Written by Demi
From employee spending cards and automated approvals to real-time reporting and accounting integrations, businesses have so much choice when it comes to managing where their money goes, with some broader spend management solutions even offering company payment cards, purchasing approvals, supplier payments and spending controls before money even leaves the business.
And with the features of these software apps growing, so has the market.
The global expense management software market is estimated to reach $8.45 billion this year, with forecasts suggesting it could grow up to $12.86 billion by 2030.
However, a growing market share doesn’t mean that every provider is getting an even spread of customer attention.
For expense management providers, being able to resonate with their potential customers is important, and understanding how businesses research their options, which competitors they’re coming across and where recommendations are being made is only going to help improve audience understanding.
Businesses are being asked to make their money stretch further, and finance leaders are feeling the pressure to rein it in as they contend with the rising cost of living, increases in software costs and National Insurance contributions.
Deloitte released an October 2026 CFO survey which found that 68% considered cost reduction a strong priority over the next year, up from 53% just one quarter earlier.
That being said, a lot of organisations are still dealing with processes that create more work than they need to, with 83% of UK CFOs believing that their spend management processes were more manual than they should be.
Businesses have clear reasons to explore automation, better spending controls and more efficient financial processes. However, investing in new software will contribute to those initial costs that their company is focusing on cutting down.
How businesses approach their research is going to depend on their immediate priorities, which won’t be the same across the board.
For some brands, it’s going to be about improving their visibility into spending across the business and having tighter controls to manage these expenses, while for others it might be about how well these platforms integrate with their existing tools, and some companies might just want to spend less time chasing employees for receipts.
Expense management software has traditionally been associated with submitting receipts, approving claims, and making sure employees get reimbursed. Those things are still core components, but the wider expectations placed on technology mean that expectations have stretched further than recording expenses.
93% of UK CFOs expected digital technology investment to increase over the following 12 months, and with efficiency being a key talking point as AI has become more widely adopted, it would make sense that businesses would be looking at time-consuming areas of their business operations as their main focus for improvement.
The sector has also evolved to suit a range of business needs, with expense management now covering elements like VAT and HMRC compliance tools, accounting integrations and configurable approval workflows.
Examples of additional needs that businesses could be looking for in expense management providers
As these platforms offer more capabilities, potential customers have more reasons to consider them beyond traditional expense tracking.
A business struggling with manual reconciliation might be looking for something very different from one trying to reduce spending across several departments, so if your marketing only focuses on saving time, you could be missing opportunities to reach businesses looking for other solutions.
When you look at it like that, businesses might be searching for a problem that these platforms solve rather than the category of expense management itself. This means providers have a much larger opportunity to reach customers at different stages of their research.
Traditional search engines have not been the only place businesses can research products and services for a long time. We’ve got platforms like YouTube, TikTok and Reddit with search capabilities, and in the accounting and auditing sector, we’ve seen a 500% increase in the share of website visits coming from AI chatbots between 2023 and 2025 (Similarweb).
Source: Similarweb
This is not to say that search engines don’t have their own place when it comes to research, but focusing solely on traditional search engines means you’re missing opportunities to reach customers elsewhere in their research journey.
In the same dataset, search engines accounted for just shy of 36% of website visits in 2023, falling to 30.26% in 2025. Search still accounts for a significant proportion of visits; AI hasn’t replaced that traffic, which is why it’s worth paying attention to other platforms that contribute to discovery.
Source: Similarweb
According to G2’s 2026 buyer research, more than 80% of surveyed B2B software buyers have used AI chatbots to source recommendations within the last 24 months.
This allows buyers to ask for specific recommendations tailored to their circumstances, compare different features with ease, and assess suitability through conversational responses. Using tools like ChatGPT means that buyers can be more specific in what they are looking for, rather than relying on established software terminology.
For expense management providers, that could mean buyers are asking much more specific solutions to the problems they’re trying to solve.
To explore how this affects AI recommendations, we used LLMLens to test questions that covered a range of business needs, including:
Examples of prompts explored in our LLMLens analysis
None of these prompts named a provider. They were written to reflect realistic business questions around specific problems and requirements.
Through multi-platform search, buyers could be introduced to brands they didn’t even know existed, and then carry out further research through website visits, review platforms, or third-party sources to assess those recommendations.
These platforms can play different but complementary roles throughout the discovery and evaluation process. Getting recommended is one thing, but brands still need to give potential customers enough information and build enough trust to take the next step.
We developed LLMLens to investigate how brands appear in AI-generated responses, using synthetic personas to model different customer requirements and test how AI platforms respond to relevant questions.
For the expense management market, we wanted to understand a few things:
Rather than looking at visibility as one overall figure, we wanted to understand where brands were appearing, where they were missing out, and how that changed as buyers became more specific about what they needed.
Using LLMLens, we analysed AI-generated responses across four commercial clusters: expense cards and spend management, expense management and business travel, accounts payable and vendor management, and cash management.
The categories we tracked using our tool, LLMLens
We tracked ten brands across these areas, testing questions that didn’t mention any providers by name. This allowed us to explore which brands AI platforms were recommending without directly prompting models towards specific businesses.
We also wanted to look beyond individual prompts and understand how recommendations changed throughout a buyer’s research journey.
We simulated five journeys, using our synthetic personas, which consisted of a finance manager, finance director, and head of operations, with different business requirements and priorities.
Rather than treating every question as a standalone prompt, we followed conversations so we could look at home recommendations developed as buyers moved towards a decision. This gave us two different ways to look at the market: which brands were appearing across different commercial opportunities, and how these recommendations changed once conversations developed.
Across all four AI models, one provider stood clearly ahead. Provider accounted for around one in five of all brand mentions we tracked, but behind them, the field was crowded. Providers B, C, and D each took between 13% and 14%, close enough that a small change in AI visibility could reorder them.
After this, visibility dropped sharply, with the most recommended provider mentioned more than six times as often as the least-recommended. That’s a big gap.
What makes this interesting is that while one provider has a clear lead, the next few brands are competing much more closely, and the rest of our analysis shows that overall position doesn’t necessarily tell us the full story.
A breakdown of AI-generated recommendations for ten expense management providers, showing each brands share of mentions
Provider A topped the rankings on every model, but by different margins. On one platform, the top three providers were only separated by 14 mentions, so a brand that looks dominant on one platform can be only slightly ahead of the pack on another.
For providers, this is why looking at one AI model in isolation could give you a very different impression of your competitive stance.
For five of the ten providers we tracked, appearances were concentrated in expense card and spend management questions, while most of the remaining mentions came from expense management and business travel prompts.
Accounts payable and cash management made up only 12% of mentions, with spend management providers often appearing as card data sources feeding into accounting software, but not as solutions to problems like daily cash positions and forecasting.
Many of these platforms offer invoice payments and cash visibility – but it’s just not reflected in AI models.
That creates an interesting gap between what providers can offer and what AI platforms are recommending them for. A business researching cash visibility or invoice payments could be presented with a different set of solutions when expense management providers have the relevant capabilities.
This is a clear reminder that being visible for your core product category doesn’t automatically mean you’re being recommended for everything your platform can do.
We followed Provider A through five simulated buyer journeys, with 35 turns in total. They appeared in 34.3% of them. They were part of the opening answer in three of the five journeys, but were absent from the final turn in every one.
Findings from Provider A’s AI search visibility analysis
The opening answer often had some influence on which providers remained a part of the conversation, but appearing early wasn’t enough to guarantee a place in the final recommendation. When requirements from the user become more specific, the brands being considered can easily change. For expense marketers, it’s important to understand where in the journey you are being surfaced, and how close this is to the user making that final decision.
This links back to the cost pressure finance leaders are under.
In the agency journey, AI didn’t compare actual pricing, but instead sorted providers into ‘per-seat’ and ‘no platform fee’ buckets, steering the buyer according to their wants. With 68% of CFOs prioritising cost reduction, being placed in the wrong pricing bucket in the opening response can rule a provider out before a buyer even visits the website.
It’s an interesting example of how AI’s interpretation of a provider and determine which options stay in the conversation. Even if you offer competitive pricing, if the AI model doesn’t gather that from your structured data or content, it’s not going to help you.
Our research using LLMLens has shown several areas worth looking into, but the right priorities will depend on where your brand is already appearing, which competitors you’re facing off against and the commercial opportunities that mean the most to your business.
Yes, LLMLens helps us find the patterns, but deciding what to do about them takes more than a set of AI recommendations, which is where our experience in SEO, PR, Content, Web Development, Paid Media, Creative and wider digital strategy comes in.
It’s great to make sure your product is showcased, but making sure you build trust and transparency in the financial services industry is essential.
For example, providers lost ground on the depth of integrations to their systems, like multi-entity coding, VAT splits, and ERP mapping. Having detailed guides and testimonials can help to give AI models something concrete to cite in their responses. If you also offer extra elements in your software like multi-currency visibility or invoice payments, make that clear in your website content, in a structured way that AI can easily source and understand.
We’d also suggest looking beyond your own website. If third-party sources are shaping how your brand is understood, it’s worth looking into whether they are reflecting your brand in an accurate way, or where stronger coverage or supporting evidence could help through the likes of PR activity.
Clear, published pricing information, including how costs change as more employees use the platforms or as businesses move between tiers, helps to keep you in the running for those cost-conscious buyers.
It’s not just about putting the price front and centre of the website, but about relaying to audiences what you get for that price, where additional fees might apply, and how the value of your platform compares to others on the market.
Getting into the first answer is only part of the job.
As buyers narrow down their options, they’re more likely to have detailed questions about integrations, implementation, pricing, support, and whether the platform is the right fit for their specific business needs. That means giving them access to comparison content, realistic implementation timelines, and relevant case studies or testimonials will give them everything they need to make the case to their internal team.
We’ve done a lot of looking into the finance industry, and know that in this new marketing landscape, being mentioned frequently isn’t the same as being recommended for the business requirements that matter most, or remaining part of the conversation as buyers remain closer to conversion.
