The Forward-Deployed Engineer Vendor Column
Don't buy AI GTM tools on features—buy on who will actually get on the phone and deploy it with you.
- Difficulty
- Easy
- Time to result
- ~weeks to results
- Steps
- 4
- Confidence
- 90%
A vendor-selection rule for AI GTM agents: because every leading agent is 'more similar than different' under the hood and all require a month of training, the deciding factor is who will help you deploy. Add a column to your evaluation matrix for the forward-deployed engineer (FDE) and test it before signing. Use it whenever picking an AI sales/support vendor.
Origin
Lemkin explains SaaStr's two most successful early agents (Artisan and Qualified) won not on features but because they 'just did the work with us.' Rival vendors demanded $100k up front or were scared of bad PR. He generalizes this into adding an FDE column to the classic feature matrix.
Core principles
- 01Leading AI GTM agents are more alike than different under the hood — most run on the same underlying model and stitched APIs.
- 02Since every agent needs ~30 days of training, the world's best software with no deployment help is not what 99% of buyers should choose.
- 03The best vendors turn away business they can't make successful — that's a buy signal, not a red flag.
- 04Test the FDE relationship before you write the check, not after.
How to run it
- 1
Build the feature matrix as usual
Do the classic evaluation — compare features across leading vendors as you always would.
- 2
Add an FDE / solution-architect column
Add a column for who will actually help you deploy: the forward-deployed engineer or solution architect who ensures you get trained and onboarded for real.
Pro tip The FDE's number one job in this new world is to make YOU the customer a success, inverting the old 8-reps-to-1-SE ratio.
- 3
Get on the phone before you buy
Talk to the actual person who'll do your deployment. Confirm they'll roll up their sleeves. If a vendor with worse features will genuinely help and the better one won't, pick the one who helps.
Pro tip Favor vendors already close to you (e.g. a sponsor, a warm relationship) who offer to help the most.
Watch out If the better-featured vendor won't help you deploy, don't do it — you'll be in the 5% failure bucket.
- 4
Weight helpfulness over pure capability
Because training takes a month, deployment support outweighs marginal feature superiority for almost everyone.
In the wild
SaaStr didn't know if Artisan was best, but they were a sponsor and offered the most help. One rival demanded $100k up front; another was 'scared of SaaStr' and didn't want bad PR if it failed. Artisan and Qualified simply did the work alongside SaaStr.
→ Both became successful production agents; Qualified closed an inbound sponsor at 11pm Saturday.
Common mistakes
Choosing on features alone
The old matrix compares features, but with agents that all need training, ignoring deployment help lands you in the 5% who never go live.
Hiring an agency to do it and disappearing
Agencies don't know how to do this. The vendor's FDE plus your own hands-on work is what makes it succeed.
Is it for you?
Best for
Founders and GTM leaders selecting an AI sales, support, or SDR vendor
Not ideal for
Buyers with a full-time GTM engineer who genuinely wants to build the agent in-house
From the transcript
“You got to do another column which is your forward deployed engineer or solution architecture your SE and talk to them and say who is…”
“before you write a check get on the phone with Lenny and see if Lenny's really going to do deployment”
“The best ones turn away a lot of business today”
From the episode
We replaced our sales team with 20 AI agents—here’s what happened
Jason Lemkin (SaaStr)