New format
Yandex Direct strategies: start from economics, not the button label
In July 2017 Direct retired “Show in the block at minimum price” — the mode many used to “just try ads” without a CPA target. Strategy names and screens have changed again since then.
The durable lesson isn’t about one 2017 button: without a clear cost per lead or order and feedback from sales, both automation and manual bids fly blind. Here’s how to think about strategies now.
What changed back then
“Show in the block at minimum price” felt like control for beginners: “we’re in the block, the bid isn’t crazy.” For agencies and experienced advertisers it was already too blunt — no CPA, ROI, or competitive footing.
Yandex moved toward richer strategies and metrics, closer to other large ad platforms. Fewer buttons for people who don’t count; more pressure to set goals and feed clean data.
Why “just try it” got harder
Without an average cost per call, lead, or deal, a bid is a lottery. “There were calls — good” never answers what a lead costs or whether ads pay off.
Some teams skip analytics on purpose (“no bandwidth”). Others tried “standard” strategies, saw nothing, and blamed paid search — while goals, keywords, or the landing were usually broken.
The usual gap:
- Direct spends on impressions and clicks
- the business only asks “were there calls?”
- nobody ties lead cost to lead quality
- the strategy gets swapped at random
What to decide before you pick a strategy
Economics and the goal first; strategy type second. You need a target action, enough conversions for learning, an acceptable CPA or ad-spend share, and budget limits.
Manual bidding fits launches, tests, and thin data. Auto-strategies fit when goals are clean and volume is high enough that the algorithm isn’t guessing.
Minimum before you go live:
- an analytics goal that actually fires
- a ceiling for cost per lead
- a landing that matches the query cluster
- negatives and a check on junk traffic
- a budget cap and a test window
Feedback beats the strategy button
Even a strong media buyer is stuck if CRM only says “someone called” with no status or deal value. Cleaner lead and sales data means sharper bids and better auto-strategies.
A strategy allocates budget. Campaign result = demand × offer × landing × measurement. No UI button replaces that chain.
FAQ
What disappeared in 2017?
The “show in the block at minimum price” strategy — a beginner-friendly “stay in the block, keep the bid down” mode. That magic button is gone; auctions and auto-bidding are still here.
Can I run Direct without numbers?
You can burn budget. You can’t steer results. Floor: a goal action, cost per lead or order, and margin.
Are auto-strategies better than manual bids?
Depends on volume and goal quality. Sparse conversions and dirty goals — the algorithm learns noise. Clean conversions at scale — auto-strategies earn their keep.
Where should a beginner start today?
A tight keyword set, a matching landing, analytics goals, and a hard budget cap. Not “pick the simplest strategy and hope.”
Is “we got calls / we didn’t” enough?
As a rough signal — yes. As the only KPI — no. You need cost per inquiry, lead quality, and a line to sales.
Should I chase premium placement?
Block height isn’t the goal. Economics is. See the premium placement article.
When do I need a specialist?
When spend is material and nobody inside joins bids, analytics, and sales. Otherwise you stall at “we’re fine” or “paid search doesn’t work.”
Does the right strategy guarantee leads?
No. A strategy only allocates budget. Without demand, offer, and landing it won’t carry the campaign.
Direct strategies without a clear CPA — and the cabinet button keeps changing?
We’ll lock goals, lead cost, and feedback first — then pick manual or auto mode from data, not auction fear.
Discuss the task
