Nineteen years ago, a vendor’s PR team asked me to cover their news. I asked them questions instead — and then tracked the answers for years. This week it happened again.
This blog carries the following words as a badge of honor: “Truth Is Believing.” Accuracy Is Knowing.
In 2007, Ariba’s PR firm contacted me to discuss a milestone — a healthcare contract win — and to put me on the phone with a senior executive. I went in with an open mind. What I found was, in the language of the time, an infomercial: a self-congratulatory press release built on growth percentages with no point of reference by which the growth could be quantified or compared.
So I didn’t relay it. I asked the questions the announcement was designed to skip — what re-engineering to an on-demand model would actually mean, whether it would cannibalize the existing business, how customers would respond over the long run. Ariba couldn’t answer all of them on the call, and promised to send more later. That exchange became one of the most-read series this blog has ever run, and it set a posture I’ve kept for nineteen years: take the claim seriously enough to test it, give the vendor a fair chance to answer on the record, and then let time grade the result.
This week, Kate Wilson of Vertice contacted me about Vertice’s acquisition of Vendr — and asked, generously, for my thoughts if it was something I wanted to cover. Same posture as 2007. So I’m doing the same thing, nineteen years later.
(2007) The Ariba Interviews
The original arc, for anyone who wants to see how the method works over time:
The value of the series was never the original post. It was the willingness to keep checking — to let the record, not the press release, have the last word.
(2026) The Vertice Interviews
The facts, as stated by Vertice: the company has acquired Vendr, and describes the combination as creating the world’s largest procurement intelligence dataset — by their figures, more than $75 billion in indirect spend, over two million pricing data points, 250,000-plus negotiated contracts, and 32,000-plus vendors — to power autonomous AI negotiation through its agent, Ana. These are the company’s own launch-day numbers, self-reported, and I’m treating them as claims to be examined rather than results to be repeated.
I won’t write up the announcement. I’ll test the structural question underneath it. And in keeping with the method, I’ve put the following questions to Vertice and invited them to answer on the record. Their responses — in their own words — will form the follow-up piece.
1. The headline is the world’s largest procurement intelligence dataset. The question isn’t whether it’s larger — it plainly is. It’s whether larger intelligence resolves the conditions that determine whether a procurement decision succeeds once it meets a real operating environment. What in the combined platform speaks to those conditions, rather than to the volume of pricing and negotiation data?
2. Ana negotiates autonomously on the buyer’s behalf. Does autonomous negotiation resolve a misaligned operating model — or does it execute the buyer’s existing, unvalidated assumptions faster and at greater scale? What stops a confident wrong assumption from being applied across any renewal?
3. A 1998 engagement I worked had all the data it believed it needed to automate, and delivery still sat at 51% against a 90% requirement. The variable that moved it to 97.3% wasn’t in any dataset — it was the off-the-map discovery that the time of day orders came in to procurement from the service department governed customs windows and courier schedules downstream. The data didn’t contain it because the data didn’t know it mattered. So: how does more procurement intelligence equip an organization to identify the misalignment that no dataset contains — before the data, not after?
4. You describe a compounding intelligence flywheel — every negotiation feeding the next. That compounds pricing and vendor intelligence. But how do you track vendor performance in the areas beyond price — product quality, timely delivery, and ultimate impact on your client’s end customer — and what compounds the buyer’s readiness to act on all of it? Price is the one variable that lives in the data by default. The ones that determine whether the decision was actually good tend not to.
As soon as I receive Vertice’s answers, I’ll write the follow-up. If the team takes the questions on, they’ll get a full and fair hearing in their own words. If the honest answer to some of them is “that’s outside what the platform addresses,” that is a legitimate and useful answer too.
Either way, the method is the same one it was in 2007. Take the claim seriously. Ask the question the announcement is built to skip. Give the vendor the floor. And let accuracy — measured over time — have the last word.
To be continued.
Jon Hansen is the founder of Hansen Models™ and publisher of Procurement Insights, an independent research practice operating on a nineteen-year archive with zero vendor sponsorships.
-30-
Related
The New Ariba Interviews Are the Vertice Interviews
Posted on June 4, 2026
0
Nineteen years ago, a vendor’s PR team asked me to cover their news. I asked them questions instead — and then tracked the answers for years. This week it happened again.
This blog carries the following words as a badge of honor: “Truth Is Believing.” Accuracy Is Knowing.
In 2007, Ariba’s PR firm contacted me to discuss a milestone — a healthcare contract win — and to put me on the phone with a senior executive. I went in with an open mind. What I found was, in the language of the time, an infomercial: a self-congratulatory press release built on growth percentages with no point of reference by which the growth could be quantified or compared.
So I didn’t relay it. I asked the questions the announcement was designed to skip — what re-engineering to an on-demand model would actually mean, whether it would cannibalize the existing business, how customers would respond over the long run. Ariba couldn’t answer all of them on the call, and promised to send more later. That exchange became one of the most-read series this blog has ever run, and it set a posture I’ve kept for nineteen years: take the claim seriously enough to test it, give the vendor a fair chance to answer on the record, and then let time grade the result.
This week, Kate Wilson of Vertice contacted me about Vertice’s acquisition of Vendr — and asked, generously, for my thoughts if it was something I wanted to cover. Same posture as 2007. So I’m doing the same thing, nineteen years later.
(2007) The Ariba Interviews
The original arc, for anyone who wants to see how the method works over time:
The value of the series was never the original post. It was the willingness to keep checking — to let the record, not the press release, have the last word.
(2026) The Vertice Interviews
The facts, as stated by Vertice: the company has acquired Vendr, and describes the combination as creating the world’s largest procurement intelligence dataset — by their figures, more than $75 billion in indirect spend, over two million pricing data points, 250,000-plus negotiated contracts, and 32,000-plus vendors — to power autonomous AI negotiation through its agent, Ana. These are the company’s own launch-day numbers, self-reported, and I’m treating them as claims to be examined rather than results to be repeated.
I won’t write up the announcement. I’ll test the structural question underneath it. And in keeping with the method, I’ve put the following questions to Vertice and invited them to answer on the record. Their responses — in their own words — will form the follow-up piece.
1. The headline is the world’s largest procurement intelligence dataset. The question isn’t whether it’s larger — it plainly is. It’s whether larger intelligence resolves the conditions that determine whether a procurement decision succeeds once it meets a real operating environment. What in the combined platform speaks to those conditions, rather than to the volume of pricing and negotiation data?
2. Ana negotiates autonomously on the buyer’s behalf. Does autonomous negotiation resolve a misaligned operating model — or does it execute the buyer’s existing, unvalidated assumptions faster and at greater scale? What stops a confident wrong assumption from being applied across any renewal?
3. A 1998 engagement I worked had all the data it believed it needed to automate, and delivery still sat at 51% against a 90% requirement. The variable that moved it to 97.3% wasn’t in any dataset — it was the off-the-map discovery that the time of day orders came in to procurement from the service department governed customs windows and courier schedules downstream. The data didn’t contain it because the data didn’t know it mattered. So: how does more procurement intelligence equip an organization to identify the misalignment that no dataset contains — before the data, not after?
4. You describe a compounding intelligence flywheel — every negotiation feeding the next. That compounds pricing and vendor intelligence. But how do you track vendor performance in the areas beyond price — product quality, timely delivery, and ultimate impact on your client’s end customer — and what compounds the buyer’s readiness to act on all of it? Price is the one variable that lives in the data by default. The ones that determine whether the decision was actually good tend not to.
As soon as I receive Vertice’s answers, I’ll write the follow-up. If the team takes the questions on, they’ll get a full and fair hearing in their own words. If the honest answer to some of them is “that’s outside what the platform addresses,” that is a legitimate and useful answer too.
Either way, the method is the same one it was in 2007. Take the claim seriously. Ask the question the announcement is built to skip. Give the vendor the floor. And let accuracy — measured over time — have the last word.
To be continued.
Jon Hansen is the founder of Hansen Models™ and publisher of Procurement Insights, an independent research practice operating on a nineteen-year archive with zero vendor sponsorships.
-30-
Share this:
Like this:
Related