Steer73 - Digital Transformation Agency

We often speak with IT leaders weighing off-the-shelf solutions because they promise faster implementation. And it’s true, even with efficient delivery processes like ours at Steer73, off-the-shelf systems usually go live a few months sooner than custom builds. But that’s all it is: a few months. Meanwhile, many of our custom systems are still running strong into their 10th year. When you consider the long-term impact, shaving 2–3 months off the launch can come at a real cost: reduced ROI, higher ongoing expenses, less flexibility, supplier lock-in, and limited control over your future roadmap. For a system that may underpin your operations for the next decade, those trade-offs deserve a second look.

Why SaaS is getting more expensive. SaaS pricing isn’t broken. The model was always to acquire market share, then focus on profitability. For many SaaS companies, we are now firmly in the ‘focus on profitability’ stage. This means that your license fees are no longer subsidised by investor money, and you might be seeing: - Higher subscription costs - More restrictive usage/user-based pricing - More features being classed as ‘add-ons’ We've just had a >25% price increase from one of our SaaS providers, with a pretty poor justification for the increase. What is your experience?

Our team's put together a whitepaper on AI in construction. If you'd find it useful, DM me or comment below. AI has been quietly revolutionising sectors such as construction, bringing about transformative changes that are reshaping how projects are conceived, managed, and delivered. Our whitepaper shows how leading construction firms are utilising AI to reduce costs, improve safety and prevent project delays. If you want me to send it to you, just comment on this post or drop me a DM and I will send it over.

Custom development used to be expensive. Not anymore. Here’s why: • Pre-built components mean systems are never built from scratch • Open-source frameworks assist significantly with standard features • Cloud platforms like Azure eliminate hardware spend • Mature delivery processes reduce waste and speed up development In many cases, you can now get a tailored system for less than SaaS.

Mobile APIs face a different security landscape. Devices are more easily lost or compromised, and the connection layer can be targeted. Minimum security standards should include: - OAuth2 or JWTs with short-lived tokens and secure refresh - Encrypted transport via TLS - Optional certificate pinning for high-risk use cases - Local encryption for sensitive on-device data - Minimal caching of secure content

Designing APIs for mobile requires different priorities than web or desktop. Mobile environments introduce constraints that are often underappreciated: - Unreliable networks - Device variability - Data sensitivity - Long-tail version adoption Every aspect of the API, from payload structure to auth handling, must reflect these realities. When done well, mobile APIs improve responsiveness, reliability, and user trust, even under adverse conditions.

Optimising mobile API’s for performance. A well-thought-out API can dramatically improve your app’s user experience, making it more responsive, reliable, and secure. When designing an API for mobile, consider that bandwidth can be limited and latency can be high. Here are some options for optimising mobile APIs for real-world conditions, as used by our team: - Reduce round-trips: Minimise the number of API calls needed to retrieve data by structuring endpoints to return all necessary information in fewer requests. Implement caching: - Client-side caching - Avoid redundant network calls by storing frequently accessed data on the device. - Server-side caching/CDNs - Serve static or semi-static content closer to the user to reduce latency. - Optimise payload size: Use efficient formats (e.g. JSON, protocol buffers). Strip out unnecessary or redundant fields. Compress responses to minimise bandwidth usage. - Design for connection resilience: Support retry logic, especially using exponential backoff, to avoid overwhelming the server after temporary failures. Implement mechanisms for resumable transfers when a connection is interrupted. - Handle partial success & recovery: Allow clients to interpret partial responses and recover from interrupted interactions. Return clear, structured error messages to help the client respond appropriately. Mobile users expect speed, even when the network doesn’t cooperate. Smart API design helps makes that possible. If you want to add a mobile app into your business and want to speak, this is one of the areas we support out clients in, please get in touch.

Secure-by-design should be your default approach when developing custom systems. Taking a proactive, structured approach to security early in a project can significantly reduce risk, improve team velocity and prevent expensive rework later. Here are five practices that aren’t just technical checklists – they’re the foundation of resilient software delivery. 1. Establishing clear security requirements from the start Before beginning any development work, establish a clear understanding of the solution’s security objectives. Align on access, security measures, compliance, hosting, and deployment rules, so you avoid surprises, rework, and scope creep later. 2. Setting up a secure architecture Build security into your architecture from day one, don’t retrofit it later. For example, centralise authentication with Entra, integrate key services like Azure Key Vault early, assign group based access to resources rather than individual users and leverage Infrastructure as Code. 3. Leverage all available tools Developers cannot match the breadth of expertise available from dedicated security firms. Specialist knowledge can be accessed with a range of integrated tools e.g. Static application security testing, Dynamic application security testing, Software composition analysis and Secret scanning. 4. Collaborative oversight and transparency Every code change should be peer-reviewed and every new feature demonstrated internally. This process not only catches issues early but builds trust and fosters continuous improvement. 5. Logging and monitoring Even with diligent design and development, issues can still arise over time. When setting up a system, it’s important to have a mature logging strategy in place, so you can respond quickly. Some key factors to consider include real-time visibility, proactive alerting, compliance & auditing, centralised log management and continuous improvement. Security isn’t a single feature or a one-time checklist, it’s a mindset that should guide every phase of software development. By embedding secure practices from day one, teams not only reduce risk but also build faster, more confidently and with fewer costly surprises down the line. As threats continue to evolve, adopting a secure-by-design approach isn’t just best practice, it’s essential. #SoftwareDevelopment #DigitalTransformation

Having an internal team doesn’t mean you can’t benefit from a Digital Transformation partner. We (Steer73) help some of the world’s leading tech teams: • Accelerate the development of new systems, handing over to their internal teams once complete • Accelerate the development of new systems and then fully manage the system once complete • Supplement missing skillsets e.g. mobile app development If you're considering working with a digital transformation partner, or have a project blocked by lack of internal team resource, please get in touch, I'd love to chat. #DigitalTransformation #Software

ERPs are powerful, but they are not always the right solution when you have more specific business processes or innovation needs. Customising your ERP means: ⚠️ Vendor lock in ⚠️ Limited innovation ⚠️ Complex, less user-friendly solutions Custom solutions at Steer73 mean: ✅ No supplier lock in ✅ Complete freedom and flexibility to build what your business needs ✅ Best-in-class employee and customer experience We help companies go beyond their ERP.

Only about 15-40% of a software project is coding*. This is particularly important for those with internal teams. Do you have the other skillsets required to build, launch and manage successful digital products? • Business analysis & requirements gathering/prioritisation • Product design (UX/UI, architecture) • QA • Project management • Roll-out & training You don’t necessarily need to hire each of these roles internally but you do need these skills. * By the way, this figure sometimes surprises people, who point to established product teams and the role mixes you see in those. But those aren't software projects, they're established product teams. When businesses need to build software, they start with a software project, and the majority of the effort in that project is not the writing of code. #digitaltransformation #SoftwareEngineering

AI is quietly transforming construction. Although construction is still one of the least digitised industries (ref: McKinsey), firms like Kier, Balfour Beatty & Ferrovial are using AI to cut costs, reduce risk, and deliver projects more reliably. Here are four real-world applications of AI in the construction industry: 1️⃣ AI for Predictive Risk Management AI is being used to analyse historic data like incidents, project timelines, and cost estimates to predict potential risks before they become issues. This allows teams to take pre-emptive actions, such as reallocating resources or adjusting schedules. 2️⃣ AI-Powered Safety Image recognition technology is being used to analyse photos taken on site, identifying safety hazards such as workers not wearing the appropriate PPE, blocked emergency exits etc. The system flags these issues in real-time, allowing for immediate corrective action. 3️⃣ Admin Automation Construction projects involve significant documentation, from contracts to change orders to daily logs and inspection reports. AI tooling is perfect for unlocking huge efficiencies in the way teams interact with, complete and share this sort of documentation. We’ve seen systems automatically categorising and organising documents when they are uploaded, making it easier for teams to find the information they need quickly. Removing manual data entry and document management is another huge opportunity for increased efficiency. 4️⃣ Smarter Collaboration Communication breakdowns are a common cause of delays and errors in construction projects. Some construction firms are using AI-powered tools to analyse communication patterns and project data to identify potential misunderstandings or gaps in information that could lead to issues down the line. With the pace of AI development and the volume of consumer-facing tools coming to market, it can be hard to understand exactly how it should be applied to your business. If you ever want to explore ways to reduce costs, risk and inefficiency with AI-enhanced tooling, drop me a message. #digitaltransformation #construction #ai

Planning to modernise or migrate any digital systems in 2025? If so, migrating data will likely be critical to your project’s success. The cost of getting data migration wrong include: Financial costs Reputational damage Operational disruption Data integrity issues Legal and compliance issues Strategic impact When modernising systems we carefully analyse how to migrate data. Here are 7 data migration strategies from our Playbook that could help you plan 1️⃣ Parallel operation with one-way sync. Legacy and new systems run simultaneously, with data synced in one direction ✅ Reduces risk by keeping legacy systems operational during migration ✅ Provides fallback ❌ Syncing becomes difficult if legacy system data changes rapidly ❌ Requires additional resource (2 systems) 2️⃣ Parallel operation with bi-directional sync Legacy and new systems remain operational with data syncing both ways ✅ Ensures data consistency across both systems ✅ Enables flexible testing and phased user migration ❌ Can introduce data conflicts ❌ More complex to implement, requiring robust synchronisation protocols 3️⃣ Big Bang migration Data is migrated all at once, with the legacy system retired. ✅ Quick and efficient for straightforward datasets ✅ Reduces the operational overhead of running two systems concurrently ❌ High-risk approach as all data is migrated at once ❌ Leaves little room for error or rollback if issues arise 4️⃣ Big bang migration with data update monitoring A single-step migration followed by a monitoring phase to ensure data integrity ✅ Adds an extra layer of assurance by monitoring and correcting data post-migration ✅ Reduces the risk of data integrity issues compared to a pure big bang approach ❌ Monitoring requires additional time and resources post-migration ❌ Delays full system utilisation until monitoring is complete 5️⃣ Hybrid/combination approach Uses multiple migration techniques to accommodate specific needs. ✅ Flexible and customisable for complex migration needs ✅ Can tackle challenges incrementally with tailored techniques ❌ Can be resource-intensive to coordinate multiple approaches ❌ May prolong migration timelines if not well-managed 6️⃣ Phased migration Migrates data in stages, often based on functionality or priority ✅ Minimises disruption by migrating data in manageable chunks ✅ Allows for testing and adaptation at each stage of the migration ❌ Prolonged timelines can increase overall migration costs ❌ May complicate operations if parts of the business rely on both systems during the transition 7️⃣ On-demand access via legacy systems New systems access legacy data dynamically instead of full migration ✅ Avoids the complexity and risk of full data migration ✅ Useful for retaining access to legacy data that isn’t frequently required ❌ Legacy systems must remain operational, potentially increasing long-term costs ❌ On-demand access can create performance bottlenecks if not optimised For help making the right decision, get in touch. #DigitalTransformation #SoftwareDevelopment