94% of B2B buyers now use AI to research purchases. The shortlist is made before your rep gets the first call. A CRM upgrade doesn't fix that. Five moves that pay off today and position you for what's coming.
Every PE value creation plan for a commercial-stage company includes the same line item: modernize the sales tech stack. Swap the CRM. Add intent data. Deploy AI copilots for reps. Layer in conversation intelligence.
It's a familiar workstream that feels like progress and usually is. The old CRM was a mess. The data was dirty. Reps were logging calls in spreadsheets. Fixing that is real and necessary.
But it's solving the seller's problem while the buyer's world changes underneath you.
The Buyer Moved First
Forrester's January 2026 State of Business Buying report, drawn from over 18,000 buyer responses, found that 94% of B2B buyers now use generative AI as a primary research tool, up from 89% twelve months earlier.¹ More important: 81% immediately validate AI recommendations with trusted humans, whether peers, analysts, or influencers. The buying committee has exploded. The average deal now involves 13 internal and 9 external stakeholders. For anything involving AI or new technology, that number doubles. Procurement sits at the table from day one in 53% of deals.
6sense's 2025 study of nearly 4,000 B2B buyers found that buyers shortlist roughly four out of five vendors on Day One, and 95% of the time the winner comes from that initial shortlist.² The decision is largely made before a seller is contacted. G2's 2025 survey of 1,169 decision-makers confirmed that AI chatbots are now the number one source influencing vendor shortlists, ahead of review sites, vendor websites, analyst firms, and salespeople.
Google's research across 2,063 senior US buyers found that 58% who made a B2B purchase in the past six months also switched vendors during that period. Three-quarters complete the buying journey in 12 weeks or less. When buyers can evaluate alternatives in an afternoon using AI, brand familiarity and rep relationships don't protect you the way they used to.
A new CRM doesn't fix this. Conversation intelligence on calls you never get doesn't fix it either.
The Uncomfortable Audit
Most stack overhauls start with tools. The harder question is one layer beneath: what does your sales team actually do all day, and how much of it is information logistics that AI now does better and cheaper?
The activities that actually require a human (complex negotiation, consultative discovery, trust earned over years) are under more pressure because the validation committee is bigger and harder to satisfy. Everything else (spec questions, pricing lookups, follow-up decks, CRM updates) is overhead you're paying $150K-$300K per head to maintain.
A stack overhaul that doesn't start with this audit just makes your reps faster at doing things that shouldn't require a rep.
Five Moves That Pay Off Now and Position You for What's Coming
None of this requires a bet on timing. Every move here pays off in the next quarter. They just also happen to be the same moves that prepare you for what's coming.
Structure your product data like your revenue depends on it. It does.
Your product specs, competitive differentiators, and pricing logic live in PDFs, in your senior reps' heads, in a SharePoint folder nobody can find, and in a pitch deck from 2023 that's "mostly still accurate." Reps waste hours assembling this for every deal. Channel partners get it wrong constantly.
Structuring this data into machine-readable, centrally governed formats makes your reps faster and your channel partners more accurate. It also makes your products discoverable by the AI tools now shaping every shortlist. If your product data isn't structured in a way those tools can parse and recommend, you're invisible in the fastest-growing research channel in B2B. That's not a 2028 problem.
Separate your reps from your information.
Once product data is structured and accessible, you can see which reps add value through expertise and judgment and which ones are functioning as a delivery mechanism for information the buyer could get elsewhere. A rep who spends 60% of their time answering spec questions and 40% on consultative selling is a misallocated resource. Flip that ratio by making information self-serve and you get more selling capacity without adding headcount.
Buyers are 95% locked into their shortlist before your rep picks up the phone. The only way to influence that list is to be present in the research phase with structured, findable information.
Stress-test your pricing against transparency.
If an algorithm compared your pricing to three competitors right now, across every product, every segment, every channel, do you win or lose? Most companies have never asked this because the friction of manual comparison protected them.
That friction is gone. Google's research shows 58% of B2B buyers switched vendors in the past six months. When buyers can benchmark your pricing against alternatives in an afternoon, relationships and inertia aren't worth what they used to be. Cleaning up your pricing architecture improves win rates today and builds a structure that survives algorithmic scrutiny. This doesn't mean lowering prices. It means knowing where your pricing reflects genuine value and where it reflects the buyer's inability to compare.
Build API-first commerce infrastructure.
Your buyers' systems want to talk to your systems. This started with EDI in the 1980s. It continued with e-procurement platforms and punchout catalogs. The current version is API-enabled quoting, availability, and ordering. The next version is autonomous agents calling those APIs without a human in the loop.
A clean API for real-time quoting improves your e-commerce conversion today. When agents start handling routine purchasing, the companies with that infrastructure will be in the consideration set. The ones that require a phone call to get a quote won't.
"Buy from yourself" using AI
Open ChatGPT, Perplexity, or whatever your buyers are using. Ask it to recommend a solution in your category for a specific use case. See if you show up. See what it says about you versus competitors. Then try to get a quote from your own company without calling a human. Most companies have never done this. The results are usually brutal. It exposes every gap in one sitting: missing product data, invisible pricing, broken digital paths, competitors who show up better. It's the fastest diagnostic you can run and it costs nothing.
The Value Creation Plan That Actually Creates Value
The safe PE playbook (clean CRM, AI copilots for reps, better pipeline reporting) will deliver efficiency gains in two quarters. It's not wrong. It's incomplete.
Buyer journeys increasingly begin inside an LLM and get validated by committees that are larger, more informed, and harder to reach than they were two years ago. Every move you make to serve that process (structured data, transparent pricing, API commerce) also makes you better at selling to humans today.
Do both and the future-proofing work becomes the highest-ROI line item in the plan. Stop at the CRM refresh and you'll have the cleanest pipeline in the industry, of deals that are being decided somewhere else.