Jev for SEO and marketing teams: what it can do and what it costs

Jev, TypeSafe AI's decision model, answers a question about text with a value your software can use: one option from a list, a score, or a yes/no probability. That suits the small, repeated decisions in the middle of this work.

These are the jobs on this site that fit SEO and marketing teams best. Each has its own page with the questions to ask, the pitfalls and a worked example.

• Label keyword lists by intent and relevance, at any volume. Search intent is a textbook Choice question with four well-known classes: informational, commercial, transactional and navigational.
• Tag ads by angle and hook strength, and pre-check them against policy. Jev can tag each ad by angle, rate hook strength and check it against your policy list in a single request, which makes it practical to label thousands of competitor or historical ads.
• Surface only the papers and headlines that match what you care about. Relevance filtering is a natural Score question: describe what you care about, and Jev rates each abstract or headline against it.

• Jev for SEO search-intent classification
• Jev for ad and creative classification
• Jev for research and news filtering

Jev lists $0.042 per million input tokens with output free, so the cost depends on how much text you send and how many questions you ask about it. At the typical volumes the use-case pages assume, the calculator estimates:

• Keywords (SEO search-intent classification): about under 1¢ per 1,000, at 2 questions and about 40 tokens each.
• Ads (ad and creative classification): about 1.3¢ per 1,000, at 3 questions and about 120 tokens each.
• Papers (research and news filtering): about 2.1¢ per 1,000, at 3 questions and about 350 tokens each.

• Cost calculator
• Jev pricing and real runs

A plain way to begin, in order:

• Choose the label set you already use, such as the four search intents, and write it as a Choice question.
• Run an export of a few thousand rows and spot-check a sample against your own judgement.
• Add a Score question for how relevant each row is to your product.
• Save the labels next to the data, so the next export only needs the new rows labelled.

• How to use Jev

Jev labels text. It cannot see the page behind a keyword or the picture in an ad, so judge intent and ad angle from the words you give it.

From the use-case pages, Jev is not the right choice when:

• You need search volume or difficulty. That data comes from SEO tools, not a model.
• The creative is image- or video-only with little text.
• You need summaries of what you keep. That's a language model's job.

• What Jev can't do

Labelling at scale is the unglamorous job these models are cheapest at. The money is in the labels you can now afford to make, such as intent on every keyword, which a team would never have paid a person to do.

The figures above are list-price estimates, not measured bills, and the accuracy claims on the linked pages are the published runs' own. Test Jev on a labelled sample of your own data before you rely on it.

• What is a decision model?