TL;DR: Key Takeaways
Sales teams are increasingly pairing ChatGPT, Google Maps, and spreadsheets into a DIY prospecting stack - but this combination creates four critical coverage gaps that prevent pipeline growth.
ChatGPT compresses research time but operates on public data only. It can't access gated property information like permit history, building ownership, or verified decision-maker contacts.
In commercial services, the building is the lead, not always the company. Free tools can't tell you a building's age, square footage, recent permits, when a tenant leaves, or whether the facilities director changed two weeks ago.
The hidden cost isn't the tools themselves. It's the 10+ hours per rep per week spent being the integration layer between disconnected platforms - time that should be spent selling.
A complete prospecting workflow connects property data, buyer intent signals, verified contacts, and pipeline tracking in one system, so reps go from research to outreach in minutes instead of hours.
The Cost of Tool Switching
Your reps aren't slacking. Far from it…that's what makes this so frustrating.
They're researching prospects in ChatGPT. They're pulling up buildings on Google Maps and scanning neighborhoods from their truck. They're logging everything in a spreadsheet - names, numbers, notes, next steps - and updating it after every call.
It looks like a modern sales workflow. It feels productive. And when you ask how prospecting is going, they have an answer: they've been busy.
But “busy” doesn’t build a sales pipeline. And somewhere between research and revenue, something keeps falling through the cracks.
The challenge isn't effort. It's coverage.
Your team is working with tools that each do one thing well - but none of them can see the full picture.
They can't tell your rep which building in their territory just pulled a mechanical permit. They can't surface a verified direct line for the facilities director who manages that building. They can't show whether that account has been actively researching HVAC replacement services for the last three weeks.
Those gaps don't show up in ChatGPT or on a spreadsheet. But they show up in your pipeline review.
Taj Shaw, Customer Success Manager at Convex, has spent five years working with commercial services sales teams as they adopt new prospecting workflows. The pattern she describes is specific - and familiar.
"Most of the customers are pairing, like, ChatGPT with a Google Maps and an Excel spreadsheet," Taj says. "Where Convex can give you all of that in one place."
This article isn't about replacing the tools your team already uses. It's about what those tools can't see - and how those blind spots quietly limit your territory coverage, your outreach quality, and your pipeline velocity.
34% selling time: Salespeople spend only 34% of their time actually selling - the rest goes to research, admin, and tool-switching. (HubSpot, 2025 Sales Trends Report)
28% of the day selling: Despite record tech spending, modern reps still spend only 28% of their day in active selling activities. (Nimble / industry benchmarks, 2026)
73% tool overlap: 73% of sales teams report tool overlap in their tech stack, wasting an estimated $2,340 per rep per year. (Prospeo, 2026 Sales Tech Stack Guide)
43% quota attainment: Quota attainment fell to 43% in 2025, down from 52% the previous year. (Salesforce, State of Sales, 2025)
94% of cold emails unanswered: Approximately 94% of cold emails go unanswered - volume alone doesn't solve the pipeline problem. (Belkins, 2025 Cold Email Benchmarks)
What Does a DIY Sales Prospecting Stack Actually Look Like?
A growing number of commercial services sales teams are building their prospecting workflow from three free tools: ChatGPT for research and email drafts, Google Maps for territory visualization, and a spreadsheet for pipeline tracking.
This approach includes AI (a trendy topic right now) which feels modern and productive - but it creates structural gaps in your data and sales workflow.
The workflow usually starts the same way. A rep opens Google Maps, scans a zip code or drives a route, and spots buildings that look like they could be a fit. Maybe it's a warehouse complex off the interstate. Maybe it's a row of medical offices near a hospital campus. They note addresses.
Then they move to ChatGPT. They paste in the company name or address and ask for background - who owns it, what they do, who might handle facilities decisions. ChatGPT returns a list. It might be right. It might be six months out of date. There's no way to verify without another 10 minutes of digging.
Then they update the spreadsheet. Address, contact name (if they found one), a note about the building, and a column for "next step."
They'll come back to it tomorrow. Or next week. Or they won't - because 15 new buildings caught their eye and the list keeps growing while outreach activities dwindle.
This is what resourceful reps do when they don't have a purpose-built system. And it works. To a point.
The challenge is that each tool operates in isolation. ChatGPT doesn't know what Google Maps is showing. The spreadsheet doesn't know what ChatGPT researched. And none of them know which buildings in the rep's territory actually need their services right now.
Taj sees this play out in customer success conversations regularly.
Customers arrive already stitching these tools together, and her team has to reframe expectations - not because the tools are bad, but because they're missing critical layers.
"They want to see us (Convex) as a data tool only," she explains, "versus adapting to the workflows that can help their management a little better."
The illusion is that three tools covering three functions equals a complete workflow. The reality is that the spaces between those tools are where the pipeline opportunities disappear.
Definitions
Property Intelligence: A data layer that consolidates building attributes, permit history, ownership records, tenant information, and decision-maker contacts into a single searchable platform covering millions of commercial properties. Unlike general contact databases, property intelligence is organized around the building - not the company.
Buyer Intent Signals: Data indicators showing that a specific property or account is actively researching services like yours. Intent signals reveal who needs you right now - not just who exists in your territory.
DIY Prospecting Stack: An informal combination of free or consumer-grade tools - typically ChatGPT for research, Google Maps for territory visualization, and Excel or Google Sheets for tracking - used by sales teams as a substitute for a purpose-built prospecting platform.
Can You Actually Use ChatGPT for Sales Prospecting?
Yes - ChatGPT compresses prospect research from 30 minutes to a few minutes and drafts competent outreach emails.
But it operates entirely on publicly available, non-gated data. It cannot access gated information about buildings, permits, verified contacts, or buyer-intent signals that commercial services teams need to prioritize and personalize their outreach.
Give ChatGPT credit. It's remarkable for certain prospecting tasks.
Ask it to research a company's background, summarize their services, identify likely decision-maker titles, or draft an email - and it returns something useful in seconds.
For reps who are new to prospecting or teams trying to scale outreach without adding headcount, it's a real time-saver.
But there are three places where it falls apart for commercial services.
It doesn't know the building. In commercial services, the building is the lead. A 150,000-square-foot industrial facility with a chiller permit from 2009 is a fundamentally different sales opportunity than a 15,000-square-foot strip mall.
ChatGPT can tell you about the company that occupies that building. It most likely can't tell you the building's age, square footage, recent permit activity, ownership structure, or tenant makeup - unless that information is publicly available. And without those details, your rep's AI-drafted email references the company - not the property.
That's the difference between outreach that sounds relevant and outreach that sounds generic.
It fabricates contacts. David Vroblesky, Principle Product Manager at Convex and ServiceTitan, puts it directly: "There's a lot of proprietary data that ChatGPT is not aware of - whether that's people's email addresses and direct phone lines that we get through proprietary data sources, or its permit applications that have been filed in local municipalities."
Ask ChatGPT for a facilities manager's email at a specific building, and it will often return a confidently formatted guess (example: firstname.lastname@company.com) that bounces because it was never verified. Or worse, the contact left that role three months ago and now your emails sent to a general inbox that’ll never get a reply.
It can't prioritize timing. ChatGPT treats every prospect the same because it has no access to buyer intent signals. It can't distinguish between a building where the property management group is actively researching janitorial services this week and one that renewed its cleaning contract six months ago.
Without intent data, every call your rep makes is a cold call - regardless of how personalized it sounds.
None of this makes ChatGPT a bad tool. It makes it an incomplete one. And in a market where 94% of cold emails go unanswered, the difference between a researched email and a truly relevant one is the difference between noise and a booked meeting.
Where Does the Free Tool Stack Fall Apart?
Free tools create four critical coverage gaps for commercial services sales teams:
No buyer intent signals to prioritize outreach timing,
No gated property data to personalize messaging,
No verified contacts at the building level,
No unified pipeline workflow to track and manage deals.
Gap 1: No Buyer Intent Signals
Google Maps shows you where buildings are. It doesn't show you which ones need you.
Your rep can scan a territory and identify 200 commercial properties in their zip codes. But without intent signals, they're treating a hospital that's actively searching for fire safety inspection services the same as one that renewed its contract last quarter.
They're spending equal time on a distribution center that just filed an HVAC permit and on one with no activity.
Buyer intent data solves the “prioritization problem” that effort alone can't fix. It surfaces the accounts that are showing research behavior right now - so your rep's first call of the day goes to the warmest opportunity, not the next name on a list sorted alphabetically.
Without this layer, your team is prospecting blind. They're covering geography without covering the market.
Gap 2: No Gated Property Data
ChatGPT can research a company. It struggles to research a building.
Permit history, square footage, building age, ownership structure, tenant makeup, equipment install age - this data exists, but it's gated. It lives in municipal databases, county records, and proprietary data sources that require aggregation and verification.
ChatGPT doesn't have access to it. Most of the time, neither does Google.
This matters because, in commercial services, the property details are what make outreach specific and personalized enough to earn a response.
Consider the difference between these two emails:
"Hi Sarah, I noticed your company manages several properties in the metro area. We specialize in commercial HVAC services and would love to connect."
Versus:
"Hi Sarah, I saw the mechanical permit filed on your 180,000-square-foot facility on Industrial Parkway - looks like the rooftop units were installed in 2011. Most systems that age are approaching the point where maintenance costs start exceeding replacement value. Worth a quick conversation next week?"
The first one can be written by anyone with a name and a ChatGPT prompt. The second email can only be written when you have specific insights that property intelligence can surface.
Gap 3: No Verified Decision-Maker Contacts
This is the gap that stalls the most deals. Your rep has researched the building. They've drafted the email. Now they need to send it to the right person - and the free tool stack can't tell them who that is.
Taj hears this consistently in customer onboarding conversations. The thing that trips teams up most? "When they can't find the contact that they're looking for - like, the decision maker," she says.
ChatGPT will suggest titles. LinkedIn will show you names. Google will give you the main office number. But none of these provide a verified direct dial or email for the specific facilities director, property manager, or operations lead who handles decisions at that building.
Terri Reddan, Director of Operations at Stratus Building Solutions in Pittsburgh, put it bluntly: "The hardest part of this job is getting the person who's in charge of cleaning…”
If you don’t have access to that person, it’s a deal breaker.
In her own words, Terri’s franchise had been paying third-party appointment vendors "hundreds of dollars a lead, and it might not even be decent.”
The leads often came with the wrong contact or no contact at all - just a building address and a phone number that rang the front desk.
Within four months of switching to Convex, her team used the verified decision-maker contacts tied to properties in her to generate over $125,000 in annual revenue directly attributable to the platform - while reducing cost per lead compared to the third-party providers.
The math Terri described is the same math every commercial services rep faces. If you can get to the right person, you can have a real conversation. If you can't, you're calling the main line and hoping someone transfers you.
Gap 4: No Unified Workflow or Pipeline
A spreadsheet tracks a list. It doesn't actively track a deal.
Spreadsheets can’t set follow-up reminders. They don't log call outcomes automatically. They can’t show your sales manager which accounts are progressing and which have gone cold. And, unless noted, they can't tell you that a rep promised to call back on Thursday and didn't.
David Vroblesky sees this gap clearly from the product side: "It's great if ChatGPT gives you a list of 5 people to call, but you're going to get voicemails for 3 of those people. How are you going to remember to follow up again 3 days from now? Because it might take 3 to 6 interactions before you actually get a booked meeting."
When your workflow is spread across three disconnected tools, your reps become the integration layer.
They're the one copy-pasting data between tabs, manually logging activity, updating notes by hand, and trying to remember which accounts need follow-up.
That's not a system. It's a workaround. And workarounds break under volume.
The workflow that was supposed to save them time is taking more time, and tasks get forgotten along the way - because there's no pipeline structure to move them from research into action.
What's the Real Cost of a Stitched-Together Workflow?
The hidden cost of a DIY prospecting stack isn't the tools - it's the hours your reps spend being the integration layer between them.
At a fully loaded cost of $75 per hour, a team of ten reps can lose over $15,000 per month on research and admin activities that a unified prospecting platform eliminates.
Work through the math for your own team.
If a rep spends 25 minutes per prospect researching across ChatGPT, Google Maps, and a spreadsheet - pulling building details, hunting for contacts, drafting outreach, and logging everything manually - and they're targeting 20 new prospects per week, that's over 8 hours of pure research time. Per rep. Per week.
That's a full day every week spent building a list that a property intelligence platform delivers in minutes.
Multiply that across a team of ten, and you're looking at 80 hours per week in research labor. At $75/hour fully loaded, that's $6,000 a week. Over $24,000 per month. On list-building and data entry.
Now layer in the accuracy problem.
A significant portion of those prospects aren't qualified because the rep didn't have intent signals to filter by. They looked at a building, it seemed like a fit, and they added it to the spreadsheet. No way to know whether that property actually needs their services. No way to know whether the contact they found is still at that company. No way to prioritize their finite calling hours toward the opportunities most likely to convert.
The Stratus franchise in Pittsburgh was spending hundreds of dollars per lead through third-party appointment centers - and many of those leads arrived with the wrong contact or a mismatch between the lead value and the building's cleanable square footage.
After switching to a platform with property data and verified contacts, they not only generated $125,000 in new annual revenue in four months but also substantially reduced the cost per lead.
That's the trade-off most sales leaders haven't fully calculated. The free tools are free. But the time your team spends compensating for what those tools can't do is the most expensive line item in your sales workflow - it just doesn't show up on any invoice.
"Most of the customers are pairing, like, ChatGPT with a Google Maps and an Excel spreadsheet, where Convex can give you all of that in one place." - Taj Shaw, Manager of Customer Success, Convex
What Does a Complete Sales Prospecting Workflow Actually Include?
A complete commercial services prospecting workflow connects five capabilities in one system:
Territory visualization layered with property data,
Buyer intent signals that prioritize outreach timing,
Verified decision-maker contacts at the building level,
Pipeline tracking with follow-up management
Accessible from any device, in the office or the field.
Consider what a Tuesday morning looks like with a unified workflow versus the DIY stack.
With the DIY stack: Your rep opens Google Maps, drives a territory, and spots three buildings that look promising. Back in the truck, they open ChatGPT and research each one. They find a company name for the first building, but no decision-maker contact. They find a LinkedIn profile for the second, but the person's title doesn't quite match - is she the facilities director or the office manager? The third building shows nothing useful.
An hour has passed and they've made zero calls.
With a complete prospecting platform: Your rep logs into Convex, opens the map view, and immediately sees which properties in their territory are showing intent signals.
A 200,000-square-foot distribution center in their zip code has been actively researching mechanical services.
They click into the property. Building age: 2007. Last permit filed: rooftop unit replacement, 2019. Ownership: managed by a regional property group. Verified contact: the facilities director, direct phone number, email address, and maybe even a LinkedIn profile.
They click "Draft Email," and the platform drafts outreach that references the building's permit history, age, and likely service needs. They review it, write an additional line, hit send, and set a follow-up for Thursday.
Five minutes. One qualified prospect. One warm conversation started.
That's the gap a DIY stack can't close. Not because the individual tools are bad - but because the workflow requires property data, intent signals, verified contacts, and pipeline tracking to operate as a single system, not four separate activities stitched together by a rep who's also supposed to be selling.
Taj sees the shift happen when customers move from treating Convex as a data source to actually working inside the workflow. "When they understand the value of the workflows… and functionality that goes way beyond just surfacing a contact," she says, " that's when they can really see the value in the platform."
The workflow doesn't replace the rep's judgment. It replaces the hours of manual research that sit between the rep and the conversation they should be having.
How Do You Know When Your Team Has Outgrown Free Tools?
There are three signals that tell you your team's DIY prospecting stack has hit its ceiling: reps are active, but pipeline isn't growing proportionally, research time per prospect exceeds 20 minutes, and you, as a manager can't see the team's prospecting activity and deal progression in one place.
Signal 1: Activity without pipeline. Your reps are logging hours. They're researching. They're making calls. But when you sit down for a pipeline review, the numbers don't match the effort.
This usually means the research is consuming the time that should be going to conversations - or the prospects they're reaching out to aren't qualified because there's no intent data to filter by.
If quota attainment is falling while activity stays steady, the problem isn't effort. It's the workflow.
Signal 2: Research eats selling time. HubSpot's 2025 research found that the average salesperson spends only 34% of their time actually selling.
For reps using a DIY stack, that number is often worse - because every new prospect requires a multi-tool research cycle before a single call gets made.
If your reps are spending more time finding people to call than actually calling, the tools aren't scaling with them.
Signal 3: No manager visibility. You can't coach what you can't see. If your only view into prospecting activity is a shared spreadsheet that is inconsistently updated, you don't know which reps are actively prospecting, which accounts are in play, which follow-ups are overdue, or where deals are stalling.
Thefield sales teams that are moving away from disconnected tools are doing it for this reason - management needs pipeline visibility that a spreadsheet can't provide.
If your team is resourceful enough to have built a DIY stack from ChatGPT, Google Maps, and a spreadsheet, they're resourceful enough to do more with a system that connects the pieces they're currently stitching together by hand.
The question isn't whether free tools got them started. It's whether those tools can get them where they need to go next.
In Summary
The three-tool stack your reps built isn't a problem. It's a sign of initiative. They found free tools that each handle a piece of prospecting - research, territory views, tracking - and they made it work.
But at some point, the gaps between those tools start costing more than the tools save. No intent signals means every call is cold. No property data means outreach stays generic. No verified contacts means your best email lands in the wrong inbox. No unified pipeline means follow-ups slip, deals stall, and your only visibility into what's happening is a spreadsheet that's already out of date.
Commercial services prospecting doesn't look like standard B2B outbound. The building is the lead. The facilities director - not the CEO - is the buyer. The permit history, building age, and ownership structure are what make outreach relevant. And those are exactly the data layers that free tools can't access.
If your team has outgrown the DIY stack, Convex brings everything they're currently stitching together into one workflow - property intelligence, buyer intent signals, verified decision-maker contacts, Generative AI outreach, and a full CRM - built specifically for commercial services sales teams.
FAQ
Can ChatGPT replace a CRM for sales prospecting?
No. ChatGPT is a research and drafting tool - it compresses the time spent gathering information and writing outreach. But it can't track deals, set follow-up reminders, log call outcomes, or provide pipeline visibility for managers. Sales teams need a separate system to manage the prospecting workflow from first touch through booked meetings.
What are the limitations of using ChatGPT for B2B sales?
ChatGPT operates on publicly available data. It cannot access gated information like permit records, verified email addresses, direct phone numbers, buyer intent signals, or property-specific details. It also fabricates contact information with confidence, guessing email formats that often bounce, which can damage a company’s domain reputation over time.
Is Google Maps useful for field sales prospecting?
Google Maps shows where buildings are, which is genuinely useful for territory visualization and route planning. But it provides no data about what's inside those buildings: building age, tenant information, ownership, permit history, or whether the property is actively looking for services. It covers geography without covering the market.
What is buyer intent data, and why does it matter for prospecting?
Buyer intent data reveals which accounts are actively researching services in your category right now. Instead of treating every property in your territory equally, intent signals let reps prioritize outreach to accounts showing real buying behavior - which converts at significantly higher rates than cold outreach to static lists.
How do commercial services teams find decision-maker contacts?
The most common methods are LinkedIn searches, Google, and Hunter.io or similar email-finder tools. These approaches return unverified results that may be outdated or inaccurate. Purpose-built platforms aggregate verified contacts - facilities directors, property managers, operations managers - tied directly to specific commercial properties, with direct phone numbers and email addresses.
What's the difference between a free sales tool and a sales intelligence platform?
Free tools handle individual tasks: research, mapping, tracking. A sales intelligence platform connects property data, contact information, intent signals, outreach tools, and pipeline management into a single workflow. The distinction isn't just features - it's whether the system eliminates the manual work between steps or forces the rep to do it themselves.
How much time do sales reps spend on prospecting research?
Industry benchmarks show reps spend 15–30 minutes per prospect on manual research, and that 66% or more of their total work time goes to non-selling activities. For commercial services teams using disconnected tools, the per-prospect research time often runs higher because building-level data requires multiple sources to compile.
What is property intelligence for commercial services?
Property intelligence combines building data - square footage, age, usage type, permit history - with ownership records, tenant information, and verified decision-maker contacts in a searchable platform. For commercial services sales teams, it replaces the multi-tool research cycle with a single view of every relevant property in their territory, including signals showing which ones are most likely to need services now.
Related Reading
AI in Sales Workflows: The Truth About What Works for Commercial Services
Beyond the Gatekeeper: Sales Intelligence Strategies for Reaching Facility Managers
Declining Cold Outreach Effectiveness: What Commercial Services Teams Need to Know
Prospecting Software vs. AI Tools: Why Sales Intelligence Beats Guesswork
Share
