Full Flow

Insights / August 26, 2026 / content / pipeline / search & AI visibility

Your website is a brochure and your buyer is running an evaluation

Utility engineers are browsing vendor sites with a short list of technical questions they need answered before they can advance you through the procurement process. Most water company websites answer none of them.

A process engineer at a mid-size utility opens your website on a Tuesday afternoon. They have a manganese problem, a capital request due in the Fall, and four vendors in a browser tab each. They spend about four minutes on your site, find a hero video, a page called Solutions, and a form. They close the tab and go back to the vendor whose installation list included a plant their size.

You will never see that visit in any meaningful way. It shows up as one session in analytics with no conversion, but it happened during the phase of the purchase where the shortlist actually gets built. By the time an RFP hits the street, the evaluation that mattered was finished months earlier, sometimes by an AI assistant summarizing the same pages the process engineer was reading.

The visit is a work session

The reason water company websites underperform is a mismatch of purpose. The site is built to introduce the company; the visitor is trying to finish a task.

That task is usually one of three things: filling in an alternatives table for a preliminary design report, assembling enough evidence to justify a vendor internally, or checking whether a vendor name someone mentioned is credible enough to keep in the consideration process. All three are technical work, and all three have a specific list of questions that must be answered. In water, the buying committee runs anywhere from one, to more often, upwards of nine people and the engineer visiting your site is the one who has to defend the recommendation to the other eight.

The questions are consistent across the industry, and I've watched them come up on nearly every deal I worked at my own ventures:

  • Have you done this at my scale, with water or assets that look like mine?
  • Who has it running today, and can I call them?
  • What does it require from me once it's installed, in power, chemicals, footprint, and operator hours?
  • Roughly what does it cost, in a range I can put in a budget request?
  • Will it clear procurement, meaning certifications, standards compliance, financial stability, and bonding?

A vendor site that answers those five plainly will beat a better-looking site that answers none of them. That is most of the strategy.

What sites do instead

The typical water technology website opens with a claim nobody can evaluate. "Innovative water solutions for a sustainable future" tells an engineer nothing, and they've read it on nine other sites this month. The Solutions page describes a category instead of a product. The technology page explains the science and stops before the operating envelope. Nowhere on the site is a number an engineer can put in a table.

Then there is the gate. The spec sheet, the case study, and the design guide sit behind a form that asks for company, phone, and job title. That gate exists because someone needs to report MQLs. Its actual effect is to remove your document from the evaluation. A consulting engineer building an alternatives analysis on a Thursday deadline will not fill out five forms; they will use the competitor whose PDF opened in one click, or ask an AI assistant, which cannot fill out your form either. Gating technical documentation in this industry trades the deal for a lead record.

The other pattern is the case study written by marketing without a utility's permission. 400 words, anonymized plant, gives no flow rate, and concludes that the customer was pleased. It reads as fiction to the person it was written for.

Getting the evidence you are allowed to publish

The reference problem is real and it's the objection that comes up every time. Public utilities generally cannot endorse a vendor, procurement rules often forbid it, and a plant superintendent who agrees to a glowing quote can create a problem for themselves.

The way around it is to stop asking for an endorsement and start asking for a technical record. Utilities that will never approve a testimonial will frequently co-present a paper at a state section meeting, sit on a conference panel about the project, or let their engineer describe the results factually in a write-up their public information officer reviews. A WEFTEC or state AWWA paper with a utility co-author is as strong (or stronger, in many cases) than any testimonial, and it's the work you were already doing for the trade show anyway. Publish it on your site, with the flow rates, the influent chemistry, the commissioning timeline, and the problems you hit during startup.

Include the problems. An engineer who reads that your first install had a fouling issue in month three and how you solved it trusts the rest of the document. Case studies where nothing went wrong are the ones that get discounted.

The pages that do the work

Most water companies do not need a redesign. They need four or five pages that currently do not exist.

An installation list, with plant size, location by state, year, and application, is the single highest-value page you can publish. It answers "have you done this at my scale" in ten seconds and is defensible.

Application pages organized around the buyer's problem instead of your product line catch the search and the AI query both. "Manganese removal for groundwater systems under 5 MGD" is how the problem is described by the person who has it. Your product name is not.

Ungated spec sheets, with operating ranges, utility requirements, and dimensions. An O&M page describing what the operator does daily, weekly, and annually, because operations staff kill more purchases than they approve and nobody markets to them. And a pricing signal, even a wide one. "Systems in this range typically run $400k to $1.2M installed depending on pretreatment" is enough to survive a budget conversation, and vendors who publish nothing get eliminated by engineers who can't size the capital request.

Judging whether it worked

Bounce rate will not tell you anything here, and neither will time on page. The measures that matter attach to deals.

Watch which pages get visited by accounts that later enter real opportunities, ask the self-reported attribution question the way the long sales cycle demands, and listen for a specific answer: "your installation list had two plants in our state." Track repeat visits from the same organization, which in water is usually a committee circulating a link. And ask your sales team the direct version of the question, which is whether they have ever sent a prospect a page from the website instead of a PDF attachment. When the answer turns to yes, the site has started doing its job.

Fixing this is also what makes the rest of the budget work. Paid search sends the click, the conference sends the name, the executive post sends the curious reader, and every one of them lands on the same set of pages. A site that cannot answer five questions turns all of that spend into untrackable, unremarkable, and un-saleable page views.


Adam Tank is the founder of Full Flow Marketing, a marketing agency built for the water industry. If your traffic looks fine and the inquiries never come, ask for a free pipeline diagnostic and get a read on what your site needs to change to start converting.