At myYearbook, Vadim Telyatnikov encountered an awkward fact about online advertising. A campaign sold by a salesperson could earn less than an ad sold through the supposedly less desirable remnant market. The labels sounded reassuring. The money told a different story. For someone responsible for a publisher’s revenue, that was a useful irritation.
- LiftDNA helped publishers earn more from the advertising space they already had.
- Its software worked inside existing ad servers, backed by a hands-on operations team.
- OpenX acquired it in February 2012; the work fed into later OpenX products.
A page view creates a fleeting opportunity: somebody can show an advertisement to that visitor. Once the moment passes, the opportunity expires. Managing that business means deciding which buyer should get which impression, while honoring commitments already made. LiftDNA’s proposition was to improve that decision where the publisher was already making it.
01 / The expensive word was “remnant”
Telyatnikov had built yield-management tools inside myYearbook. He expected outside supply-side platforms to improve the results. In his account, they failed to outperform the internal system. That disappointment changed his question: perhaps other publishers needed the advantage of working inside the ad server, too. LiftDNA launched in September 2009.
The important distinction was architectural. An ad server decides what advertisement appears. An exchange connects buyers and sellers. A yield optimizer helps a publisher choose among earning opportunities. LiftDNA occupied that last role while integrating with the first. It could work with systems such as Google’s DFP and OpenX, preserving the publisher’s existing relationships.
Consider what that means for adoption. A publisher need not begin by tearing out its infrastructure or dismissing every existing partner. The proposed improvement sits within a familiar workflow. That is a lesson a software company in almost any industry can borrow: make the next decision better before asking the customer to rebuild the whole department.
02 / Sell it at the mine
In February 2012, Telyatnikov explained the approach with a mining analogy. Shipping raw materials to different markets meant some could fall off the train along the way. His alternative placed the sale at the point of production.
“We sell all the raw material at the mine.”
Vadim Telyatnikov · February 2012
OpenX executive Jason Fairchild told The Philadelphia Inquirer that publishers could lose up to 10 percent of inventory moving it from internal servers to outside exchanges and networks. That was an executive’s estimate, not a universal measurement. Still, it helps explain why the location of a decision mattered as much as the number of potential buyers.
Publishers were LiftDNA’s customers. Advertisers supplied the demand, but the company’s job was to help the seller make better use of it. Its initial customer base leaned toward growing mid-tier publishers handling more than 100 million impressions monthly. At that scale, a small improvement per impression can matter; so can the time spent managing the machinery.
03 / The software came with people
LiftDNA also handled the work that made monetization function: ad trafficking, billing, reconciliation, collections, inventory segmentation and data analysis. These are unglamorous jobs with real consequences. A clever allocation decision is less useful if a campaign is configured incorrectly or nobody reconciles what the buyer actually owes.
That made the business a combination of publisher software and managed operations. The service component belongs in any assessment of its economics. A revenue improvement is useful only after the cost of delivering it. For a small publisher, intensive outside help might be hard to justify. A larger publisher with a capable internal team would ask a different question: what additional value does this arrangement produce?
Additional revenue
− fees − added operating cost
= actual benefit
There is a practical discipline here: compare net results against a credible baseline. More buyers do not automatically mean more money. Neither does a more complicated stack. The relevant outcome is what the publisher keeps, with acceptable delivery and advertising quality. A business without competing demand or meaningful monetizable inventory has less for an optimizer to improve.
Later founder portraits, published by Global Court Collective.
04 / A partner becomes the buyer
OpenX and LiftDNA had worked together at a product level since 2010. By the acquisition announcement on February 27, 2012, their technologies were already integrated. OpenX offered LiftDNA by OpenX both within its Enterprise platform and as a standalone service. The deal brought optimization together with ad serving and an exchange.
The financial scale deserves care. Contemporary reporting put LiftDNA’s outside capital at about $1 million, including $250,000 from Ben Franklin Technology Partners. The company had about 30 employees at acquisition, 19 in King of Prussia. The purchase price was undisclosed. Funding, headcount and acquisition proceeds are different numbers; confusing them produces a much louder, much less useful story.
By April 2013, Telyatnikov was leaving, with integration nearing completion under OpenX’s Qasim Saifee. In June 2014, OpenX introduced Demand Fusion, built on the acquisition and subsequent development. It brought real-time bidding and ad-network demand together; launch partners included YP, TVGuide.com and The New York Observer. Those were partners for the later OpenX product.
Competition continued. PubMatic had introduced a unified auction in 2010, and Google offered dynamic allocation within its ecosystem. LiftDNA’s story therefore rewards precision: an early, particular approach to improving publisher decisions, followed by a larger platform’s investment. Its most portable idea is wonderfully practical. Before treating inventory as leftovers, find out whether the selling system is giving it a fair chance.