumstances be relied upon when making a decision to invest in any fund or investment, including those managed by Activant. Certain information contained in here has been obtained from third-party sources, including from portfolio companies of funds managed by Activant. While taken from sources believed to be reliable, Activant has not independently verified such information and makes no representations about the current or enduring accuracy of the information or its appropriateness for a given situation.

Activant does not solicit or make its services available to the public. The content provided herein may include information regarding past and/or present portfolio companies or investments managed by Activant, its affiliates and/or personnel. References to specific companies are for illustrative purposes only and do not necessarily reflect Activant investments. It should not be assumed that investments made in the future will have similar characteristics. Please see “full list of investments” at https://activantcapital.com/companies/ for a full list of investments. Any portfolio companies discussed herein should not be assumed to have been profitable. Certain information herein constitutes “forward-looking statements.” All forward-looking statements represent only the intent and belief of Activant as of the date such statements were made. None of Activant or any of its affiliates (i) assumes any responsibility for the accuracy and completeness of any forward-looking statements or (ii) undertakes any obligation to disseminate any updates or revisions to any forward-looking statement contained herein to reflect any change in their expectation with regard thereto or any change in events, conditions or circumstances on which any such statement is based. Due to various risks and uncertainties, actual events or results may differ materially from those reflected or contemplated in such forward-looking statements.

It's Time for DataOps 

And time to 10x your data platform

It’s Time for DataOps

Safeguarding the Mona Lisa is a tough job – a prized masterpiece that could be subject to any number of elaborate heists. That’s why it is protected by powerful security tools like security cameras, motion detectors and bulletproof glass. However, the Mona Lisa does not change, replicate, or move, which makes the security job a lot easier. Data Security teams, on the other hand, must protect an asset that flies around the globe through clouds and third-party apps, where hundreds of employees demand daily, unrestricted access to that data to get their jobs done. While the Mona Lisa hangs safely in the Louvre, 

data breaches are occurring at a rate of five every single day


In this report, we explore what makes that problem so hard, why it’s about to get even harder, and how we see the solution evolving through 

. Today, companies invest over $200bn on traditional security tools, but the $5bn Data Security market is too often an afterthought. To overcome existing data breach challenges, companies need to make Data Security the building block of their security strategy.

We’ve all been the subject of phishing emails. From personal experience, we know 

. And for large enterprises with valuable IP, customer data and multi-million-dollar transaction flows, it’s an order of magnitude worse. In 2022, the average enterprise experienced over 1,000 cyber-attacks per week, or 150 every single day – more than doubling since 2020.

With sophisticated actors frequently attacking, data breaches occur at an alarming rate, with 1,862 breaches recorded in 

 These breaches impact a wide range of companies– 53% of organizations stated that they had experienced a 

 And there are many recognizable examples: 

 all experienced breaches in the last few months.

 That bill could rise to $10mn+ for industries with extensive sensitive data, like healthcare.

 due to downtime, management distraction, loss of intellectual property and reputational damage: 73% of consumers would reconsider using a company’s product if their personal data was exposed.

’s systems were down for over a month, creating a backlog of $14bn in payments that were due through its claims 

Over 80% of impacted businesses pay ransom fees, which totaled $1.1bn in 

For example, Equifax’s 2017 breach saw them pay £11mn to the UK Financial Conduct Authority and $575mn to the US Federal Trade 

In the information age, data is a company’s most critical asset. Unlike physical assets, data cannot be replaced once it’s lost. And for customers, they can’t change critical identifiers like names and social security numbers. Once revealed, that information is public forever. You can’t un-breach data, so the stakes are incredibly high. 

To understand how we arrived at this fraught situation, we must examine the key factors making data security an intensely difficult challenge.

1. More data, duplicated across more systems, accessed more widely:

New data is growing rapidly – estimates suggest that the number of sensitive data files organizations must manage will increase ~5x by   

Even a well-architected security policy can become inundated with a growing stock of data to manage over time, 

yet 66% of staff believe that their data has already outpaced their ability to secure

 This is worsened by growing demands to be data-driven, which create a larger attack surface across different analytics systems and SaaS apps. Such demands also force companies to have more lax data access policies, lest the data security team becomes inundated with access requests and teams are left waiting too long for critical decision-making material. 

2. Disparate data sources create a lack of visibility into data:

With organizations managing an average of 187+ data sources and systems, it is incredibly difficult to appropriately classify and tag all incoming 

 As a result, organizations are often unaware of the sensitive information they possess and where it resides, hindering their ability to safeguard it adequately

. Hence only 4% of companies have dedicated storage for sensitive 

3. Inconsistent governance and security posture:

39% of companies use more than one cloud computing vendor (“multi-cloud”), and 84% mix cloud and on-premises resources 

 Also consider that companies may combine numerous data tools, such as using both Snowflake and Databricks. As a result, it is difficult to manage a consistent and unified policy, leaving vulnerabilities, especially in the cloud. Cloud resources can become vulnerable due to poor access controls, unrestricted ports and unsecured backups – 99% of cloud IDs are “excessively 

80% of data breaches involve data stored in the cloud

137 countries have instituted data protection legislation, the most prominent being 

 Individually, these regulations burden data teams with extensive considerations surrounding how they handle data, with equally painful costs for non-compliance (GDPR penalties can be as much as 4% of global turnover). But collectively, these disparate regulations create a maze that is near impossible for global businesses to navigate. With the recent proposal of an 

, all signs point to this situation intensifying.

 models will 10x the number of data users, while AI co-pilots drive more SaaS systems to demand enterprise data access. 64% of companies report being “under pressure” to adopt Generative AI, but 84% see cybersecurity as the primary 

 Those roadblocks exist across the model lifecycle:

Model training requires centralizing training data which creates a honeypot for attackers.

Model development often involves the use of open-source models as well as testing with numerous models, resulting in some enterprises running over 100 different 

 This makes it difficult to detect security vulnerabilities in all the models being used over time, leading to 

more than 40% of companies suffering from privacy or security issues 

The inference phase exposes companies to prompt injection, where attackers can use specific prompts to drive the model to reveal sensitive information or inject malicious code into its underlying database to enable hackers to gain access to a system. For example, researchers at Google forced ChatGPT to reveal sensitive information from its training 

 Companies need control over what data employees provide to AI systems, which is difficult to manage at scale, particularly when employees make use of non-sanctioned AI systems, giving rise to “shadow AI.”

It should be no surprise then that 40% of chief security officers feel that they are unprepared for the current threat environment. Implementing stronger data governance and security controls ranked as the number one priority for data leaders, above modernizing their data infrastructure or integrating AI into business 

Companies must prioritize data security, not just for compliance reasons but to maintain business continuity, brand value and competitive positioning. Data breaches cannot be treated as routine - they pose existential threats with the potential to cripple businesses.

Today, cybersecurity is a $200bn+ market and at just ~$5bn, Data Security appears to be an afterthought for CISOs. Putting data first promises a more sound strategy. Once you know where your sensitive data is, you can apply tools like access control and governance, but also have a much more targeted strategy for tools like threat intelligence, network security and cloud security. Perhaps, with efficiency gains from this approach, companies that start with Data Security could even see some overall savings on their cybersecurity budget. 

It's why we see data security becoming the fundamental building block of the broader security stack.

That sentiment is echoed by Anshu Sharma, whose highest priority security actions start with data:

That’s why we see the market evolving from its current estimates of ~$5bn to $47bn over the next decade (25% CAGR)

. We estimated our TAM based on industry-specific ACVs, considering the level of sensitive data and data complexity for that industry. From conversations with CISOs, companies are likely to spend anywhere from $500k - $5mn on data security, depending on their unique needs. We narrowed this range down to $500k - $1.5mn, noting that ACV’s of $5mn would likely only apply to extreme outliers by size and complexity. ACV is multiplied out by the number of enterprises (firms with 1000+ FTEs) per the Bureau of Labor Statistics.

This is an emerging market segment with so much potential – data security has numerous exciting components and an ecosystem of start-ups that are leading the way forward. In the next section, we walk through how we break down the data security market in the form of our Activant Value Chain.

Our View on Data Security

Working from left to right, we’ll discuss the most critical pieces of this value chain, the jobs to be done for those software products, and highlight a few of the companies leading the way in each segment.

Lack of visibility remains one of the core pain points when trying to secure data. Companies often don’t know what data they have, making it impossible to adequately secure it. 

Discovery and classification serve as the crucial first step.

 These tools automatically locate and catalog all data, from sources like databases and SaaS apps. Classification models can tag data based on its type (PII, health, etc.) and risk level. Once companies understand their risk, the right 

 scan enterprise data in any data store without installing agents on your devices, while working with both structured and unstructured data, where 

A comprehensive encryption strategy can help companies comply with regulations and serve as a last line of defense if data is breached. Sensitive data should always be subject to some form of encryption, but traditional encryption can make data 

Companies might consider technologies like tokenization (a key tool used by 

), format preserving encryption, and homomorphic encryption (which could power machine learning on encrypted data).

 is building a new piece of this architecture – the data privacy vault, a centralized point of control for sensitive data­– to ensure that data doesn’t proliferate across a company’s systems. 

 Microshard technology protects data from ransomware by shredding it into 4-byte pieces and distributing it across cloud storage locations.

Office

Nina
Matthews

Research Analyst

Office

Research

Relevant Experiences

Office

Nina Matthews

Research Analyst

Office

Research

Relevant Experiences

Nina
Matthews