The challenge
MFT is a crowded category with well-known incumbents. Buyers asking ChatGPT, Perplexity, or Google "what's the best MFT solution?" mostly heard someone else's name.
The insight
The data split two things that usually get lumped together. AI engines already cited our pages often: they trusted the content. But they named competitors more often.
"We're better cited. They're better known."
That changed the plan. Content quality wasn't the bottleneck. Access, coverage, and name recognition were.
What I did
Fixed access
I fixed the 403 errors that were blocking AI crawlers, turned on instant Bing crawl notifications, and added machine-readable "last updated" dates to every blog post.
Shipped the missing pages
New product and solution pages, comparison pages against Fortra, MOVEit, and Axway, and pages for the category questions where we were invisible.
Built authority off-site
The G2 report, eBooks, customer stories, and manufacturing content, plus a PR push to position the company as a leader in enterprise MFT, not just "security-first" MFT, so AI connects the brand to the whole category.
Made it a loop, not a launch
A weekly report tracks citation share, citation rate, and mention rate across eight AI engines, plus organic search and on-site conversion. Each month I review it with the product owner and leave with the top three actions. The same loop is now being copied to the MDSS and ICAP product lines.
Swipe sideways to inspect the full diagram →
The results
April baseline to late summer 2026
AI referral traffic, from about 1,800 to 4,000+, driven by the new product pages and technical fixes.How measured: AI-engine referrals in analytics, April vs. August.
Mention rate: AI naming the product, up from the April baseline.How measured: AirOps, MFT category prompts, eight AI engines.
MFT pages moved up from Google's second page toward the first.
Blog-to-contact conversion versus the site average on this product line.
What carries forward
Pick one product, separate "cited" from "known," fix access before content, and run it as a weekly loop with the product owner in the room. Once it works on one line, it's a template for the rest.