So based on any business I've ever been associated with, those would be seen as strong growth prospects.
So it would take heavier weight for most recent and most volume data. And then I apologize for beating a dead horse here with the slowdown of the Medicare-approved submissions this quarter. So going into Q4, we expect sequential LTVs to be up given that pattern.
We have broadened its mandate to focus on not only enrollment but extending it to proactive post-enrollment engagement as customers start utilizing their Medicare policies. The number of third-quarter Medicare-approved members grew 17% compared to a year ago. Just curious on your thoughts there, and I'm assuming you guys use Walmart as one of your channel partner, and if this has any implication on that relationship. So better recommendation algorithms, better cues for them on the call, things they could be looking out for. So in terms of kind of how that measures the guidance, I think you're right. This is Tim. As a result, the first quarter is the most meaningful period in terms of persistency data. As a reminder, given that we observed churn on a lag, our quarter-end membership figures measure the estimated number of effective policies paid by members based primarily on cash collections and historical retention data. Most notably, when consumers were calling to cancel, 90% of the time, we were able to keep them as eHealth customers and the overwhelming majority in their current plan.
Our fulfillment mix has significant implications for the average policy persistency given that online enrollments tend to retain the product longer, with first-year retention rates higher by approximately 25% to 30% than members who enrolled telephonically. Your line is now open. They don't have their drugs in front of them.
And we have been able to change the targeting within all of our channels. Yeah, Greg. Agent cost per approval were favorable compared to last year due to higher percentage of online enrollments. We have now extended our platform to serve thousands of independent and community pharmacies and providers nationwide. And so we promote directly to them. 36% of our applications for these products were submitted by our customers online. Our enrollment volumes through Q3 was in line with our expectations as was the overall top-line expectations. This has been a busy and successful quarter preparing for the most important Medicare-selling season of the year and deploying our retention enhancement initiatives across the entire organization. So if you go back to Q3 '18, the year-over-year growth compared to Q3 '17 for Medicare Advantage was 0%. In-house eHealth agents now represent roughly 45% of our total sales force compared to just 30% last AEP. I was wondering if you might be able to comment more broadly on your partnership strategy. And at the same time, we are, as you have heard here, making good progress in our final implementation of retention initiatives, which won't show up on our reported metrics because we do need observations on how persistency has improved in order for LTVs to increase again. So that is one of the main reasons why we expanded our retention team in the run-up to AEP to add in agents to do outbound calling to our high-risk customers and to be able to handle a larger inflow with licensed agents for our existing customers. In line 23, the word "importing" could most accurately be restated as. As always, our customers are at the center of everything we do at eHealth, and we are acutely focused on enhancing and simplifying their experience by meeting them where they want to be, whether online through interaction with our licensed agent or customer care specialists or through a hybrid agent-assisted online enrollment. The IFP business has performed well.
So implied in our guidance based on year-to-date results, our Medicare revenue in Q4 growth will be close to 50%. Good afternoon. Or is it hard to characterize it in one bucket?
So we do look at our retention across as many different dimensions as we possibly can, and what I would say is that the overwhelming drivers were factors within our control within our sales funnel and less to do with exogenous changes happening around us. I'm sorry. And then the last thing I'd say, which we'll reiterate, is really given the choice of investing for growth in Q3 versus Q4, we would push to Q4 given the much better economics of how we deploy the capital. We expect it to double this year in aggregate.
So if you enrolled in Q3, you have the opportunity to switch plans again in Q4.
And in the last year, we had the AEP and the OEP.
Most of my questions have been answered, but I did want to go back to your assumptions around the growth in the online channel. But certainly, this year, we've gone to market not just with the customer center but with an entire redesign of our e-commerce funnel, taking into account customer research that we've done, first-party research, as well as extensive A/B testing.
As you heard on the call, we feel very good about all those plans and everything that we've seen relative to how we have been executing.
Even though we try, we exert maximal effort to extract all of that information. Please note that we report churn as any planned switching by a paying member even if the change is made on our platform and eHealth remains the broker of record.
... (line 13) most likely represents the point of view of. As a result, we expect that this shift will have positive implications for our overall conversions and per member acquisition costs compared to a year ago, as well as for how well we retain members post-enrollment. Thank you. So we have been able to leverage our data and analytics to better understand what drives persistency.
We will be presenting certain financial measures on this call that are considered non-GAAP under SEC Regulation G. For reconciliation of each non-GAAP financial measure to the most directly comparable GAAP financial measure, please refer to the information included in our press release and in our SEC filings, which can be found in the About Us section of our corporate website under the heading Investor Relations. Good afternoon. Perhaps you could give us an update on how you see that looking forward because, as I said, the stock is down in the aftermarket.
Good afternoon, and thanks for taking my question. And from an investment perspective, our goal is to prepare for the annual enrollment period both for sales, marketing, and also for technology investments.
So there was a gap in how we were serving our customers that was very apparent very quickly. 4. We entered the annual enrollment period from a position of strength and are expecting to generate significant Medicare enrollment while reducing our per-member acquisition costs, a powerful combination. These are inbound calls, so customers will call with a concern of some kind and previously would be routed to a sales agent who was attempting to sell them something new when that probably wasn't appropriate. These results were in line with our expectations for the quarter and reflect an investment made in our Medicare telesales capacity with an emphasis on expanding our in-house agent force, as well as marketing investments ahead of the AEP. So we had a very, very light comp in Q3 2019.
So you're correct.
On the call this afternoon, we'll have Scott Flanders, eHealth's chief executive officer; and Derek Yung, chief financial officer. And given most of the -- you've hired a lot of internal folks this past quarter, and you generally don't get productivity until the second AEP period.
But didn't your second quarter include like some 20,000-member turnover catch-up from the first quarter, so on an apples-to-apples basis, not that big decline? So I think you're wise to see us within the revenue guidance range if that's what you're trying to interpret. Good afternoon, guys, and thanks for taking the question. Thank you. If we think of the multipronged approach you have to improving customer retention, does getting back to the prior LTV represents sort of the best outcome? And I do agree, Elizabeth, it is an underappreciated part of our story. I mean, we do watch the provider channel and follow the public systems, and clearly, they are struggling with the lack of elective surgeries.
Your line is now open. And so when somebody would call in and expressed that they wanted to cancel or make a change, 90% of those calls resulted in them remaining an eHealth customer. Within the first year, churn has been quite front-end-loaded. Please go ahead, sir. So maybe I could take one more stab at trying to understand that.
And it's certainly our hope and expectation that they will. Right. Learn vocabulary, terms, and more with flashcards, games, and other study tools.
Can you flesh out a little bit more, or did we understand that wrong?
Year to date, our Medicare segment profit was $19.4 million, compared to $5.9 million for the same period last year or an increase of over 200%.
I understand second quarter had some SEP benefit but still a sequential decline from around 60,000 approved members in the second quarter to 45,000 in third quarter. Q3 has always been a big investment quarter for Medicare as we expand our telesales capacity and deploy technology upgrades in preparation for the annual enrollment period. So just curious how you think about that from an execution standpoint and investing standpoint. So you mentioned that you're going to have less focus on "high conversion customer" volume and more focus on higher LTV opportunities, meaning it looks like folks that would churn at a slower pace. So you're right. Market data powered by FactSet and Web Financial Group. Please go ahead, ma'am. Calculate adjusted EBITDA by adding stock-based compensation expense, change in fair value of earnout liability, depreciation and amortization, amortization of intangible assets, other income, and benefit from income taxes to our GAAP net loss. This was primarily driven by a 37% increase in approved IFP members for major medical plan products, accompanied by a continued trend of longer duration for these products and an 18 point million in residual or tail revenue that we booked in this segment during the quarter. At the same time, our total agent number, including outsourced call center agents, increased approximately 40%, which is well below the expected growth in fourth-quarter Medicare enrollments. After management completes its remarks, we'll open the lines for questions. Thank you, everyone, for your informed questions. Frank Morgan -- RBC Capital Markets -- Analyst. Kate Sidorovich -- Vice President of Investor Relations.
This was primarily driven by seasonal factors and was within our expected range. I'd like for Tim to answer that because he is singularly accountable to us for those results.
Yeah, I can take that one. Your next question comes from the line of Dave Styblo from Jefferies. In general, our commission cash collections in Medicare continue to be favorable when compared to initial estimates used for LTV at the time we recognize revenue.