‍Desk Notes is a series of insights from our experience working actively with over 40 SaaS companies. You can find more of Allen's thoughts on tech through his Medium, Twitter, or reach out directly at allen@evp.com.au

‍

****

Introduction

Recent debates about AI's potential to disrupt software business models have prompted me to examine how SaaS companies construct and maintain their competitive edges, also known as "moats". The questions arise: What exactly are these moats? Do they still exist? And how should we consider them while building resilient SaaS companies?

All startup companies begin life vulnerable and exposed. There are few barriers preventing competitors from entering the market or replicating what they do. However, as companies mature, the strongest ones have set up "moats" - barriers that demand an overwhelming commitment of resources – financial, time, or logistical – to replicate their operations.

Early stage moats are fragile structures that harden as the company scales. I want to hone in on the foundational moat concepts relevant to B2B software companies, discuss their relevance at various stages and analyse some of the strategic ‘plays’ we’ve observed. The moat definitions are drawn from the book 7 Powers and are applied specifically to the current generation of emerging B2B software companies. This article focuses on four of the seven powers which are most relevant from company formation through to the growth stages.

B2B Software Moats

Counter-positioning - the opening play

Counter positioning is about defining your offering in a way that existing companies cannot easily replicate or substitute without significant trade-offs.

True counter positioning has to create sufficient friction for existing players to avoid replication, but it rarely stops new entrants from doing the same. It is closely related to the concept of disruptive innovation, which is about attacking existing markets via new products. I like to think of it as an opening strategy and not a strong standalone moat. Some of the playbooks below showcase common methods to counter-position a software product.

Find a vertical niche

Selling hyper specific solutions that carve out a segment of the market already occupied by others is an example of counter-positioning. Building bespoke customisations to a platform that fit the needs of a segment can require considerable investment in product which in turn builds friction to copy. The new solutions are not encumbered by legacy tech stacks, data schemas and infrastructure.

We continue to see opportunities in vertical software with new players taking slices out of underserved segments such as Nexl in legal, upgrading 90s era solutions created by LexisNexis and replacing the overly bloated current gen of CRM like Hubspot and Salesforce. Ignition carved out a position in the accounting vertical as a purpose-built proposal solution, solving very specifically for the needs of the industry. Focus ultimately enables both these EVP companies to deliver value far quicker than general purpose products.

Disrupt traditional services

Technology products tend to counter position well against service businesses where there is some human overhead. New technology can reduce the cost for customers significantly by introducing concepts like:

• Automations - Hnry automated the process of managing taxes as a sole-trader and disrupted a layer of service providers
• Self-service - The Trade Desk dislodged traditional media buyers and agency workers by introducing a self serve platform for buying media
• Always-on - Mutinex is transforming the way marketers engage with their marketing analytics by enabling them to be always-on rather than once a year that they would get with consulting firms

Shift the distribution model

The method in which you distribute and deliver your product to the end user can be a form of counter-positioning. At a macro level, the shift from on-prem to SaaS was a counter position that wiped out the older generation of solutions. The strategy of bottom-up delivery, where you enabled individuals and small teams to sign-up independently to your product, paved the way for a new generation of enterprise software to gain share.

We saw Atlassian turn the enterprise sales model on its head by executing a low-touch model. They removed friction with an easy trial process, extensive marketing / knowledge materials, quick time to value and a transparent pricing model. In the local consumer space we saw the battle of fast groceries (Milkrun, Voly, Send) against the traditional grocery stores as an attempt to counterposition on brick and mortar retail. Coles and Woolworths would have to consider cannibalising their own retail store traffic in order to defend and we saw the launch of Metro60.

Switching costs - the growth stage strategy

Switching costs arise when a consumer values compatibility across multiple purchases or interactions with a specific product over time. The cost/benefit analysis skews in favour of the existing solution when considering a switch.

If changing from one product to another is difficult (i.e. you lose embedded knowledge, routines, data records), then the cost to change can weigh heavily on any desire to switch out of your software solution, creating a gravity towards your product. I see this as a moat that progressively takes form throughout the growth phase of the product, although it requires careful consideration to set-up the right product frameworks.

Think about the shelf life of your data

I recently came across the term shelf-life or half-life of data and I really appreciate how it conveys the fluctuating usefulness of data through time. Let’s consider the data from an Applicant Tracking System (JobAdder, UKG, Employment Hero). You may discover that once this data is extracted and employed to fill a job role, its relevance quickly diminishes or becomes 'stale'.

Alternatively the system you install which stores your financial data or operational business data tends to have a longer shelf-life as you use it for short term insight and long-term compliance / reporting obligations. Hence, the 'data freshness' factor merits attention: as you collect data for your customers, ponder over strategies to extend its shelf-life. Just like perishables in a grocery store, data also has an expiration date. The challenge (and opportunity) lies in preserving its value, enhancing its freshness, and extracting the maximum utility before its inevitable 'sell-by date'.

Create an ecosystem

Having an ecosystem of third-party plugins on top of your software platform can rapidly solidify your product advantage as you draw upon users and builders to invest resources into sustaining better user experience. Some great examples of ecosystems in SaaS are found at Shopify, Notion & Figma. Three common ecosystem plays involve 1) user-generated templates, 2) experts/consultants and; 3) custom applications. Check out https://austinyang.co/saas-ecosystem-building/ for some tips. Ecosystems are incredibly sticky because they are inherently difficult to replicate (from both an ecosystem engagement perspective and the product experiences it enables).

Create contractual lock-in

This is a lever wielded by your sales team to create economic lock in. Time will probably work in your favour to create some stickiness in the data that is stored and captured, investment in upskilling the user base and general friction that comes with getting rid of an application that you’ve had around for multiple years.

Build an opinionated product

An opinionated software product believes “that a certain way of approaching business process is inherently better and provides software crafted around that approach.” Think about building products which add incremental value to the underlying data that is specific and opinionated. That way even if data is migrated there is no way to get the same level of value from another product. I see two common ways of adding opinion to your product:

• Create a system of insight that populates your product with analysis that is highly targeted towards the industry and their problems. The most obvious example from experience is the Shopify Analytics engine, full of valuable eCommerce metrics built on top of the existing data records. To frame it another way, don’t let Tableau or Excel be the de-facto analysis layer for your data records
• Incorporate a recommendation engine that surfaces contextual insight leveraging various aspects of your product. Mutinex does this well by nudging users towards certain platform actions that consider their organisation’s data and the current market conditions

Process power - the evergreen moat

Process power refers to competitive advantage that comes with an organisation’s internal capability. It’s an underlooked but powerful moat in software development. Software products can often be built to be internal facing and provide huge improvements in performance, processes and insight for a company.

My view is that process power is one of the first moats that an organisation can build and is powerful across all phases, hence being ‘evergreen’.

Support, implementation and sales automation

Software companies, from my experience, still require a level of human touch and service. It’s not possible to be completely hands off. The best I’ve seen build muscle internally to be hands-on at scale.

• Mutinex has invested heavily in decreasing the cost & time to onboard customers. Effectively becoming best-in-class at data ingestion. This puts them in an excellent position to attack incumbents who still push a lot of the burden onto their own customers and place strain on their implementation teams
• Salesforce is in another league, but to illustrate the point, they’ve invested in training and constructing an ecosystem of support engineers and implementation consultants which extends their reach to customers
• Other companies across the EVP portfolio have employed proprietary lead generation tools which automate discovery of customers, competitor monitoring, sales outreach and follow-ups

Growth infrastructure

When you get to a certain scale, you start to deal with brand new process problems. This is what happened when a team at LinkedIn developed Apache Kafka, an activity tracking and metrics logging application which they needed to manage the volume of traffic they dealt with. Today Kafka is open source and there is a large business called Confluent built around it. Even in the early stages, SaaS companies can consider code infrastructure or business-code logic as a moat which makes it challenging for competitors to replicate.

• Companies like Hnry have invested heavily in deepening their connection with the tax office.
• Freelancer.com built their own sophisticated A/B testing platform that enabled people with lower technical capabilities to design and run growth experiments across the marketplace

Network effects - late growth

I want to keep this section short because I’m sure you’ve read a lot of commentary on network effects already. The strategy which I think is particularly relevant in 2023 is how rapidly one can build a data network effect. Put simply, how quickly can your business learn from its own users? In B2B, companies are rarely optimising for network effects from day one but are positioning themselves to reap its benefits at scale.

There are a couple of considerations to set up for a data moat

• Make sure, contractually, that data insights can be owned and used by your business
• Think about creating a data trap. An activity that feels natural for users to submit their data. In enterprise software land this might look like a data contribution process where you improve the accuracy of the information for yourself while also improving that data-point for all other users.
• Look at alternative data sources. Can you kick-start your data network by acquiring bespoke pieces of data?

‍