ThriveStack citedbyAI search attribution learning guideIntroductionLive demoBook a demoStart free
Lesson 5.4 · Benchmarks, ROI and business case

AI search ROI:
how to calculate it and make the business case

AI search ROI is the value of AI-sourced customers divided by what you spend. Estimate AI visits from your total visits and an AI share, multiply by your signup and close rates, then by deal value. Use your own AI share. Studies range from 0.32% to 1.08% of traffic.

12.1%

of Ahrefs signups came from AI search, from 0.5% of visitors

Ahrefs, single site, Jun 2025
83%

higher revenue per visit for AI traffic on U.S. retail sites

Adobe, Sep 2026, published Oct 29, 2026
4.4x

value of an AI search visitor vs an organic visitor

Semrush, Jul 2025
1.08%

of website sessions from AI referrals across 10 industries

Conductor, updated Jul 6, 2026
ThriveStack citedby

A worked example:
1,000 AI visits become 3 deals you can price

Sample funnel. The rates are assumptions, so replace them with yours.

AI visits1,000Signups at 3%30Opportunities at 30%9Deals at 33%3Revenue = deals x your average contract value. Use your own number.

Assumptions for illustration: 3% visit to signup, 30% to opportunity, 33% to deal.

How to read it. Each step uses a rate you can find in your CRM. Multiply the deals by your average contract value to get AI-sourced revenue.

In ThriveStack citedby The funnel from AI visit to revenue is built from your own CRM data, so the ROI uses your rates, not a benchmark. See it for your site →

ThriveStack citedby

AI-influenced revenue can be larger than AI-sourced revenue, so show both lines

Sample monthly revenue credited to AI, in index points.

014284256JunJulAugSepOctNovAI-influencedAI-sourced

Sample data for illustration. Your own numbers will differ.

How to read it. The orange line is what analytics proves. The dark line adds deals where the buyer named AI. Keep the estimate labelled.

In ThriveStack citedby AI-sourced and AI-influenced revenue are two lines in one report, so finance sees proof and estimate separately. See it for your site →

Start with a range for your AI share, because the studies disagree.

SE Ranking found AI platforms at 0.32% of traffic across 101,574 sites (Jun 18, 2026). Conductor found 1.08% across 1,215 enterprise domains (updated Jul 6, 2026). They sample different sites and periods. Measure your own share in GA4, then run the math at both ends.

0.32%

of traffic from ChatGPT, the low end of the published range

SE Ranking, 101,574 sites, Jun 18, 2026
1.08%

of sessions from all AI referrals, the high end

Conductor, updated Jul 6, 2026

Four steps turn AI visits into a value you can compare with cost.

Here is a worked example. Every number is an assumption, so replace each with your own.

Worked example: 100,000 visits at a 1.08% AI share is worth about USD 5,400 a month at these assumptions
StepAssumptionResult
AI visits100,000 visits x 1.08%1,080
Signups5% signup rate54
Customers10% close rate5.4
ValueUSD 1,000 per customerUSD 5,400

Count AI-influenced deals on their own line.

Keep the headline on AI-sourced results. Show AI-influenced deals as an estimate beside it. A smaller number you can defend beats a bigger one you cannot.

Headline
  • AI-sourced results
  • A count you can defend
Beside it
  • AI-influenced deals as an estimate
  • With a note on how you counted

Compare the value with the full cost of the program.

Include tool cost and staff time. Divide the cost by the value of one AI-sourced customer to see how many customers pay for it.

Does the program pay for itself?
Tool cost + staff time
/ value of one AI-sourced customer
= customers needed to break even

A business case needs a low, a middle and a high case, because the studies spread wide.

Take the volume from your own GA4 if you have it. If you do not, use the published range: 0.32% of traffic from ChatGPT (SE Ranking, May 2026) to 1.08% from all AI referrals (Conductor). Take the conversion lift from a cautious source, such as 4.4 times (Semrush), and compare it with your own conversion rate.

Adobe reports AI visits earning 83% more revenue per visit on retail sites, and Ahrefs reports a 23 times conversion rate on one site. Treat those as the high case.

ThriveStack citedbyResearch table

A low, middle and high case, each input labelled with its source and date

A low, middle and high case, each input labelled with its source and date
CaseAI share of trafficConversion lift
Low0.32% (SE Ranking, May 2026)Your own, or none
MiddleBetween the twoCautious source, 4.4x (Semrush, vendor study)
High1.08% (Conductor, Jul 2026)Your own CRM data
Source: see sources belowthrivestack.ai/citedby · Nov 2026

Finance trusts a range with sources more than a single big number.

Show the three cases in a small table: visits, conversion, value and cost. Label each input with its source and date. Add AI-influenced deals as a second line with a note on how you counted them.

When your own data is ready, replace the published inputs one at a time and keep the old column for comparison.

Finance trusts ranges

A range with sources is more credible than one big number. Replace each published input with your own data as it arrives.

“We're still processing the data for a larger study across many sites.”
Patrick StoxAhrefs, on why the 23x conversion result comes from one siteAhrefs blog, Jun 16, 2025
How ThriveStack citedby helps

ThriveStack citedby reports AI-sourced and AI-influenced pipeline by engine, so the ROI math uses your numbers.

Build my AI attribution report →

AI search ROI, answered.

How do I calculate AI search ROI?

Multiply AI visits by signup rate, close rate and deal value, then compare with your cost.

What share of traffic comes from AI?

Studies range from 0.32% to 1.08%. Measure your own site.

Should I include AI-influenced deals?

Show them on a separate line, labelled as an estimate.

Sources for this lesson

  1. Ahrefs: 0.5% of visitors drove 12.1% of signups (Patrick Stox, Jun 16, 2025) Single site
  2. Adobe: AI traffic drives 83% higher revenue per visit (Oct 29, 2026) 1 trillion visits
  3. Semrush: impact of AI search on SEO traffic (Rachel Handley, Jul 21, 2025)
  4. Conductor: 2026 AEO / GEO Benchmarks Report Updated Jul 6, 2026
  5. SE Ranking: ChatGPT referral traffic, May 2026 (Jul 9, 2026)

Figures were checked against each page in Nov 2026. Vendor studies are marked by their sample. Charts marked sample data are illustrations.

All 10 questions this lesson answers

AI search ROI calculator

Visits x AI share = AI visits. Multiply by signup rate, close rate and deal value. Example with assumed rates: 100,000 visits x 1.08% = 1,080 AI visits, 5% signup gives 54, 10% close gives 5.4 customers, and at USD 1,000 each that is USD 5,400 a month. Replace each rate with your own.

AI traffic conversion rate benchmark

There is no neutral benchmark. Ahrefs reported 0.5% of visitors drove 12.1% of signups on one site (Ahrefs, Jun 16, 2025, checked Nov 2026). Adobe reported 83% higher revenue per visit on U.S. retail sites (Adobe, Oct 29, 2026, checked Nov 2026). Semrush modeled 4.4x (Semrush, checked Nov 2026). Use your own CRM rates.

ChatGPT traffic conversion rate vs Google

Results differ by site. Ahrefs saw a much higher conversion rate from AI visitors on one site, and Ahrefs says a larger study is in progress (Ahrefs, checked Nov 2026). Compare ChatGPT and Google visits in your own GA4 before you quote a ratio.

AI search share of pipeline

We found no neutral figure for share of pipeline. Ahrefs' 12.1% is share of signups on one site, not pipeline (Ahrefs, checked Nov 2026). Measure your own with a signup survey and CRM source fields.

How much revenue comes from AI search

It depends on your traffic and rates. Published traffic shares run from 0.32% from ChatGPT (SE Ranking, 101,574 sites, checked Nov 2026) to 1.08% of sessions from AI referrals (Conductor, checked Nov 2026). Multiply your own visits by these to bracket revenue.

AI search attribution business case

Show a low, middle and high case. Label each input with source and date, and show AI-sourced results first with AI-influenced deals as an estimate beside it.

How to prove AI search ROI to the CFO

Give a range, not one number. Use two lines: AI-sourced, which analytics can prove, and AI-influenced, which is an estimate with a stated method. Replace published inputs with your own as data arrives.

AI search payback period

Add tool cost and staff time, then divide by the value of one AI-sourced customer. That gives customers needed to break even. Example with assumed numbers: USD 1,500 a month in cost and USD 1,000 per customer needs 1.5 customers a month.

Cost per AI visit

Monthly cost divided by AI visits. Example with assumed numbers: USD 500 for 1,080 AI visits is about USD 0.46 a visit. Divide by AI signups too.

AI search CAC vs paid search CAC

We found no neutral benchmark. Compute it yourself: AI program cost divided by AI-sourced customers, then compare with your paid search CAC from the same period. Keep AI-influenced deals out of this number.

Go deeper
MapHover to open