Does calling a lead within minutes raise conversion?

What the research says

The best-known evidence on speed to lead is the Lead Response Management Study, presented in October 2007 by Dave Elkington, then chief executive of InsideSales.com, and James Oldroyd, then a faculty fellow at MIT. According to the study summary, it used data from InsideSales.com’s own system: three years of records from six companies that generate web leads, more than fifteen thousand leads and more than a hundred thousand call attempts.

Its headline, repeated in InsideSales.com’s press release of the time, was that the odds of qualifying a web lead called within five minutes were 21 times those of a lead first called at 30 minutes. The summary also reports that the odds of reaching and of qualifying a lead fell steeply over the first hour.

Two definitions matter. A contact was a call that reached a live person and lasted a set length of time. A lead was qualified when it agreed to enter the sales process, for example by booking an appointment, with each company using its own definition. The summary says plainly that the study did not address close rates.

Why speed might matter

• The lead asked for the call. A person who has just filled in a form is near a phone and thinking about the purchase. In our view that moment fades quickly.
• Other sellers may call too. The study’s authors note that lead providers often sell the same lead to several buyers; their own test enquiries brought calls from several companies. The first caller gets the first conversation.
• Contact is the gate. A lead nobody reaches cannot buy, so anything that lifts contact lifts the ceiling on sales.

The limits

• One vendor’s data. The study came from a company that sells lead response software, and we found it as a summary and a press release, not as a journal article.
• Correlation, not a test. The summary describes an analysis of existing records, not leads assigned to fast or slow calls at random. Companies that call fast may differ in other ways, and leads that arrive during working hours are easier to call fast and may also be easier to reach.
• Qualification is not a sale. Speed may raise contacts and appointments more than it raises sales per contact. More contacts or better conversion explains why the two need separate numbers.
• Web leads only, and in 2007. It says nothing about cold lists or aged leads, and in our view phone habits have changed since.
• A fast call is still a call. Consent, do-not-call lists and calling hours stay with your compliance team, however quickly the lead came in.

How to test it on your own floor

1. Measure your response time today: lead created to first dial, and to first contact, for every lead, by source and by hour of arrival. Look at the spread as well as the average.
2. Split incoming leads at random. One group goes to a call-now queue; the other follows your current process. Same weeks, same agents, same scripts.
3. Score on sales per lead. Speed changes who you reach, so sales per contact alone can hide the effect. Report sales per contact next to it to see whether the extra contacts close at the same rate. How to calculate close rates and pick the denominator.
4. Wait for sales to settle, including cancellations, before reading the result.
5. Count the cost. Calling within minutes needs agents free when leads arrive. Set the extra sales against the staffing it takes.

In our view speed decides how many leads you reach, and who is on the call decides much of what happens next. Measure both, and change one at a time. How to prove a lead routing change worked.

Sources

• MIT Lead Response Management Study summary (InsideSales.com and James Oldroyd, 2007): a copy of the study summary, hosted by Mortech, with the data, definitions and findings.
• InsideSales.com press release on the study (October 2007): the vendor’s announcement of the five-minute and 30-minute finding.
• The Short Life of Online Sales Leads (Harvard Business Review, March 2011): the later article by Oldroyd, McElheran and Elkington; we could not open the full text and quote nothing from it.